Re: [Matplotlib-users] numpoints in legend() function for scatter plot is not working in matplotlib 1.1.0?

2012-09-10 Thread Chao YUE
Hi Ben, I guess it's this: commit 8c57e4fe4909815092af470f4a036c80b407382c Merge: cf53d4c b6c5625 Author: Eric Firing Date: Mon Jul 9 15:25:16 2012 -0700 Merge pull request #994 from efiring/fix_pcolorfast_quadmesh Fix bug in pcolorfast introduced by #901 cheers, Chao On Mon, Sep

Re: [Matplotlib-users] numpoints in legend() function for scatter plot is not working in matplotlib 1.1.0?

2012-09-10 Thread Phil Elson
Ben, on master scatter legends are broken. A pull request (I can't remember which, and github is down, possibly #1176) fixes this (obviously, the PR is marked as release critical for 1.2). I have confirmed, that even once fixed on master, the scatter legend handler does not take notice of the "num

Re: [Matplotlib-users] numpoints in legend() function for scatter plot is not working in matplotlib 1.1.0?

2012-09-10 Thread Benjamin Root
On Mon, Sep 10, 2012 at 9:33 AM, Chao YUE wrote: > Hi Ben, > > I think I installed this version by following the instructions on > matplotlib website. But when I try to use git log, I get: > > chaoyue@chaoyue-Aspire-4750:/usr/local/lib/python2.7/dist-packages/matplotlib$ > pwd > /usr/local/lib/py

Re: [Matplotlib-users] numpoints in legend() function for scatter plot is not working in matplotlib 1.1.0?

2012-09-10 Thread Chao YUE
Hi Ben, I think I installed this version by following the instructions on matplotlib website. But when I try to use git log, I get: chaoyue@chaoyue-Aspire-4750:/usr/local/lib/python2.7/dist-packages/matplotlib$ pwd /usr/local/lib/python2.7/dist-packages/matplotlib chaoyue@chaoyue-Aspire-4750:/us

Re: [Matplotlib-users] numpoints in legend() function for scatter plot is not working in matplotlib 1.1.0?

2012-09-10 Thread Benjamin Root
On Mon, Sep 10, 2012 at 2:56 AM, Chao YUE wrote: > Hi Ben, > > I tried the "numpoints" in legend function for scatter plot, in dev > version and GTKAgg backend it works. > > In [3]: mat.__version__ > Out[3]: '1.2.x' > > In [4]: mat.get_backend() > Out[4]: 'GTKAgg' > > Chao > > Strange, when I tes