From efiring at hawaii.edu  Sun Jan  7 15:24:25 2018
From: efiring at hawaii.edu (Eric Firing)
Date: Sun, 7 Jan 2018 10:24:25 -1000
Subject: [Matplotlib-devel] image tests fail locally on my OSX system
Message-ID: <e3f8c513-400a-514b-12d9-750aa3245eba@hawaii.edu>

On my normal environment, OSX 10.12.6 on a Macbook Pro Retina, I am 
getting multiple image test failures that are *not* just the usual 
freetype problems.  (I am using 'export MPLLOCALFREETYPE=1' in my 
.bashrc.)  The failures are most visible in the diffs of the spectrum plots.

Is anyone else seeing this, or are others on OSX able to get all of the 
tests to pass?

Eric

From jklymak at uvic.ca  Sun Jan  7 15:46:40 2018
From: jklymak at uvic.ca (Jody Klymak)
Date: Sun, 7 Jan 2018 12:46:40 -0800
Subject: [Matplotlib-devel] image tests fail locally on my OSX system
In-Reply-To: <e3f8c513-400a-514b-12d9-750aa3245eba@hawaii.edu>
References: <e3f8c513-400a-514b-12d9-750aa3245eba@hawaii.edu>
Message-ID: <9D02F4FE-A489-49A4-8F2D-89E4F0C39EF0@uvic.ca>

Hi Eric,

I?m on 10.13.2 and getting only an error in

```
lib/matplotlib/tests/test_font_manager.py ....F....

========================================================== FAILURES ==========================================================
_________________________________________________ test_get_fontconfig_fonts __________________________________________________

    @pytest.mark.skipif(not has_fclist, reason='no fontconfig installed')
    def test_get_fontconfig_fonts():
>       assert len(get_fontconfig_fonts()) > 1
E       assert 0 > 1
E        +  where 0 = len([])
E        +    where [] = get_fontconfig_fonts()

lib/matplotlib/tests/test_font_manager.py:88: AssertionError
============================================= 1 failed, 8 passed in 1.73 seconds =============================================
```

All other tests pass on Master for me.  

Cheers,   Jody


> On Jan 7, 2018, at  12:24 PM, Eric Firing <efiring at hawaii.edu> wrote:
> 
> On my normal environment, OSX 10.12.6 on a Macbook Pro Retina, I am getting multiple image test failures that are *not* just the usual freetype problems.  (I am using 'export MPLLOCALFREETYPE=1' in my .bashrc.)  The failures are most visible in the diffs of the spectrum plots.
> 
> Is anyone else seeing this, or are others on OSX able to get all of the tests to pass?
> 
> Eric
> _______________________________________________
> Matplotlib-devel mailing list
> Matplotlib-devel at python.org
> https://mail.python.org/mailman/listinfo/matplotlib-devel


From efiring at hawaii.edu  Sun Jan  7 15:51:53 2018
From: efiring at hawaii.edu (Eric Firing)
Date: Sun, 7 Jan 2018 10:51:53 -1000
Subject: [Matplotlib-devel] image tests fail locally on my OSX system
In-Reply-To: <9D02F4FE-A489-49A4-8F2D-89E4F0C39EF0@uvic.ca>
References: <e3f8c513-400a-514b-12d9-750aa3245eba@hawaii.edu>
 <9D02F4FE-A489-49A4-8F2D-89E4F0C39EF0@uvic.ca>
Message-ID: <2bb5d534-9bd2-0e42-2dfe-3e065d4be256@hawaii.edu>

Jody,

Are you installing natively, or in a conda environment?  (I'm doing 
everything in a conda environment.)

Eric

On 2018/01/07 10:46 AM, Jody Klymak wrote:
> Hi Eric,
> 
> I?m on 10.13.2 and getting only an error in
> 
> ```
> lib/matplotlib/tests/test_font_manager.py ....F....
> 
> ========================================================== FAILURES ==========================================================
> _________________________________________________ test_get_fontconfig_fonts __________________________________________________
> 
>      @pytest.mark.skipif(not has_fclist, reason='no fontconfig installed')
>      def test_get_fontconfig_fonts():
>>        assert len(get_fontconfig_fonts()) > 1
> E       assert 0 > 1
> E        +  where 0 = len([])
> E        +    where [] = get_fontconfig_fonts()
> 
> lib/matplotlib/tests/test_font_manager.py:88: AssertionError
> ============================================= 1 failed, 8 passed in 1.73 seconds =============================================
> ```
> 
> All other tests pass on Master for me.
> 
> Cheers,   Jody
> 
> 
>> On Jan 7, 2018, at  12:24 PM, Eric Firing <efiring at hawaii.edu> wrote:
>>
>> On my normal environment, OSX 10.12.6 on a Macbook Pro Retina, I am getting multiple image test failures that are *not* just the usual freetype problems.  (I am using 'export MPLLOCALFREETYPE=1' in my .bashrc.)  The failures are most visible in the diffs of the spectrum plots.
>>
>> Is anyone else seeing this, or are others on OSX able to get all of the tests to pass?
>>
>> Eric
>> _______________________________________________
>> Matplotlib-devel mailing list
>> Matplotlib-devel at python.org
>> https://mail.python.org/mailman/listinfo/matplotlib-devel
> 


From tcaswell at gmail.com  Mon Jan  8 17:01:37 2018
From: tcaswell at gmail.com (Thomas Caswell)
Date: Mon, 08 Jan 2018 22:01:37 +0000
Subject: [Matplotlib-devel] 2018-01-08 call notes
Message-ID: <CAA48SF-TGbJbW9g-qZZ_av--t45T4=XLGr8zJicFDpmkbMj2SA@mail.gmail.com>

On call: Myself, Eric Firing, Jody Klymak, Paul Hobson

 - goal to have 2.1.2 by end of week
 - went over high bit-depth units in imshow
 - went over qt5 issue 9040
 - covered doc review policy
   - single review
   - merge if it is an improvement
   - do substantial copy edit via push or PR instead of in comments
 - discussed 10193 and 10033
 - talked about constraint manager

Calls are weekly at 3pm eastern time on google hangouts.  If you want to
join ping me and I'll get you an invitation.

