updates to run under v0.7, 1.0

.travis.yml

@@ -1,15 +1,21 @@
 language: julia
+
 os:
   - linux
   - osx
+
 notifications:
   email: false
+
 julia:
-  - 0.6
+  - 0.7
+  - 1.0
   - nightly
+
 matrix:
   allow_failures:
     - julia: nightly
-#script:
-#  - if [[ -a .git/shallow ]]; then git fetch --unshallow; fi
-#  - julia -e 'Pkg.init(); Pkg.clone(pwd()); Pkg.test("ImplicitEquations")'
+
+script:
+  - if [[ -a .git/shallow ]]; then git fetch --unshallow; fi
+  - julia -e 'using Pkg; Pkg.clone(pwd()); Pkg.test("ImplicitEquations")'

LICENSE.md

@@ -1,6 +1,6 @@
 The ImplicitEquations.jl package is licensed under the MIT "Expat" License:
 
-> Copyright (c) 2014-16: jverzani.
+> Copyright (c) 2014-18: jverzani.
 >
 > Permission is hereby granted, free of charge, to any person obtaining
 > a copy of this software and associated documentation files (the

README.md

@@ -1,5 +1,3 @@
-[![ImplicitEquations](http://pkg.julialang.org/badges/ImplicitEquations_0.6.svg)](http://pkg.julialang.org/?pkg=ImplicitEquations&ver=0.6)
- 
 Linux: [![Build Status](https://travis-ci.org/jverzani/ImplicitEquations.jl.svg?branch=master)](https://travis-ci.org/jverzani/ImplicitEquations.jl)
  
 Windows: [![Build Status](https://ci.appveyor.com/api/projects/status/github/jverzani/ImplicitEquations.jl?branch=master&svg=true)](https://ci.appveyor.com/project/jverzani/implicitequations-jl)
@@ -32,5 +30,3 @@ plot(f ⩵ 0)  # \Equal[tab]
 ```
 
 ![DevilsCurve](http://i.imgur.com/LChTzC1.png)
-
-

REQUIRE

@@ -1,5 +1,3 @@
-julia 0.6
-ForwardDiff 0.3.0
-ValidatedNumerics 0.5.0
-RecipesBase 0.0.6
-Compat 0.8.6
+julia 0.7
+IntervalArithmetic 0.15.0
+RecipesBase 0.5.0

appveyor.yml

@@ -1,13 +1,19 @@
 environment:
   matrix:
-  - JULIA_URL: "https://julialang-s3.julialang.org/bin/winnt/x86/0.6/julia-0.6-latest-win32.exe"
-  - JULIA_URL: "https://julialang-s3.julialang.org/bin/winnt/x64/0.6/julia-0.6-latest-win64.exe"
-  - JULIA_URL: "https://julialangnightlies-s3.julialang.org/bin/winnt/x86/julia-latest-win32.exe"
-  - JULIA_URL: "https://julialangnightlies-s3.julialang.org/bin/winnt/x64/julia-latest-win64.exe"
+  - julia_version: 0.7
+  - julia_version: 1.0
+  - julia_version: latest
+
+platform:
+  - x86 # 32-bit
+  - x64 # 64-bit
+
+## uncomment the following lines to allow failures on nightly julia
+## (tests will run but not make your overall status red)
 matrix:
   allow_failures:
-  - JULIA_URL: "https://julialangnightlies-s3.julialang.org/bin/winnt/x86/julia-latest-win32.exe"
-  - JULIA_URL: "https://julialangnightlies-s3.julialang.org/bin/winnt/x64/julia-latest-win64.exe"
+  - julia_version: latest
+
 
 branches:
   only:
@@ -21,19 +27,12 @@ notifications:
     on_build_status_changed: false
 
 install:
-  - ps: "[System.Net.ServicePointManager]::SecurityProtocol = [System.Net.SecurityProtocolType]::Tls12"
-# Download most recent Julia Windows binary
-  - ps: (new-object net.webclient).DownloadFile(
-        $env:JULIA_URL,
-        "C:\projects\julia-binary.exe")
-# Run installer silently, output to C:\projects\julia
-  - C:\projects\julia-binary.exe /S /D=C:\projects\julia
+  - ps: iex ((new-object net.webclient).DownloadString("https://raw.githubusercontent.com/JuliaCI/Appveyor.jl/version-1/bin/install.ps1"))
 
 build_script:
-# Need to convert from shallow to complete for Pkg.clone to work
-  - IF EXIST .git\shallow (git fetch --unshallow)
-  - C:\projects\julia\bin\julia -e "versioninfo();
-      Pkg.clone(pwd(), \"ImplicitEquations\"); Pkg.build(\"ImplicitEquations\")"
+  - echo "%JL_BUILD_SCRIPT%"
+  - C:\julia\bin\julia -e "%JL_BUILD_SCRIPT%"
 
 test_script:
-  - C:\projects\julia\bin\julia --check-bounds=yes -e "Pkg.test(\"ImplicitEquations\")"
+  - echo "%JL_TEST_SCRIPT%"
+  - C:\julia\bin\julia -e "%JL_TEST_SCRIPT%"

docs/examples.md

@@ -22,7 +22,6 @@ is graphed over the default region as follows:
 
