R help - Apr 2000 - dev2bitmap from a script invoked by another process

Issue:
I have a R script that is invoked periodically by another process. In the
script, a graphics is generated and I want to get a the image for the graph
as a jpeg.

Here is a prototypical version of the script (try.R):
plot(1)
dev2bitmap("try.jpg", type = "jpeg")

I am running R-1.0.0 on Solaris 7.

The script is invoked as: R --vanilla < try.R
A file named try.jpg is created but it contains a blank image, not the plot
created from the script.
Device inserted after plot(1) tells me that the current device is
postscript.
If I specify: postscript(file = "try.ps"), a file try.ps is created
that
contains a postscript rendition of the plot. The try.jpg is also created but
is as previously described.
If I run the script in interactive mode, the file try.jpg is created and
contains the plot, as expected.

I know I can create the postscript file and then use ghostscript to convert
it but I was wondering if it is possible to create the jpeg directly from R.

Thanks for any help you might have.
-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-
r-help mailing list -- Read http://www.ci.tuwien.ac.at/~hornik/R/R-FAQ.html
Send "info", "help", or "[un]subscribe"
(in the "body", not the subject !)  To: r-help-request at
stat.math.ethz.ch
_._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._

"Carel, Roland" <rcarel at netgenics.com>
writes:
> plot(1)
> dev2bitmap("try.jpg", type = "jpeg")
...
> The script is invoked as: R --vanilla < try.R
> A file named try.jpg is created but it contains a blank image, not the plot
> created from the script.
...
> I know I can create the postscript file and then use ghostscript to convert
> it but I was wondering if it is possible to create the jpeg directly from
R.
Gah. This is silly. The problem is that dev2bitmap calls dev.print to
print the current device as postscript. When run noninteractively, the
current device *is* postscript, but that is not a replayable driver. A
quick solution is to explicitly start X11() at the start of your
script (which has a number of drawbacks, but might be good enough). A
better fix would be to doctor dev2bitmap by replacing

    dev.print(device=postscript, file=tmp, width=width, height=height,
              pointsize=pointsize, paper="special", horizontal=FALSE,
...)

with

   dev.off() #close (and flush!) PS device
   system(paste("mv Rplots.ps", tmp))

(Untested!)

Or something pretty similar.

-- 
   O__  ---- Peter Dalgaard             Blegdamsvej 3  
  c/ /'_ --- Dept. of Biostatistics     2200 Cph. N   
 (*) \(*) -- University of Copenhagen   Denmark      Ph: (+45) 35327918
~~~~~~~~~~ - (p.dalgaard at biostat.ku.dk)             FAX: (+45) 35327907
-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-
r-help mailing list -- Read http://www.ci.tuwien.ac.at/~hornik/R/R-FAQ.html
Send "info", "help", or "[un]subscribe"
(in the "body", not the subject !)  To: r-help-request at
stat.math.ethz.ch
_._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._