Tom
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/matplotlib-devel/attachments/20180108/ca7a8c98/attachment.html>

From tcaswell at gmail.com  Mon Jan  8 17:31:19 2018
From: tcaswell at gmail.com (Thomas Caswell)
Date: Mon, 08 Jan 2018 22:31:19 +0000
Subject: [Matplotlib-devel] matplotilb 2.2 prep
Message-ID: <CAA48SF8Ch=K_jXRAvhS+68cKSbhjHQa050Rn3qLsr2H9pLWdLQ@mail.gmail.com>

Folks,

We are aiming to have a a 2.2.0rc1 around Jan 30.  To that end please:

  - review and merge documentation PRs.  Documentation changes can be
merged by the first reviewer with the criteria "is it better than it was?"
( https://matplotlib.org/devel/coding_guide.html#pr-review-guidelines )
 - Tag any important issues or PRs you are aware of as 'release critical'
to ensure they get looked at.
 - move any issues and PRs you don't think are critical to 2.2 and won't be
patched or reviewed in the next two-three weeks to the 3.0 milestone.

If you have a commit bit just make the changes, if not please leave
comments asking for the changes

This will be the last minor release to support python 2.7 and will be the
supported with (critical) bug fixes to 2020.  The next release after 2.2
will be 3.0 targeted for ~July 2018.  The only major API change will be
dropping python2 support and will otherwise be a normal feature release.

Tom
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/matplotlib-devel/attachments/20180108/84e648c8/attachment.html>

From tcaswell at gmail.com  Mon Jan  8 22:37:35 2018
From: tcaswell at gmail.com (Thomas Caswell)
Date: Tue, 09 Jan 2018 03:37:35 +0000
Subject: [Matplotlib-devel] call for GSoC mentors
Message-ID: <CAA48SF_Y-+ba3xEnHAbwQiBgJOvLWHEE2-HdJb-_vyv+Rj1v2Q@mail.gmail.com>

Folks,

Is anyone interested in mentoring for GSoC this year?  If we are going to
take a student I would like at least 2 people other than me to volunteer to
be a mentor per student.

We need to get an initial response to numFOCUS in the next week or so.

Tom
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/matplotlib-devel/attachments/20180109/9586096f/attachment.html>

From tcaswell at gmail.com  Thu Jan 18 00:06:11 2018
From: tcaswell at gmail.com (Thomas Caswell)
Date: Thu, 18 Jan 2018 05:06:11 +0000
Subject: [Matplotlib-devel] [REL] v2.1.2
Message-ID: <CAA48SF8mBY+4NqooeOz0Gg5vXyk4PxPcSFJ3cjW12FTHpzRxqQ@mail.gmail.com>

Folks,

Just tagged and pushed v2.1.2.

Kicked off the mac/linux wheel building system, emailed Christoph, and put
a PR into conda-forge for the new binaries.

Will upload to pypi, updated the docs, and announce widely tomorrow night.

Thanks to everyone who worked on this.

On ward to 2.2.  If people have time please help with review / triage of
what is critical for 2.2 and what needs to be pushed to 3.0.

Tom
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/matplotlib-devel/attachments/20180118/8e66eded/attachment.html>

From sandro.tosi at gmail.com  Thu Jan 18 19:26:52 2018
From: sandro.tosi at gmail.com (Sandro Tosi)
Date: Thu, 18 Jan 2018 19:26:52 -0500
Subject: [Matplotlib-devel] [REL] v2.1.2
In-Reply-To: <CAA48SF8mBY+4NqooeOz0Gg5vXyk4PxPcSFJ3cjW12FTHpzRxqQ@mail.gmail.com>
References: <CAA48SF8mBY+4NqooeOz0Gg5vXyk4PxPcSFJ3cjW12FTHpzRxqQ@mail.gmail.com>
Message-ID: <CAB4XWXwkdWv3xXEqxmLhYt4LYY3z3hjJKXAcBa3+7uFPRbjsVQ@mail.gmail.com>

hello everyone,
while preparing the package for Debian i noticed that somewhere
between 2.0.0 and 2.1.1 mpl no longer read /etc/matplotlibrc (we
install that file to give a basic settings to every user without
requiring them to create a custom .config/matplotlibrc), or doesnt
understand it correctly (but i'm not sure about this last point, as i
copied matplotlibrc.template replacing the backend value and still
doesnt work).

with 2.0.0, running

python /usr/share/doc/python-matplotlib-doc//examples/shapes_and_collections/line_collection.py

shows the 2 images just fine, but with 2.1.1 installed:

python /usr/share/doc/python-matplotlib-doc//examples/shapes_and_collections/line_collection.py
/usr/lib/python2.7/dist-packages/matplotlib/__init__.py:941:
UserWarning: could not find rc file; returning defaults
 warnings.warn(message)

and of course no image is shown. backend is TkAgg in both cases.

do you know what's going on?

thanks in advance!