 ```
 using Plots, ImplicitEquations
-pyplot()
 
 a,b = -1,2
 f(x,y) = y^4 - x^4 + a*y^2 + b*x^2
@@ -121,11 +120,11 @@ Uses a few new things: the `screen` function is used to restrict
 ranges and logical operators to combine predicates.
 
 ```
-f0(x,y) = ((x/7)^2 + (y/3)^2 - 1)  *   screen(abs(x)>3) * screen(y > -3*sqrt(33)/7) 
+f0(x,y) = ((x/7)^2 + (y/3)^2 - 1)  *   screen(abs(x)>3) * screen(y > -3*sqrt(33)/7)
 f1(x,y) = ( abs(x/2)-(3 * sqrt(33)-7) * x^2/112 -3 +sqrt(1-(abs((abs(x)-2))-1)^2)-y)
 f2(x,y) = y - (9 - 8*abs(x))       *   screen((abs(x)>= 3/4) &  (abs(x) <= 1) )
 f3(x,y) = y - (3*abs(x) + 3/4)     *   I_((1/2 < abs(x)) & (abs(x) < 3/4))    # alternate name for screen
-f4(x,y) = y - 2.25                 *   I_(abs(x) <= 1/2) 
+f4(x,y) = y - 2.25                 *   I_(abs(x) <= 1/2)
 f5(x,y) = (6 * sqrt(10)/7 + (1.5-.5 * abs(x)) - 6 * sqrt(10)/14 * sqrt(4-(abs(x)-1)^2) -y) * screen(abs(x) >= 1)
 
 r = (f0 ⩵ 0) | (f1 ⩵ 0) | (f2 ⩵ 0) | (f3 ⩵ 0) | (f4 ⩵ 0) | (f5 ⩵ 0)
@@ -186,7 +185,7 @@ plot(f ⩵ 2*3^2)
 ## now add tangent at (3,3)
 a,b = 3,3
 dydx(a,b) = -b/a             # implicit differentiate to get dy/dx =-y/x
-tl(x) = b + dydx(a,b)*(x-a)  
+tl(x) = b + dydx(a,b)*(x-a)
 plot!(tl, linewidth=3, -5, 5)
 ```
 
@@ -204,7 +203,7 @@ interval arithmetic, as possible.
 
 The package
 [IntervalConstraintProgramming ](https://github.com/dpsanders/IntervalConstraintProgramming.jl)
-also allows for this type of graphing, and momre.
+also allows for this type of graphing, and more.
 
 ## TODO
 
@@ -221,4 +220,3 @@ also allows for this type of graphing, and momre.
 * Tupper sketches out how to be more rigorous with computing whether a region is black or white.
 
