From help-octave-request at bevo dot che dot wisc dot edu  Sun Apr 12 17:02:23 1998
Subject: Re: PLplot package...plscol0 resolution
From: Mike Blatchley <mblatch at rocky dot orci dot com>
To: jcardoso at inescn dot pt
cc: help-octave at bevo dot che dot wisc dot edu
Date: Sun, 12 Apr 1998 16:01:58 -0600 (MDT)

On Sat, 11 Apr 1998, Mike Blatchley wrote:
>
> Regarding my problem with PLplot posted yesterday,
> 
> >  Same thing happens with:
> >  octave:1> grid   (or plot([1:10],[1:10]), etc.)
> >  X Error of failed request:  BadAccess (attempt to access private resource denied)
> >    Major opcode of failed request:  89 (X_StoreColors)
> >    Serial number of failed request:  47
> >    Current serial number in output stream:  62
> 
> I traced the offending statements down to 2 in pldef.m
> 
>   plscol0(15,0,0,0);
>   plscol0(0,255,255,255);
> 
> If I comment these out, I get my plot.  

Well, I've continued tracking down my problem with the plscol0 function.
It turns out that I could crash the standard plplot C demos by calling
plscol0, but only if called after plinit().  Running these demos with the
-debug flag, I see that the colormap is downgraded to r/o after calling
plinit().  

So, in pldef.m, I think the 2 plscol0 lines should be placed above plinit
and all works fine. 

Don't you all love it when I keep answering my own questions :-) !!

> However, if I move a window over 
> the plot, any area of the plot that the window passed over is "erased."  
> Is this just a problem with plplot, or do I still have some underlying 
> problem.

I believe this one has something to do with plplot itself.  If I comment
out the "plSetOpt("np","");" line in pldef.m (i.e. I'll have plplot wait
for a keypress before removing plot window and returning to octave), I can
call the "plend" function at the end of the octave demo scripts, and the
windows are properly restored after windows cross over them.  It must have
something to do with plplot not looking for a signal from X to refresh
until it's in the plend function, or something like that.  I don't
know...... 

I'll keep digging on this one, but will still be interested in any
suggestions.


Thanks,

Mike B.
mblatch at orci dot com