On Thu, Jan 18, 2018 at 12:06 AM, Thomas Caswell <tcaswell at gmail.com> wrote:
> Folks,
>
> Just tagged and pushed v2.1.2.
>
> Kicked off the mac/linux wheel building system, emailed Christoph, and put a
> PR into conda-forge for the new binaries.
>
> Will upload to pypi, updated the docs, and announce widely tomorrow night.
>
> Thanks to everyone who worked on this.
>
> On ward to 2.2.  If people have time please help with review / triage of
> what is critical for 2.2 and what needs to be pushed to 3.0.
>
> Tom
>
> _______________________________________________
> Matplotlib-devel mailing list
> Matplotlib-devel at python.org
> https://mail.python.org/mailman/listinfo/matplotlib-devel
>



-- 
Sandro "morph" Tosi
My website: http://sandrotosi.me/
Me at Debian: http://wiki.debian.org/SandroTosi
G+: https://plus.google.com/u/0/+SandroTosi

From quantum.analyst at gmail.com  Thu Jan 18 20:07:59 2018
From: quantum.analyst at gmail.com (Elliott Sales de Andrade)
Date: Thu, 18 Jan 2018 20:07:59 -0500
Subject: [Matplotlib-devel] [REL] v2.1.2
In-Reply-To: <CAB4XWXwkdWv3xXEqxmLhYt4LYY3z3hjJKXAcBa3+7uFPRbjsVQ@mail.gmail.com>
References: <CAA48SF8mBY+4NqooeOz0Gg5vXyk4PxPcSFJ3cjW12FTHpzRxqQ@mail.gmail.com>
 <CAB4XWXwkdWv3xXEqxmLhYt4LYY3z3hjJKXAcBa3+7uFPRbjsVQ@mail.gmail.com>
Message-ID: <CAJP_izet_u7K9YbA9P6aN296WXaistReiMhPszrg_1uD2hoxMg@mail.gmail.com>

Hi Sandro,

I don't think Matplotlib searches in that location by default. Doesn't
Debian also use the patch to modify the search path that we use on Fedora?

Something like this one (rebased for 2.1.1):
https://github.com/QuLogic/matplotlib/commit/58d84f95955556f7d089966e6cf7740818e9749e

On Jan 18, 2018 7:27 PM, "Sandro Tosi" <sandro.tosi at gmail.com> wrote:

> hello everyone,
> while preparing the package for Debian i noticed that somewhere
> between 2.0.0 and 2.1.1 mpl no longer read /etc/matplotlibrc (we
> install that file to give a basic settings to every user without
> requiring them to create a custom .config/matplotlibrc), or doesnt
> understand it correctly (but i'm not sure about this last point, as i
> copied matplotlibrc.template replacing the backend value and still
> doesnt work).
>
> with 2.0.0, running
>
> python /usr/share/doc/python-matplotlib-doc//examples/
> shapes_and_collections/line_collection.py
>
> shows the 2 images just fine, but with 2.1.1 installed:
>
> python /usr/share/doc/python-matplotlib-doc//examples/
> shapes_and_collections/line_collection.py
> /usr/lib/python2.7/dist-packages/matplotlib/__init__.py:941:
> UserWarning: could not find rc file; returning defaults
>  warnings.warn(message)
>
> and of course no image is shown. backend is TkAgg in both cases.
>
> do you know what's going on?
>
> thanks in advance!
>
> On Thu, Jan 18, 2018 at 12:06 AM, Thomas Caswell <tcaswell at gmail.com>
> wrote:
> > Folks,
> >
> > Just tagged and pushed v2.1.2.
> >
> > Kicked off the mac/linux wheel building system, emailed Christoph, and
> put a
> > PR into conda-forge for the new binaries.
> >
> > Will upload to pypi, updated the docs, and announce widely tomorrow
> night.
> >
> > Thanks to everyone who worked on this.
> >
> > On ward to 2.2.  If people have time please help with review / triage of
> > what is critical for 2.2 and what needs to be pushed to 3.0.
> >
> > Tom
> >
> > _______________________________________________
> > Matplotlib-devel mailing list
> > Matplotlib-devel at python.org
> > https://mail.python.org/mailman/listinfo/matplotlib-devel
> >
>
>
>
> --
> Sandro "morph" Tosi
> My website: http://sandrotosi.me/
> Me at Debian: http://wiki.debian.org/SandroTosi
> G+: https://plus.google.com/u/0/+SandroTosi
> _______________________________________________
> Matplotlib-devel mailing list
> Matplotlib-devel at python.org
> https://mail.python.org/mailman/listinfo/matplotlib-devel
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/matplotlib-devel/attachments/20180118/1a68b603/attachment.html>

From jks at iki.fi  Fri Jan 19 12:29:34 2018
From: jks at iki.fi (=?iso-8859-1?Q?Jouni_K=2E_Sepp=E4nen?=)
Date: Fri, 19 Jan 2018 19:29:34 +0200
Subject: [Matplotlib-devel] CI: only run macosx on travis cron jobs /
 only branches
References: <CAA48SF9_dD=1PEit7DGiH96ZhehKF8ff=ohiSZs0MG-249YaOw@mail.gmail.com>
 <be0e285b-c983-2046-ae67-4b66c3b95893@hawaii.edu>
Message-ID: <m2tvvhdaf5.fsf@jks.iki.fi>

Eric Firing <efiring at hawaii.edu> writes:

> On 2017/09/26 7:22 AM, Thomas Caswell wrote:
>> Folks,
>>
>> https://blog.travis-ci.com/2017-09-22-macos-update
>>
>> Due to the travis OSX infrastructure not being able to keep up with
>> demand they are going to limit concurrency of jobs that us osx.

