metab() can't be called by ddply()? confused

[rBatt]

I can run the following code fine when I pass each year-lake combo to metab() individually, but not when I try to do it all through ddply(). I think the error might be in conquerList:

metabBK <- ddply(hfData, c("lake"), metab, method="bookkeep", irr.name="irr2")

Output:

[1] "Points removed due to incomplete day or duplicated time step: 7933"
[1] "NA's added to fill in time series: 228960"
Error: length(u1) == s1 * s2 is not TRUE

traceback()

6: stop(list(message = "length(u1) == s1 * s2 is not TRUE", call = NULL, 
       cppstack = NULL))
5: .Call("plyr_loop_apply", PACKAGE = "plyr", n, f)
4: loop_apply(n, do.ply)
3: llply(.data = .data, .fun = .fun, ..., .progress = .progress, 
       .inform = .inform, .parallel = .parallel, .paropts = .paropts)
2: ldply(.data = pieces, .fun = .fun, ..., .progress = .progress, 
       .inform = .inform, .parallel = .parallel, .paropts = .paropts)
1: ddply(hfData, c("lake"), metab, method = "bookkeep", irr.name = "irr2")

[lawinslow]

What does hfData look like?

[rBatt]

@lawinslow

dput(head(hfData))

structure(list(lake = structure(c(1L, 1L, 1L, 1L, 1L, 1L), .Label = c("paul", 
"peter"), class = "factor"), datetime = structure(c(1210694400, 
1210694700, 1210695000, 1210695300, 1210695600, 1210695900), class = c("POSIXct", 
"POSIXt"), tzone = "GMT"), year = c(2008, 2008, 2008, 2008, 2008, 
2008), doy = c(134.666666666667, 134.670138888889, 134.673611111111, 
134.677083333333, 134.680555555556, 134.684027777778), do.obs = c(10.4347483191214, 
10.4528878395361, 10.4500595198614, 10.4380570411655, 10.4087379516118, 
10.4055922795976), do.sat = c(10.0747468313655, 10.0999625107891, 
10.0999846075527, 10.0807120402841, 10.0495479240296, 10.0239167387756
), wtr = c(12.379333051731, 12.2693323477265, 12.2692362745664, 
12.3532440129325, 12.4899999039994, 12.6033339093371), doSat = c(97.6598307734584, 
97.5865454017649, 97.5599284817454, 97.6332213454424, 97.6600261833021, 
97.880076761626), z.mix = c(1.5, 1.5, 1.5, 1.5, 1.5, 1.5), k.gas = c(0.492825255571682, 
0.511874243056661, 0.491352605979017, 0.492338819545, 0.475021164297611, 
0.495831029616867), irr = c(147.5, 124.701986754946, 109.867549668811, 
100.016501650173, 102.491749174914, 142.665562913905), wnd = c(1.18393660745817, 
1.40853299870392, 1.18393660745817, 1.18234004295608, 0.942855367755401, 
1.18393660745817), chl = c(2.30265083940554, 1.72931341004522, 
1.56931878608269, 2.00878286367763, 2.53598513274742, 2.22931533006012
), pH = c(6.601958991, 6.621968761, 6.621978531, 6.631988334, 
6.631998072, 6.632007842)), .Names = c("lake", "datetime", "year", 
"doy", "do.obs", "do.sat", "wtr", "doSat", "z.mix", "k.gas", 
"irr", "wnd", "chl", "pH"), row.names = c(NA, 6L), class = "data.frame")

[rBatt]

Except I think I added a column "irr2" which was just the 0L 1L integer form of irr

[lawinslow]

Can you filter the data.frame yourself and just supply it to metab? I doubt that ddply has anything to do with it.

[rBatt]

Yeah, if I subset it on my own (i.e., supply 1 lake-year) and supply it to
metab, it works fine.

I think it's crashing at the conquer list step in metab.

Might have something to do with how ddply() stores the intermediates. Not
sure.

On Sun, Nov 16, 2014 at 3:30 PM, Luke Winslow [email protected]
wrote:

Can you filter the data.frame yourself and just supply it to metab? I
doubt that ddply has anything to do with it.

—
Reply to this email directly or view it on GitHub
#70 (comment)
.

[lawinslow]

Wrap the metab function in a function and stick a breakpoint inside of
it. Then you can look at what it is trying to pass.