 * increase speed (could color 1-pixel regions better if so, perhaps; division checks; type stability).
-

src/ImplicitEquations.jl

@@ -1,12 +1,8 @@
-__precompile__(true)
-
 module ImplicitEquations
 
 
-using ForwardDiff
-import ValidatedNumerics: Interval, diam, isempty
+import IntervalArithmetic: Interval, diam, isempty
 using RecipesBase
-using Compat
 
 include("predicates.jl")
 include("intervals.jl")

src/intervals.jl

@@ -4,7 +4,7 @@ import Base: <, <=, ==, !==, >=, >,
              +, -, *, /, ^
 
 ## a few definitionsn for ValidatedNumerics that don't fit in there:
-## Validated numerics doesn't define these, as the order ins't a total order 
+## Validated numerics doesn't define these, as the order ins't a total order
 Base.isless(i::Interval{T}, j::Interval{S}) where {T<:Real, S<:Real} = isless(i.hi, j.lo)
 #<=(i::Interval{T}, j::Interval{S}) where {T<:Real, S<:Real} = <=(i.hi, j.lo)
 
@@ -14,7 +14,7 @@ Base.isless(i::Interval{T}, j::Interval{S}) where {T<:Real, S<:Real} = isless(i.
 
 
 ## BInterval represents TRUE (true, true), FALSE (false, false) and MAYBE (false, true)
-immutable BInterval <: Integer
+struct BInterval <: Integer
     lo :: Bool
     hi :: Bool
 
@@ -50,13 +50,13 @@ function negate_op(op)
 end
 
 ## OIinterval includes interval, if defined on interval and if continuous on interval
-immutable OInterval <: Real
+struct OInterval <: Real
     val::Interval
     def::BInterval
     cont::BInterval
     OInterval(val, def, cont) = new(val, def, cont)
 end
-
+(O::OInterval)(o::OInterval) = o
 Base.show(io::IO,  o::OInterval)  = print(io, "OInterval: ", o.val, " def=", o.def, " cont=",o.cont)
 
 ## some outer constructors...
@@ -66,22 +66,16 @@ OInterval(a) = OInterval(a,a)   # thin one...
 OInterval(i::Interval) = OInterval(i.lo, i.hi)
 
 Base.convert(::Type{OInterval}, i::Interval) = OInterval(i.lo, i.hi)
-Base.convert(::Type{OInterval}, x::S) where {S<:Real}= OInterval(x)
-Base.promote_rule(::Type{OInterval}, ::Type{ForwardDiff.Dual{N,B}}) where {N,B<:Real} = warn("defined to remove ambiguity")
+
 Base.promote_rule(::Type{OInterval}, ::Type{A}) where {A<:Real} = OInterval
 
 ## A region is two OIntervals.
-immutable Region
+struct Region
     x::OInterval
     y::OInterval
 end
 
-## not good for v0.5+
-#call(f::Function, u::Region) = f(u.x, u.y)
-
-
-#ValidatedNumerics.diam(x::OInterval) = diam(x.val)
-diam(x::OInterval) = diam(x.val)
+ImplicitEquations.diam(x::OInterval) = diam(x.val)
 
 ## extend functions for OInterval
 ## Notice these return BIntervals -- not Bools
@@ -116,7 +110,7 @@ rather a `BInterval` which allows for a "MAYBE" state. As such, a
 simple ternary operator, like `x > 0 ? 1 : NaN` won't work, to screen values.
 
 """
-screen(ex) = (ex == FALSE) ? NaN : 1 
+screen(ex) = (ex == FALSE) ? NaN : 1
 const I_ = screen               # indicator function like!
 