It's been pretty bad recently. Here's one incident:

https://www.traviscistatus.com/incidents/mc9x2wmpnvhg

That one is supposedly resolved, but it's going to take time for Travis
to work through the backlog.

One idea could be to separate the macosx backend into a different
repository and only run mac builds on that one.


Another idea: CircleCI offers the "Seed plan" for Mac OS open source
projects, but we'd need to contact them to request it:

https://circleci.com/pricing/#faq-section-os-x

That plan includes 500 minutes per month, which isn't very much, but
perhaps we could forgo builds of every revision and just trigger a build
every day or every two days, depending on how long the build would take.

-- 
Jouni K. Sepp?nen
http://www.iki.fi/jks


From tcaswell at gmail.com  Sat Jan 20 23:31:11 2018
From: tcaswell at gmail.com (Thomas Caswell)
Date: Sun, 21 Jan 2018 04:31:11 +0000
Subject: [Matplotlib-devel] [REL] v2.1.2
In-Reply-To: <CAJP_izet_u7K9YbA9P6aN296WXaistReiMhPszrg_1uD2hoxMg@mail.gmail.com>
References: <CAA48SF8mBY+4NqooeOz0Gg5vXyk4PxPcSFJ3cjW12FTHpzRxqQ@mail.gmail.com>
 <CAB4XWXwkdWv3xXEqxmLhYt4LYY3z3hjJKXAcBa3+7uFPRbjsVQ@mail.gmail.com>
 <CAJP_izet_u7K9YbA9P6aN296WXaistReiMhPszrg_1uD2hoxMg@mail.gmail.com>
Message-ID: <CAA48SF8N6Q0QtHsUJ8CdFee9=kDCEoeUsrugd+sGXR0BbNSD_g@mail.gmail.com>

As an update, finalizing the release is waiting on travis to finish
building the mac wheels.

Tom

On Thu, Jan 18, 2018 at 8:07 PM Elliott Sales de Andrade <
quantum.analyst at gmail.com> wrote:

> Hi Sandro,
>
> I don't think Matplotlib searches in that location by default. Doesn't
> Debian also use the patch to modify the search path that we use on Fedora?
>
> Something like this one (rebased for 2.1.1):
> https://github.com/QuLogic/matplotlib/commit/58d84f95955556f7d089966e6cf7740818e9749e
>
> On Jan 18, 2018 7:27 PM, "Sandro Tosi" <sandro.tosi at gmail.com> wrote:
>
>> hello everyone,
>> while preparing the package for Debian i noticed that somewhere
>> between 2.0.0 and 2.1.1 mpl no longer read /etc/matplotlibrc (we
>> install that file to give a basic settings to every user without
>> requiring them to create a custom .config/matplotlibrc), or doesnt
>> understand it correctly (but i'm not sure about this last point, as i
>> copied matplotlibrc.template replacing the backend value and still
>> doesnt work).
>>
>> with 2.0.0, running
>>
>> python
>> /usr/share/doc/python-matplotlib-doc//examples/shapes_and_collections/line_collection.py
>>
>> shows the 2 images just fine, but with 2.1.1 installed:
>>
>> python
>> /usr/share/doc/python-matplotlib-doc//examples/shapes_and_collections/line_collection.py
>> /usr/lib/python2.7/dist-packages/matplotlib/__init__.py:941:
>> UserWarning: could not find rc file; returning defaults
>>  warnings.warn(message)
>>
>> and of course no image is shown. backend is TkAgg in both cases.
>>
>> do you know what's going on?
>>
>> thanks in advance!
>>
>> On Thu, Jan 18, 2018 at 12:06 AM, Thomas Caswell <tcaswell at gmail.com>
>> wrote:
>> > Folks,
>> >
>> > Just tagged and pushed v2.1.2.
>> >
>> > Kicked off the mac/linux wheel building system, emailed Christoph, and
>> put a
>> > PR into conda-forge for the new binaries.
>> >
>> > Will upload to pypi, updated the docs, and announce widely tomorrow
>> night.
>> >
>> > Thanks to everyone who worked on this.
>> >
>> > On ward to 2.2.  If people have time please help with review / triage of
>> > what is critical for 2.2 and what needs to be pushed to 3.0.
>> >
>> > Tom
>> >
>> > _______________________________________________
>> > Matplotlib-devel mailing list
>> > Matplotlib-devel at python.org
>> > https://mail.python.org/mailman/listinfo/matplotlib-devel
>> >
>>
>>
>>
>> --
>> Sandro "morph" Tosi
>> My website: http://sandrotosi.me/
>> Me at Debian: http://wiki.debian.org/SandroTosi
>> G+: https://plus.google.com/u/0/+SandroTosi
>> _______________________________________________
>> Matplotlib-devel mailing list
>> Matplotlib-devel at python.org
>> https://mail.python.org/mailman/listinfo/matplotlib-devel
>>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/matplotlib-devel/attachments/20180121/8481c6b2/attachment.html>

