Error on local build(?)

[cboettig]

Hmm, my local (ubuntu-based) build is failing for reasons that aren't obvious to me. Running make gets me almost everything, but fails to make the phylogeny figure which then fails in the manuscript build.

Likewise, sourcing in wood.R gets through most of the images before it produces the following error. Wondering if this error is obvious to you?

> source("wood.R")
Loading required package: deSolve
Loading required package: ape
Loading required package: subplex
Loading required package: Rcpp
Error in n.taxa2[i] <- n.taxa : 
  NAs are not allowed in subscripted assignments

Enter a frame number, or 0 to exit   

1: source("wood.R")
2: withVisible(eval(ei, envir))
3: eval(ei, envir)
4: eval(expr, envir, enclos)
5: wood.R#334: fig.fraction.on.phylogeny(phy.o, res.b)
6: plot-tree.R#55: diversitree:::plot2.phylo(phy.o, type = "fan", cex = 0.5, n
7: pp.node.coords(x)
8: node.height(phy.p, phy$n.taxa)

(Had to toggle the LaTeX \pdfminorversion up to 4 manually on my machine, not idea why that needs specifying in the first place... everything else just worked which is very impressive!) Sorry to be bugging you.

[richfitz]

Can you try with github diversitree? Looks like I switched over to use that on Travis (see here) so there may be a bug there.

The \pdfminorvecsion thing is a workaround to deal with Manuscript Central.

[cboettig]

Thanks, silly me, I should have noticed from the travis.yml that I needed the github version.

Ah, Manuscript Central; good times. It's exercises like submitting to Manuscript Central that maintain my faith that any researcher can learn to use GitHub if only given the incentive.