R-beta: Encapsulated Postscript output with R-0.49

Steve Shiboski (steve@biostat.ucsf.edu)
Sun, 8 Jun 97 15:12:36 PDT

Date: Sun, 8 Jun 97 15:12:36 PDT
From: steve@biostat.ucsf.edu (Steve Shiboski)
Message-Id: <9706082212.AA01994@biostat.ucsfbiostat>
To: r-help@stat.math.ethz.ch
Subject: R-beta: Encapsulated Postscript output with R-0.49

On Sun, 8 Jun 1997, Thomas Lumley wrote:

>I haven't had any of these problems.  My postscript files come up the
>right way around in ghostview and import beautifully into LaTeX with
>\epsfig.   They are also in colour (which I have only just worked out how
>to do in S-PLUS).
>Most, if not all, of my postscript files were produced by save.plot()
>rather than by postscript(). This may explain the difference, but they do
>include the %%Orientation, %%Page, /ep, and so on.   Also, in earlier
>versions of R I used postscript() with no problems.
>What LaTeX commands and dvi converter are you using? I have used both the
>NTeX (on Linux) and teTeX versions of LaTeX2e with the dvips converter,
>and used the epsfig package.

Thanks for your comments.

I use the following LaTeX commands to import an eps file
into a document:

I have used this method for years with no problems.

I mainly use my old NeXT and the version of LaTeX and the
dvi previewer (and dvips) which was bundled with the system.
This could account for some of the weird behavior. Perhaps
something is different with the version of Postscript used.

I just tried the same commands on my Sun at work running a
recent version of teTex (LaTeX2e), and the imported figure
(in either the dvi file or the postscript image produced by dvips)
is still oriented and sized improperly (when viewed on my NeXT). 
I'll have to try viewing these on the Sun tomorrow. If I use the
on the Sun, things look better, but the orientation is still off.

In any case, after I run R postscript files through my Perl filter, 
they import properly and can also be used to paste into conventional
word processing and graphics programs (e.g. WordPerfect). I looked 
around on the Web and saw references to a program called "ps2epsi"
which is supposed to work well for similar problems encountered
with postscript output produced by S-PLUS. Apparently it includes
a bitmap vesrion of the eps image so that it looks approximately
correct in previewers like Ghostview.

It's very possible that I am the only R user experiencing these 
problems. If so, all this should probably be ignored.

Stephen Shiboski                          email: steve@biostat.ucsf.edu
Division of Biostatistics                 voice: 415-476-0533
University of California San Francisco    fax:   415-476-6014
500 Parnassus Avenue, MU 420 West;        San Francisco, CA 94143-0560

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@stat.math.ethz.ch