plot and image demos (growing window)

[Daniel J Sebald]

Rik, Ben,

I've had an opportunity to try this out with numerous versions of gnuplot.  There is randomness to the window expansion.  That is, it isn't on ever plot that the window expands.  That fact seems to point to gnuplot as the source, and the problem seems to exist as far back as gnuplot version 4.2.0.  However, the expansion is at a much slower rate, approximately once with every 200 to 300 plots.  With gnuplot version 4.2.5 the expansion rate is about once every 3 or 4 plots.  There could be a timing issue made worse by new additions to gnuplot over the years; one that shows up only when the CPU is being completely taxed.

I'll look into the gnuplot side of things (and if you want to send something to the gnuplot discussion list, feel free), but my advice would be to not spend too much effort tracking down exactly where Octave might cause a problem.  It might just be a random arrangement of code that works better than other versions.

Dan
 

Ben Abbott wrote:
> On Jun 1, 2009, at 9:11 PM, Rik wrote:
> 
>>>> The problem definitely seems to be gnuplot related.  For gnuplot  
>>>> version
>>>> 4.2.2, revision 9097 works.  For gnuplot version 4.2.5, revision  9097
>>>> fails.  4.3 won't compile on my Hardy Heron box.  I suppose we  
>>>> could try
>>>> to work-around their bug if we could figure out how we are  
>>>> triggering it
>>>> or we will simply have to let it pass as being outside of Octave.
>>>
>>>
>>> I'm not saying there isn't a bug in gnuplot, but I'm still trying to
>>> understand what the method is for controlling the size of the x11
>>> window and/or mouse.  (Unfortunately, savanah server isn't working so
>>> that I can view the latest code.)
>>>
>>> There is too much in between Octave and gnuplot to conclude this is a
>>> gnuplot bug based on your observation.
>>
>> I really wasn't trying to conclude definitively who was at fault but
>> with both code bases changing simultaneously it is difficult (and
>> possibly not worthwhile) to discover where the fault lies.  In
>> particular, for gnuplot 4.2.2, things break between Octave versions  9097
>> and 9098.
> 
> 
> Can you clarify? What breaks for gnuplot 4.2.2?
> 
>> For gnuplot 4.2.5 and 4.3.0, however, the breakage occurs
>> between Octave versions 8223 and 8224.  I already sent the debug.gp
>> files for gnuplot 4.2.2.  Attached are different debug.gp files for  the
>> breakage between 8223 and 8224.  The difference is the addition in  8224
>> of the following preamble lines:
>>
>>> set multiplot;
>>> set origin 0, 0
>>> set size 1, 1
>>
>>
>> According to the log these changes were done to make the postscript
>> bounding box accurate.
>> changeset:   8224:62d90e049d4f
>> user:        David Bateman <dbateman at free.fr>
>> date:        Thu Oct 16 10:31:37 2008 +0100
>> summary:     Always use multiplot with gnuplot to ensure correct
>> postscript bounding box
> 
> 
> Ok, I'll try experimenting with those three lines to see if I can  
> create a simple gnuplot example to produces the growing window.
> 
> Ben