Re: changes to graphics.cc and subplot.m

[Ben Abbott]

On Jul 21, 2013, at 7:49 PM, Ben Abbott wrote:

> On Jul 21, 2013, at 1:53 PM, Stefan Mahr wrote:
> 
>>> Looking a second time, I see Dimitri's result only differs in the 4th 
>>> decimal place, while Stefan's differs in the first decimal place.  Stefan, 
>>> can you try a fresh build and confirm your result persists?
>>> 
>> 
>> I get the same result with a fresh rebuild. My Ubuntu PPA build (always
>> fresh clone+bootstrap) shows this behavior as well.
>> 
>> Dimitri's plot looks very similar to the output of my first plot. The
>> y-axis of the lower plot is smaller than the y-axis of the plot above.
>> 
>> Just found out that the error occurs every time when a new plot window
>> is opened, not only on fresh restart of octave as I said before. It
>> doesn't matter if I close the old figure or open a new one with figure(x).
>> 
>> graphics_toolkit fltk
>> figure
>> subplot(2,1,1); plot(rand(10,1)); subplot(2,1,2); plot(rand(10,1))
>> %% plot looks normal
>> 
>> close all
>> subplot(2,1,1); plot(rand(10,1)); subplot(2,1,2); plot(rand(10,1))
>> %% the y-axis of lower plot is to small
>> 
>> subplot(2,1,1); plot(rand(10,1)); subplot(2,1,2); plot(rand(10,1))
>> %% both plots have small y-axis
>> 
>> Stefan
> 
> Thanks! ... I'm now able to write a test script that clearly demonstrates the 
> problem.  "test stefan" fails for me as well.
> 
> I'll push a change to add this test to graphics.cc and file a bug report.
> 
> Ben
> 
> <stefan.m>

oops!  The test only fails for me the 1st time it is run, and only because the 
axes handles are in the wrong order.

Ben