[Rd] XFIG color output (PR#7491)

[Prof Brian Ripley]

On Mon, 10 Jan 2005 chris.haidinyak at amd.com wrote:

> Full_Name: Chris Haidinyak
> Version: 2.0.1 & 1.9.0
> OS: Linux x86_64
> Submission from: (NULL) (139.95.251.9)
>
>
> Hi,
>
>   I am using R 1.9.0 (on Linux Redhat AS 3 - x86_64) and 2.0.1 (on Linux Fedora
> Core 3 - x86_64) and am having problems using graphs created with the xfig
> option.
>
>
> 1.9.0 problem - sometimes the graph output puts the color definitions in the
> wrong place in the file. According to the xfig program, color definitions must
> come before any object. For whatever reason, R sometimes produces xfig files
> with color definitions in the wrong place. Run demo('graphics') to recreate the
> problem.

I can't.  What exactly are you doing, and in which file does this occur?
I have just tested all of them.

If perchance you were using onefile=TRUE, do read the comment in the help 
file.

Please produce a reproducible example.

> 2.0.1 problem - This release seems worse that 1.9.0. All of the graphic output
> from R to xfig (using the following script
>   xfig()
>   demo('graphics')
>   dev.off()
> produces only b/w graphs or no graph at all.

Paul Murrell altered the alpha channel handling in 2.0.0, and missed this 
one.  I have a patch tested and about to go into 2.0.1-patched.


-- 
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