From sandro.tosi at gmail.com  Sun Jan 21 13:10:44 2018
From: sandro.tosi at gmail.com (Sandro Tosi)
Date: Sun, 21 Jan 2018 13:10:44 -0500
Subject: [Matplotlib-devel] [REL] v2.1.2
In-Reply-To: <CAJP_izet_u7K9YbA9P6aN296WXaistReiMhPszrg_1uD2hoxMg@mail.gmail.com>
References: <CAA48SF8mBY+4NqooeOz0Gg5vXyk4PxPcSFJ3cjW12FTHpzRxqQ@mail.gmail.com>
 <CAB4XWXwkdWv3xXEqxmLhYt4LYY3z3hjJKXAcBa3+7uFPRbjsVQ@mail.gmail.com>
 <CAJP_izet_u7K9YbA9P6aN296WXaistReiMhPszrg_1uD2hoxMg@mail.gmail.com>
Message-ID: <CAB4XWXxazmajx3eLeK_WgpPzAhsnhJ=snTYQaLGJJc0ouzY=ew@mail.gmail.com>

thanks Elliott! we had a similar (but outdated) patch, now it works just fine!

looking at your fedora-patches branch, i noticed u to have several
change to adapt image_comparison threshold. we are in a similar
position: this is the result of the test suite while building the
package:

950 failed, 5955 passed, 26 skipped, 2 deselected, 5 xfailed, 1
xpassed, 5 warnings in 600.83 seconds
950 failed, 5940 passed, 28 skipped, 2 deselected, 7 xfailed, 1
xpassed, 5 warnings in 1543.66 seconds
951 failed, 5955 passed, 25 skipped, 2 deselected, 5 xfailed, 1
xpassed, 14 warnings in 574.67 seconds
951 failed, 5940 passed, 27 skipped, 2 deselected, 7 xfailed, 1
xpassed, 14 warnings in 838.53 seconds

(2.7, 2.7-debug, 3.6, 3.6-debug) and the count of the
ImageComparisonFailure failures is 3800 (which is basically hiding any
possible other relevant failure). sometimes RMS is as high as 111!

i'm not really sure what the right approach would be but it seems this
set of tests is currently not providing much useful information: from
a distribution POV what should we do? keep bumping the threshold until
they stop failing? (and i know we discussed this multiple times, sorry
to rehash it again :) )



On Thu, Jan 18, 2018 at 8:07 PM, Elliott Sales de Andrade
<quantum.analyst at gmail.com> wrote:
> Hi Sandro,
>
> I don't think Matplotlib searches in that location by default. Doesn't
> Debian also use the patch to modify the search path that we use on Fedora?
>
> Something like this one (rebased for 2.1.1):
> https://github.com/QuLogic/matplotlib/commit/58d84f95955556f7d089966e6cf7740818e9749e
>
> On Jan 18, 2018 7:27 PM, "Sandro Tosi" <sandro.tosi at gmail.com> wrote:
>>
>> hello everyone,
>> while preparing the package for Debian i noticed that somewhere
>> between 2.0.0 and 2.1.1 mpl no longer read /etc/matplotlibrc (we
>> install that file to give a basic settings to every user without
>> requiring them to create a custom .config/matplotlibrc), or doesnt
>> understand it correctly (but i'm not sure about this last point, as i
>> copied matplotlibrc.template replacing the backend value and still
>> doesnt work).
>>
>> with 2.0.0, running
>>
>> python
>> /usr/share/doc/python-matplotlib-doc//examples/shapes_and_collections/line_collection.py
>>
>> shows the 2 images just fine, but with 2.1.1 installed:
>>
>> python
>> /usr/share/doc/python-matplotlib-doc//examples/shapes_and_collections/line_collection.py
>> /usr/lib/python2.7/dist-packages/matplotlib/__init__.py:941:
>> UserWarning: could not find rc file; returning defaults
>>  warnings.warn(message)
>>
>> and of course no image is shown. backend is TkAgg in both cases.
>>
>> do you know what's going on?
>>
>> thanks in advance!
>>
>> On Thu, Jan 18, 2018 at 12:06 AM, Thomas Caswell <tcaswell at gmail.com>
>> wrote:
>> > Folks,
>> >
>> > Just tagged and pushed v2.1.2.
>> >
>> > Kicked off the mac/linux wheel building system, emailed Christoph, and
>> > put a
>> > PR into conda-forge for the new binaries.
>> >
>> > Will upload to pypi, updated the docs, and announce widely tomorrow
>> > night.
>> >
>> > Thanks to everyone who worked on this.
>> >
>> > On ward to 2.2.  If people have time please help with review / triage of
>> > what is critical for 2.2 and what needs to be pushed to 3.0.
>> >
>> > Tom
>> >
>> > _______________________________________________
>> > Matplotlib-devel mailing list
>> > Matplotlib-devel at python.org
>> > https://mail.python.org/mailman/listinfo/matplotlib-devel
>> >
>>
>>
>>
>> --
>> Sandro "morph" Tosi
>> My website: http://sandrotosi.me/
>> Me at Debian: http://wiki.debian.org/SandroTosi
>> G+: https://plus.google.com/u/0/+SandroTosi
>> _______________________________________________
>> Matplotlib-devel mailing list
>> Matplotlib-devel at python.org
>> https://mail.python.org/mailman/listinfo/matplotlib-devel



