[Rd] How to point R toward Ghostscript on a Windows XP system
Prof Brian Ripley
ripley at stats.ox.ac.uk
Sat Jan 15 08:59:24 CET 2011
On Fri, 14 Jan 2011, Jeremy Koster wrote:
> A colleague designed a script for a bar plot, which I'd like to
> export to my directory via the barplot command:
>
> bitmap(file="barplot.tif", type="tifflzw", height=4, width=6.5, res=1250)
>
> Unfortunately, this command produces the following error message:
>
> Error in system(paste(gsexe, "-help"), intern = TRUE, invisible = TRUE) :
> 'gswin32c.exe' not found
>
>> From consulting the archives and the help file, I deduced that I
>> needed to install Ghostscript, which I did. Subsequent messages in
>> the archives suggest directing R toward Ghostscript using an
>> environmental variable, specifically R_GSCMD.
Please do read the help page. You didn't tell us if you added
Ghostscript to your path: if you had done so this would not be needed
(as the help page says).
> Unfortunately, I haven't figured out how to use this command. Can
> anyone provide assistance?
Sys.setenv(GS_CMD="C:/Program Files/gs/gswin32c.exe")
if what you say below is correct, but
Sys.setenv(GS_CMD="C:/Program Files/gs/bin/gswin32c.exe")
would be more standard for recent versions of Ghostscript.
> This is for R version 2.12.0 using Windows XP on a Dell laptop.
OK, but next time please consult the posting guide and give us the
'at a minimum' information asked for there (we need to know if this is
32- or 64-bit Windows on a 32- or 64-bit OS -- and although rare,
64-bit Windows XP does exist).
> The path for ghostscript in my program files folder is: C:\Program
> Files\gs
>
> Many thanks,
> Jeremy
> UC Anthropology
BTW, asking how to set environment variables is covered in the rw-FAQ
and is not an R-devel topic.
--
Brian D. Ripley, ripley at stats.ox.ac.uk
Professor of Applied Statistics, http://www.stats.ox.ac.uk/~ripley/
University of Oxford, Tel: +44 1865 272861 (self)
1 South Parks Road, +44 1865 272866 (PA)
Oxford OX1 3TG, UK Fax: +44 1865 272595
More information about the R-devel
mailing list