[R] R 2.7.0: pdf() > pdf.options versus formals

[Duncan Murdoch]

Hans-Joerg Bibiko wrote:
> Hi,
>
> I have a tiny question about the graphics change in R 2.7.0.
>
> If I write a script à la:
>
> options(device="pdf")
> formals(pdf)[c("file","onefile","width","height")] <- list("~/Rplot 
> %03d.pdf", FALSE, 8, 8)
> plot(3)
> plot(4)
>
> both plots will be written into ONE pdf file. (not two as expected)
> This worked fine under 2.6.x
>
> In R 2.7.0 I have to write:
>
> options(device="pdf")
> formals(pdf)[c("file","onefile","width","height")] <- list("~/Rplot 
> %03d.pdf", FALSE, 8, 8)
> pdf.options(onefile=F)
> plot(3)
> plot(4)
>
> In the NEWS doc for R 2.7.0 is written:
> pdf() now takes defaults from the new function pdf.options() ...
>
> Fine. But is there no way to use formals in that context (or more  
> precisely to preset 'onefile' because other arguments are taken from  
> formals)?
>   
As Brian said, no.  But if you hadn't tried to be so clever, you 
wouldn't have this problem.  You can write a new function that calls 
pdf(), and
give it whatever defaults you like.  For example,

pdf <- function(file="foo", onefile=FALSE, width=8, height=8,...)
  grDevices::pdf(file=file, onefile=onefile, width=width, height=height, 
...)

As far as the original device function is concerned, file, onefile, 
width and height will never be missing, so the original default
values will never be used.  And because this approach doesn't modify the 
implementation of a base function, it
should be more robust to changes there:  we're fairly unlikely to change 
the interpretation of any of those 4 parameters, and would
not do so without NEWS notice in advance.

Duncan Murdoch

 
> I have several scripts written for R 2.6 and that would mean that I  
> have to rewrite all of them.
>
> R version 2.7.0 (2008-04-22)
> powerpc-apple-darwin8.10.1
>
> locale:
> en_US.UTF-8/en_US.UTF-8/C/C/en_US.UTF-8/en_US.UTF-8
>
> attached base packages:
> [1] stats     graphics  grDevices utils     datasets  methods   base
>
>
> Many thanks in advance,
>
> --Hans
> ______________________________________________
> R-help at 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.
>