On Sun, Sep 12, 2010 at 8:47 AM, Duncan Murdoch <murdoch.dun...@gmail.com> wrote: > On 12/09/2010 11:41 AM, Michael Friendly wrote: >> >> Just a follow-up on this thread, now with R 2.11.1. I was happy back then >> to use Deepayan's solution for this, under earlier R versions; but it now >> gives an error and the Sweave-generated .tex file no longer compiles. >> >> <<ortho-xyplot1-code, keep.source=TRUE, eval=FALSE>>= >> library(nlme) >> library(lattice) >> xyplot(distance ~ age|Sex, data=Orthodont, type='b', groups=Subject, >> pch=15:25, >> col=palette(), cex=1.3, main="Orthodont data") >> @ >> <<ortho-xyplot1,fig=TRUE, include=FALSE, echo=FALSE, width=6, height=6>>= >> plot(trellis.last.object()) >> @ >> >> Now, I get the following from Sweave: >> >> Error: chunk 46 (label=ortho-xyplot1) >> Error in !lattice.getStatus("current.plot.saved", prefix = prefix) : >> invalid argument type >> >> Did something in lattice change so that plot(trellis.last.object()) no >> longer works or needs to be specified differently? > > I don't know if something in lattice has changed, but that doesn't look as > though it should work: the code to do the plot was never evaluated. I > think you would need eval=TRUE in the first chunk, or > <<ortho-xyplot1-code>> > ahead of the plot command in the second.
Yes. I notice that it was me who introduced the eval=FALSE (I'm not sure why). Hopefully that should fix it. However, there _has_ been a recent lattice update, where the internals have undergone some change, so there is a chance that is to blame. I won't be able to take a look before day after tomorrow. -Deepayan ______________________________________________ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.