-- 
Sandro "morph" Tosi
My website: http://sandrotosi.me/
Me at Debian: http://wiki.debian.org/SandroTosi
G+: https://plus.google.com/u/0/+SandroTosi

From rmay31 at gmail.com  Mon Jan 22 17:24:22 2018
From: rmay31 at gmail.com (Ryan May)
Date: Mon, 22 Jan 2018 15:24:22 -0700
Subject: [Matplotlib-devel] 2018-01-22 Call notes
Message-ID: <CAKH0P+XnRvwjtY2zvs0ioeaGg+VT8y706KouPjxGuA=YTynQnA@mail.gmail.com>

On call: Myself, Anthony Lee, Eric Firing, Hannah, Jody Klymak, Ryan May

- 2.2 priorities - 827 open issues tagged for this release, but these are
mostly just carried forward from past milestones
- Looked over release critical issues
- 10033 (Axes sharing improvements) looks just about ready, needs some
reviews
- Punting 10216 (GDB gone on Travis) for now -- Anthony submitted 10291 for
setting fault handler
- Discussed 10210 (Qt Cairo): Who else is well-versed enough in
Qt/Matplotlib to give a review
- Need to get 9082 (geometry constraint manager) in and start kicking the
tires on it
- Discussed 10206 (units/converters and axes): decided it?s fine to not
convert floats (numbers) passed in on axes that have converters
- 8820 (pcolor of arrays of bools broken by numpy 1.13): Decided it?s
numpy?s fault (changing behavior), not ours, so we're not going to go out
of our way to fix this in matplotlib itself. There?s a reasonable
workaround with `arr.astype(np.int)`.

Ryan

-- 
Ryan May
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/matplotlib-devel/attachments/20180122/7a18e30e/attachment.html>

From ben.v.root at gmail.com  Tue Jan 23 20:42:26 2018
From: ben.v.root at gmail.com (Benjamin Root)
Date: Tue, 23 Jan 2018 20:42:26 -0500
Subject: [Matplotlib-devel] SciPy 2018 Tutorial Proposals Due Feb. 9th
Message-ID: <CANNq6FmM6hjDFEMqQndZxszPx=dr-JYihUUeSYjribZAUs0GJQ@mail.gmail.com>

Dear matplotlib developers,

The SciPy 2018 <https://scipy2018.scipy.org/ehome/index.php?eventid=299527&>
call
for proposals is open and I am emailing on behalf of the Tutorial Committee
to encourage a submission from you all. Tutorials will be presented July
9-10 and, as usual, the conference is in Austin. You may find additional
information here  <https://scipy2018.scipy.org/ehome/299527/648139/>and the
deadline for submission is February 9th. (Please note that this is
significantly earlier than past years.)

Please feel free to email me with any questions.

Sincerely,
Ben Root
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/matplotlib-devel/attachments/20180123/6f7f6d35/attachment.html>

From Nicolas.Rougier at inria.fr  Wed Jan 31 10:18:22 2018
From: Nicolas.Rougier at inria.fr (Nicolas Rougier)
Date: Wed, 31 Jan 2018 16:18:22 +0100
Subject: [Matplotlib-devel] Article about matplotlib
Message-ID: <A0AF2E4D-4597-4DC2-ACCC-52C71C66811C@inria.fr>


Dear all,

Ralf Gommers proposed a few days ago to start the writing of a long article about scipy and I wonder if there is some plan for matplotlib as well? I?m always citing John?s article but it would be nice to have a more up to date article (John?s article is 11 years old and matplotlib has evolved a lot since then). What do you think?

Nicolas

From jklymak at uvic.ca  Wed Jan 31 13:41:32 2018
From: jklymak at uvic.ca (Jody Klymak)
Date: Wed, 31 Jan 2018 10:41:32 -0800
Subject: [Matplotlib-devel] Call notes 29 Jan
Message-ID: <9159D15C-00C2-4857-BE57-CD71BE149966@uvic.ca>

Hi all,

Call notes from Monday?s call.

Cheers,   Jody


On call: Thomas, Eric, Ryan, Jody, Antony (later half)

## Documentation

Thomas reported that there is a fund for a mini-workshop to scope out documentation changes.  Plan was to use that to write a proposal to fund a book.
  - Eric expressed some reservations about a book, given the state of flux some parts of Matplotlib are in.  
  - Jody noted that beyond the book, a shorter ?getting started? guide on the webpage would still be desirable:
	
	ACTION: Jody to co-ordinate a couple of outline documents that could start outlining both of these projects before any meeting.  

There was also some discussion of online documentation technology:
  - Notebooks (via nbsphinx: https://nbsphinx.readthedocs.io)
	- hard to make PRs with.  Diffs are a mess in GitHub.  
      - Can?t use emacs (easily)
  - RST-based, with plot directive: 
	- python authoring less than desirable
  - Sphinx-gallery based: 
	- rst authoring less than desirable 
	- sphinx gallery has very restricted ordering methods (file size and alphabetical)

## Release 2.2

Various Release-Critical PRs were looked at.


