3.0.4 call for patches

[Ben Abbott]

On Dec 12, 2008, at 4:42 PM, Ben Abbott wrote:

> On Friday, December 12, 2008, at 04:20PM, "Thomas Weber" <thomas.weber.mail at gmail.com 
> > wrote:
>> On Fri, Dec 12, 2008 at 02:31:29PM +0100, Jaroslav Hajek wrote:
>>> On Thu, Dec 11, 2008 at 11:41 PM, Thomas Weber
>>> <thomas.weber.mail at gmail.com> wrote:
>>>> On Mon, Dec 08, 2008 at 07:49:34AM +0100, Jaroslav Hajek wrote:
>>>>> hello,
>>>>>
>>>>> I think time has come to get 3.0.4 out. We have >30 patches since
>>>>> 3.0.3 and maybe more omitted. Anyone cares of a particular fix  
>>>>> to be
>>>>> applied to 3.0.4? Please point me to it.
>>>>
>>>> Okay, here's one (octgpr's demos fail, so you might be  
>>>> interested :) )
>>>> 0a48abc35932
>>>>
>>>> Test case is contourf's documentation:
>>>>       [x, y, z] = peaks (50);
>>>>       contourf (x, y, z, -7:9)
>>>> fails in 3.0.3 for me.
>>>>
>>>>       Thomas
>>>>
>>>
>>> I don't have a clue what is causing this error. Does someone else?
>>> Even the current development tip seems to be affected.
>>
>> Uh, you guys have lost me completely. The error message I was talking
>> about has nothing to do with gnuplot, but is pure Octave:
>>
>> =====================================================================
>> octave:2>  contourf (x, y, z, -7:9)
>> warning: meaning may have changed due to change in precedence for  
>> && and || operators
>> error: patch: X and Y must be of same size
>> error: evaluating if command near line 230, column 3
>> error: called from `contourf:parse_args' in file `/usr/share/octave/ 
>> 3.0.3/m/plot/contourf.m'
>> error: called from `contourf' in file `/usr/share/octave/3.0.3/m/ 
>> plot/contourf.m'
>> =====================================================================
>>
>> Sorry for being unclear in the first place.
>>
>> 	Thomas
>
> Line 230 of contourf is
>
> if (isvector (X) || isvector (Y) && any (size (X) != size(Y)))
>
> Replace that with
>
> if ((isvector (X) || isvector (Y)) && any (size (X) != size(Y)))
>
> and the problem should be fixed (provided I understand the intended  
> logic).
>
> Ben

Ok, I'm not sitting at my mac with a recent build from the developer's  
sources. All I see is a cyan rectangle as well.

I thought the problem might be the order of the patch rendering, but  
reversing them did not good ... then I noticed that problem has been  
fixed by *me* a few months back.

David, I've cc'd you. Since you applied your patch for adding contour  
group objects something has changed, any thoughts?

Ben