[R] The best way to end up with WMF files

[Pikounis, Bill]

Hi Peter,

> And if I do create another format first and eventually end up with
> a wmf, what is the best way to get there without loosing quality
> on the way through the conversion(s)?  
> 
> So my question is this:  How can I end up with a wmf files using
> linux, without sacrificing too much (any?) quality on the way?

If I understand your questions correctly, one other potential option comes
from the libEMF ( http://libemf.sourceforge.net/ and
http://sourceforge.net/projects/libemf ) library and a tool called pstoedit
( http://www.pstoedit.net/pstoedit/ ). With these I have had some success
converting PostScript formatted (single) data graphs into Enhanced Windows
Metafile format (EMF) under Linux. My tests have not been extensive,
however, just some simple scatterplots with lowess curves and a couple of
the lattice/grid examples.  The quality of the graphs was acceptable to me
in these simple tests, and the desirable resolution characteristics of
vector-based graphs remained.  

Some fiddling was needed to get pstoedit to recognize libEMF at build time -
I use Mandrake 9.0.  The libEMF library in particular required some patches
to handle gcc 3.x. These patches have been submitted but the libEMF author
has not updated the original tarball yet.  The patches were fairly simple to
apply, although I did not document my path in doing so.  

Also, pstoedit runs under Windows, and the conversion to WMF/EMF is implicit
in its already built installer, so perhaps the biologists would not object
to doing the conversion from PostScript -> WMF directly on their machines?
(No GUI, of course, and ghostscript installation may be required if I
remember correctly, so I understand this may not be feasible for your target
audience. I also recall that the a flag to rotate the graphs was needed.)

Hope that helps,
Bill

----------------------------------------
Bill Pikounis, Ph.D.
Biometrics Research Department
Merck Research Laboratories
PO Box 2000, MailDrop RY84-16  
126 E. Lincoln Avenue
Rahway, New Jersey 07065-0900
USA

v_bill_pikounis at merck.com

Phone: 732 594 3913
Fax: 732 594 1565


> -----Original Message-----
> From: Peter Dunn [mailto:dunn at usq.edu.au]
> Sent: Wednesday, March 19, 2003 6:57 PM
> To: R-help at stat.math.ethz.ch
> Subject: [R] The best way to end up with WMF files
> 
> 
> Hi all
> 
> I am doing some stats work for a group of biologists
> who require windows metafiles (*.wmf) for their publications.
> To create these, I appear to have two choices:
> 
> 1.  Restart my machine in Windows and use  savePlot
> 2.  Keep my machine in linux, save as another format,
>     then convert.
> 
> I'd rather stay in linux; but how do I get wmf files?  I looked
> at using ImageMagick's  convert, but it appears only to *read*
> wmf files and not write them according to the help (not that
> I could get it to read them when I tried... but that's another
> story).  
> 
> And if I do create another format first and eventually end up with
> a wmf, what is the best way to get there without loosing quality
> on the way through the conversion(s)?  
> 
> So my question is this:  How can I end up with a wmf files using
> linux, without sacrificing too much (any?) quality on the way?
> 
> I looked in the Mail archives and couldn't find anything useful.
> 
> Thanks as always,
> 
> P.
> 
> -- 
> Dr Peter Dunn          (USQ CRICOS No. 00244B)
>   Web:    http://www.sci.usq.edu.au/staff/dunn
>   Email:  dunn @ usq.edu.au
> Opinions expressed are mine, not those of USQ.  Obviously...
> 
> ______________________________________________
> R-help at stat.math.ethz.ch mailing list
> https://www.stat.math.ethz.ch/mailman/listinfo/r-help
> 

------------------------------------------------------------------------------