--
Jody Klymak    
http://web.uvic.ca/~jklymak/






From anntzer.lee at gmail.com  Wed Jan 31 15:29:52 2018
From: anntzer.lee at gmail.com (Antony Lee)
Date: Wed, 31 Jan 2018 12:29:52 -0800
Subject: [Matplotlib-devel] Article about matplotlib
In-Reply-To: <A0AF2E4D-4597-4DC2-ACCC-52C71C66811C@inria.fr>
References: <A0AF2E4D-4597-4DC2-ACCC-52C71C66811C@inria.fr>
Message-ID: <CAGRr6BFQ7hnWkery+k7cKdg1M2oZv_-oqR7bYebGhiu1y3E49g@mail.gmail.com>

I don't think there is a plan, but I like the idea...

On Jan 31, 2018 7:18 AM, "Nicolas Rougier" <Nicolas.Rougier at inria.fr> wrote:

>
> Dear all,
>
> Ralf Gommers proposed a few days ago to start the writing of a long
> article about scipy and I wonder if there is some plan for matplotlib as
> well? I?m always citing John?s article but it would be nice to have a more
> up to date article (John?s article is 11 years old and matplotlib has
> evolved a lot since then). What do you think?
>
> Nicolas
> _______________________________________________
> Matplotlib-devel mailing list
> Matplotlib-devel at python.org
> https://mail.python.org/mailman/listinfo/matplotlib-devel
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/matplotlib-devel/attachments/20180131/415f1a44/attachment.html>

From jklymak at uvic.ca  Wed Jan 31 16:10:58 2018
From: jklymak at uvic.ca (Jody Klymak)
Date: Wed, 31 Jan 2018 13:10:58 -0800
Subject: [Matplotlib-devel] Call notes 29 Jan
In-Reply-To: <9159D15C-00C2-4857-BE57-CD71BE149966@uvic.ca>
References: <9159D15C-00C2-4857-BE57-CD71BE149966@uvic.ca>
Message-ID: <4EB5E55B-9FCD-41A1-9776-C3909472B583@uvic.ca>

More call notes, this time from a more definitive source @tacaswell!

- discussion about how to manage
- discussion about docs
  - TAC: we have money to organize a mini-conference (from active state)
  - EF: are we settled enough to do this? Do we want to leave it more dynamic?
    - discoverability is mostly just google
      - our API is overly complicated and organically grown
      - major barrier to discoverability / usability
    - maybe we want to clean up the API (aggressively?) first
    - leads to questions about to define the agenda of mini-workshop
  - JK: need an intermediate between what we have and a 200 page book
    - need to order the sphinx-gallery entries
  - EF: consider reverting the tutorials as sphinx-gallery
    - need more control of ordering, but works well for examples
    - wrong tool for tutorials
    - look into sphinx-notebook
    - move back to plain sphinx?
  - TAC: we need to develop agenda for mini-workshop that addresses the above
  - EF: before workshop is there a way to gather user requirements?
    - TAC: user interviews, plausible, may also be part of work, can start now.
    - JK: we can start collecting comments ahead of time
      - JK will work on managing this!

> On 31 Jan 2018, at 10:41, Jody Klymak <jklymak at uvic.ca> wrote:
> 
> Hi all,
> 
> Call notes from Monday?s call.
> 
> Cheers,   Jody
> 
> 
> On call: Thomas, Eric, Ryan, Jody, Antony (later half)
> 
> ## Documentation
> 
> Thomas reported that there is a fund for a mini-workshop to scope out documentation changes.  Plan was to use that to write a proposal to fund a book.
>  - Eric expressed some reservations about a book, given the state of flux some parts of Matplotlib are in.  
>  - Jody noted that beyond the book, a shorter ?getting started? guide on the webpage would still be desirable:
> 	
> 	ACTION: Jody to co-ordinate a couple of outline documents that could start outlining both of these projects before any meeting.  
> 
> There was also some discussion of online documentation technology:
>  - Notebooks (via nbsphinx: https://nbsphinx.readthedocs.io)
> 	- hard to make PRs with.  Diffs are a mess in GitHub.  
>      - Can?t use emacs (easily)
>  - RST-based, with plot directive: 
> 	- python authoring less than desirable
>  - Sphinx-gallery based: 
> 	- rst authoring less than desirable 
> 	- sphinx gallery has very restricted ordering methods (file size and alphabetical)
> 
> ## Release 2.2
> 
> Various Release-Critical PRs were looked at.
> 
> 
> --
> Jody Klymak    
> http://web.uvic.ca/~jklymak/
> 
> 
> 
> 
> 
> _______________________________________________
> Matplotlib-devel mailing list
> Matplotlib-devel at python.org
> https://mail.python.org/mailman/listinfo/matplotlib-devel

--
Jody Klymak    
http://web.uvic.ca/~jklymak/






From nelle.varoquaux at gmail.com  Wed Jan 31 16:54:04 2018
From: nelle.varoquaux at gmail.com (Nelle Varoquaux)
Date: Wed, 31 Jan 2018 22:54:04 +0100
Subject: [Matplotlib-devel] Call notes 29 Jan
In-Reply-To: <9159D15C-00C2-4857-BE57-CD71BE149966@uvic.ca>
References: <9159D15C-00C2-4857-BE57-CD71BE149966@uvic.ca>
Message-ID: <CAE-UAvRJ7pYYR5M9O54f9E-VBxBZpw4wFoFj45P4pRn1GNC+AQ@mail.gmail.com>

>   - Jody noted that beyond the book, a shorter ?getting started? guide on the webpage would still be desirable:

There was an attempt at integrating Nicolas Rougier's excellent
matplotlib tutorial to the doc. I believe he had given permission to
do so, and that the amount of work required to merge the tutorial was
extremely small.

