matplotlib-devel Mailing List for matplotlib

>>>>> "Ryan" == Ryan Krauss <rya...@gm...> writes:
 Ryan> One thing I don't seem to be able to do is make the diff
 Ryan> pick up the matplotlibrc changes. The following lines need
 Ryan> to be added to make this feature useful:
I don't know if I failed to mention this somewhere else or not, but
this is because the rc is now autogenerated by setup.py from the
matplotlibrc.template file. So that is where changes like
### Legend
legend.isaxes	:	True
legend.numpoints 	:	4 # the number of points in the legend line
legend.fontsize : 14
legend.pad 	:	0.2 # the fractional whitespace inside the legend border
should go. I added these and they are no in CVS.
Thanks,
JDH

>>>>> "Jouni" == Jouni K Seppanen <jk...@ik...> writes:
 Jouni> Hi, I just read in another thread that the recommended way
 Jouni> to submit patches is to both post them here and on the
 Jouni> Sourceforge tracker. Here's one that I had only posted on
It's a good reminder, since I follow the lists more closely than the
sf site....
 Jouni> the tracker a while ago but that still seems to apply
 Jouni> cleanly to current cvs. It fixes a bug in my previous
 Jouni> boxplot patch (division by zero if you only plot one box),
 Jouni> and changes the return value of boxplot so that you can
 Jouni> modify the different parts of the plot separately, as
 Jouni> suggested in John's commentary to David Haas's original
 Jouni> boxplot patch. Here's an example:
OK, committed to CVS. Thanks!
Checking in lib/matplotlib/axes.py;
/cvsroot/matplotlib/matplotlib/lib/matplotlib/axes.py,v <-- axes.py
new revision: 1.152; previous revision: 1.151
done
JDH

Hi,
I just read in another thread that the recommended way to submit
patches is to both post them here and on the Sourceforge tracker. 
Here's one that I had only posted on the tracker a while ago but that
still seems to apply cleanly to current cvs. It fixes a bug in my
previous boxplot patch (division by zero if you only plot one box),
and changes the return value of boxplot so that you can modify the
different parts of the plot separately, as suggested in John's
commentary to David Haas's original boxplot patch. Here's an example:
In [1]:boxplot(rand(100,2))
Out[1]:
{'boxes': [<matplotlib.lines.Line2D instance at 0x2deaaa8>,
 <matplotlib.lines.Line2D instance at 0x2deae68>],
 'caps': [<matplotlib.lines.Line2D instance at 0x2deaaf8>,
 <matplotlib.lines.Line2D instance at 0x2deab70>,
 <matplotlib.lines.Line2D instance at 0x2dea8f0>,
 <matplotlib.lines.Line2D instance at 0x2def170>],
 'fliers': [<matplotlib.lines.Line2D instance at 0x2dead00>,
 <matplotlib.lines.Line2D instance at 0x2deab48>,
 <matplotlib.lines.Line2D instance at 0x2def2b0>,
 <matplotlib.lines.Line2D instance at 0x2def148>],
 'medians': [<matplotlib.lines.Line2D instance at 0x2deaa58>,
 <matplotlib.lines.Line2D instance at 0x2def210>],
 'whiskers': [<matplotlib.lines.Line2D instance at 0x2dea8a0>,
 <matplotlib.lines.Line2D instance at 0x2dea9b8>,
 <matplotlib.lines.Line2D instance at 0x2deae90>,
 <matplotlib.lines.Line2D instance at 0x2deafd0>]}
In [2]:setp(Out[1]['medians'], 'linewidth', 3)
Out[2]:[None, None]
The patch also changes the boxplot demo to demonstrate this.
-- 
Jouni

Can you send me a short example that shows the problem?
I'm not able to duplicate it here -- but I'm probably not doing the 
right thing.
John
John Gill wrote:
> I'll take a look -- the legends for scatter plots fix was one of mine.
>
> John
>
> Stefan Kuzminski wrote:
>> I notice that my legend in a scatter plot ( the axis legend ) does not
>> display a graphic indication next to the text ( as it does with line
>> plots ). The difference seems to be if the first argument to 'legend'
>> is a list of 'Line2D' objects it works ( the legend looks good ) but if
>> the first argument is a list of 'RegularPolyCollection' the legend
>> contains just text and there is no way to know what it is 'legending'
>>
>> Thanks,
>> Stefan 
>>
>> __________________________________________________
>> Do You Yahoo!?
>> Tired of spam? Yahoo! Mail has the best spam protection around 
>> http://mail.yahoo.com 
>>
>>
>> -------------------------------------------------------
>> This SF.net email is sponsored by: Splunk Inc. Do you grep through log files
>> for problems? Stop! Download the new AJAX search engine that makes
>> searching your log files as easy as surfing the web. DOWNLOAD SPLUNK!
>> http://sel.as-us.falkag.net/sel?cmd=lnk&kid=103432&bid=230486&dat=121642
>> _______________________________________________
>> Matplotlib-devel mailing list
>> Mat...@li...
>> https://lists.sourceforge.net/lists/listinfo/matplotlib-devel
>>