 ## Functions which are continuous everywhere
@@ -159,7 +153,7 @@ Base.tanh(x::OInterval) = sinh(x)/cosh(x)
 Base.coth(x::OInterval) = 1/tanh(x)
 
 function Base.asin(x::OInterval)
-    if x.val.hi < -1.0 ||  x.val.lo > 1.0  
+    if x.val.hi < -1.0 ||  x.val.lo > 1.0
         OInterval(x.val, FALSE, FALSE)
     elseif (x.val.lo < -1.0) & (x.val.hi > 1.0)
         OInterval(Interval(-pi/2, pi/2), x.def & MAYBE, x.cont)
@@ -208,7 +202,7 @@ Base.exp(x::OInterval) = OInterval(exp(x.val), x.def, x.cont)
 ## /
 ## division is slow
 function /(x::OInterval, y::OInterval)
-    ## 0 is the issue. 
+    ## 0 is the issue.
     if 0.0 ∈ y.val
         ## maybe defined, maybe continuous
         OInterval(x.val/y.val, x.def & MAYBE, x.cont & MAYBE)
@@ -218,7 +212,7 @@ function /(x::OInterval, y::OInterval)
 end
 
 ## log
-function Base.log(x::OInterval) 
+function Base.log(x::OInterval)
     if x.val.hi <= 0
         OInterval(x.val, FALSE, FALSE)
     elseif x.val.lo <= 0
@@ -254,7 +248,6 @@ function ^(x::OInterval, q::Rational)
     OInterval(val, x.def, x.cont)
 end
 
-^(x::OInterval, r::ForwardDiff.Dual) = warn("defined to resolve ambiguity")
 function ^(x::OInterval, r::Real)
     r < 0 && return(1/x^(-r))
     if x.val.hi < 0
@@ -287,7 +280,7 @@ end
 
 Base.max(x::OInterval, y::OInterval) = OInterval(max(x.val, y.val), x.def & y.def, x.cont & y.cont)
 Base.min(x::OInterval, y::OInterval) = OInterval(min(x.val, y.val), x.def & y.def, x.cont & y.cont)
-    
+
 
 ## others sign, mod, ...
 function Base.sign(x::OInterval)
@@ -304,7 +297,7 @@ function xy_region(u, L, R, B, T, W, H)
     c = B + py.lo * (T - B) / H
     d = B + (py.hi) * (T - B) / H
     delta = sqrt(eps())
-    
+
     x, y  = OInterval(a+(u.x.cont==TRUE)*delta,b-delta), OInterval(c+0*delta,d-delta)
     x, y
 end
@@ -324,7 +317,7 @@ function compute(p::Pred, u::Region, L, R, B, T, W, H)
 
     (fxy.def == FALSE) && return (FALSE)
     isempty(fxy.val) && return (FALSE & fxy.def)
-    
+
     if p.op === ==
         return((p.val ∈ fxy.val) ? MAYBE : FALSE)
     elseif negate_op(p.op) === ==
@@ -340,7 +333,7 @@ end
 ## build up answer
 function compute(ps::Preds, u::Region, L, R, B, T, W, H)
     vals = [compute(p, u, L, R, B, T, W, H) for p in ps.ps]
-    val = shift!(vals)
+    val = popfirst!(vals)
     for i in 1:length(ps.ops)
         val = ps.ops[i](val, vals[i])
     end
@@ -351,14 +344,14 @@ end
 
 Does this function have a zero crossing? Heuristic check.
 
-We return `TRUE` or `MAYBE`. However, that 
+We return `TRUE` or `MAYBE`. However, that
 leaves some functions showing too much red in the case where there is no zero.
 
 """
 function cross_zero(r::Pred, u::Region, L, R, B, T, W, H)
     x, y = xy_region(u, L, R, B, T, W, H)
     dx, dy = diam(x), diam(y)
-    
+
     n = 20                      # number of random points chosen
     λ1s, λ2s = [0.0; 1.0;rand(n)], [0.0; 1.0; rand(n)]
     β1s, β2s = [1.0; 0.0; rand(n)], [1.0; 0.0; rand(n)]
@@ -370,7 +363,7 @@ function cross_zero(r::Pred, u::Region, L, R, B, T, W, H)
         val = (r.f(ll...) - r.val) * (r.f(ur...) - r.val)
         ((val <= 0)==TRUE) && return(TRUE)
     end
-    return(MAYBE)              
+    return(MAYBE)
 end