pcolor and "interp" [changeset-2]

[Ben Abbott]

On Sep 29, 2008, at 5:33 PM, David Bateman wrote:

> Ben Abbott wrote:
>> On Sep 29, 2008, at 3:34 PM, David Bateman wrote:
>>> John W. Eaton wrote:
>>>
>>>> Perhaps it is now time to think about removing the  
>>>> have_newer_gnuplot
>>>> variable from __go_draw_axes__?  This was useful when we were  
>>>> making
>>>> the Octave 3.0 release around the time that gnuplot 4.2 was  
>>>> released,
>>>> but now it has been a while, and I think we can expect people to  
>>>> have
>>>> the newer version.  Removing this variable would simplify
>>>> __go_draw_axes__ somewhat.  I think we will cause more trouble for
>>>> ourselves if we proceed down the path of checking every micro  
>>>> version
>>>> of gnuplot and trying to support them all.
>>>>
>>> I'd also suggest dumping the code for gnuplot versions older than   
>>> 4.2 in __go_draw_axes__. It made sense to have it for Octave  
>>> 3.0.x,  but if you are using a bleeding edge version of Octave you  
>>> should  have a version of gnuplot that hasn't been obsolete for 18  
>>> months.  gnuplot 4.2.0 was released 18 months ago.
>>>
>>> Regards
>>> David
>> John,
>> I have no problem with requiring a fully functional gnuplot. As  
>> I've  managed to patch 4.2.3 the version checking is a disabler,  
>> rather than  an enabler.
>> David,
>> Are you suggesting that we drop the present use of   
>> "have_newer_gnuplot" but still handle the differences between  
>> versions  < 4.2.4 and those >= 4.2.4, or suggesting that *all*  
>> version checking  for gnuplot should be dropped?
>
> The first of your options..
>
> D.

ok, then is there any objection to patching __gnuplot_version__ to  
include the patch level?

Ben