Cheers,
N

>
>         ACTION: Jody to co-ordinate a couple of outline documents that could start outlining both of these projects before any meeting.
>
> There was also some discussion of online documentation technology:
>   - Notebooks (via nbsphinx: https://nbsphinx.readthedocs.io)
>         - hard to make PRs with.  Diffs are a mess in GitHub.
>       - Can?t use emacs (easily)
>   - RST-based, with plot directive:
>         - python authoring less than desirable
>   - Sphinx-gallery based:
>         - rst authoring less than desirable
>         - sphinx gallery has very restricted ordering methods (file size and alphabetical)
>
> ## Release 2.2
>
> Various Release-Critical PRs were looked at.
>
>
> --
> Jody Klymak
> http://web.uvic.ca/~jklymak/
>
>
>
>
>
> _______________________________________________
> Matplotlib-devel mailing list
> Matplotlib-devel at python.org
> https://mail.python.org/mailman/listinfo/matplotlib-devel

From antony.lee at berkeley.edu  Wed Jan 31 18:07:09 2018
From: antony.lee at berkeley.edu (Antony Lee)
Date: Wed, 31 Jan 2018 15:07:09 -0800
Subject: [Matplotlib-devel] Call notes 29 Jan
In-Reply-To: <4EB5E55B-9FCD-41A1-9776-C3909472B583@uvic.ca>
References: <9159D15C-00C2-4857-BE57-CD71BE149966@uvic.ca>
 <4EB5E55B-9FCD-41A1-9776-C3909472B583@uvic.ca>
Message-ID: <CAGRr6BGdjzzF_6JFA8hrMc7Gf5i0_jw2XVYMos0OjrrDrwffQQ@mail.gmail.com>

I missed the discussion about the docs, so
1. s-g is great for standalone examples but I would also prefer plain rst
for long tutorials (with e.g. vim-restructuredtext I can have both syntax
highlighting of rst parts in rst and python code-blocks in python (and
likewise for other languages) if the file is rst; for a .py file there is
no special recognition of rst markup by the editor).
2. any specific points about "python authoring less than desirable" for
rst?  what functionalities would be desirable?
  possible relevant links (some already mentioned above):
https://nbsphinx.readthedocs.io/en/0.3.1/
https://github.com/ngoldbaum/RunNotebook
https://github.com/matthew-brett/nb2plots

Antony

2018-01-31 13:10 GMT-08:00 Jody Klymak <jklymak at uvic.ca>:

> More call notes, this time from a more definitive source @tacaswell!
>
> - discussion about how to manage
> - discussion about docs
>   - TAC: we have money to organize a mini-conference (from active state)
>   - EF: are we settled enough to do this? Do we want to leave it more
> dynamic?
>     - discoverability is mostly just google
>       - our API is overly complicated and organically grown
>       - major barrier to discoverability / usability
>     - maybe we want to clean up the API (aggressively?) first
>     - leads to questions about to define the agenda of mini-workshop
>   - JK: need an intermediate between what we have and a 200 page book
>     - need to order the sphinx-gallery entries
>   - EF: consider reverting the tutorials as sphinx-gallery
>     - need more control of ordering, but works well for examples
>     - wrong tool for tutorials
>     - look into sphinx-notebook
>     - move back to plain sphinx?
>   - TAC: we need to develop agenda for mini-workshop that addresses the
> above
>   - EF: before workshop is there a way to gather user requirements?
>     - TAC: user interviews, plausible, may also be part of work, can start
> now.
>     - JK: we can start collecting comments ahead of time
>       - JK will work on managing this!
>
> > On 31 Jan 2018, at 10:41, Jody Klymak <jklymak at uvic.ca> wrote:
> >
> > Hi all,
> >
> > Call notes from Monday?s call.
> >
> > Cheers,   Jody
> >
> >
> > On call: Thomas, Eric, Ryan, Jody, Antony (later half)
> >
> > ## Documentation
> >
> > Thomas reported that there is a fund for a mini-workshop to scope out
> documentation changes.  Plan was to use that to write a proposal to fund a
> book.
> >  - Eric expressed some reservations about a book, given the state of
> flux some parts of Matplotlib are in.
> >  - Jody noted that beyond the book, a shorter ?getting started? guide on
> the webpage would still be desirable:
> >
> >       ACTION: Jody to co-ordinate a couple of outline documents that
> could start outlining both of these projects before any meeting.
> >
> > There was also some discussion of online documentation technology:
> >  - Notebooks (via nbsphinx: https://nbsphinx.readthedocs.io)
> >       - hard to make PRs with.  Diffs are a mess in GitHub.
> >      - Can?t use emacs (easily)
> >  - RST-based, with plot directive:
> >       - python authoring less than desirable
> >  - Sphinx-gallery based:
> >       - rst authoring less than desirable
> >       - sphinx gallery has very restricted ordering methods (file size
> and alphabetical)
> >
> > ## Release 2.2
> >
> > Various Release-Critical PRs were looked at.
> >
> >
> > --
> > Jody Klymak
> > http://web.uvic.ca/~jklymak/
> >
> >
> >
> >
> >
> > _______________________________________________
> > Matplotlib-devel mailing list
> > Matplotlib-devel at python.org
> > https://mail.python.org/mailman/listinfo/matplotlib-devel
>
> --
> Jody Klymak
> http://web.uvic.ca/~jklymak/
>
>
>
>
>
> _______________________________________________
> Matplotlib-devel mailing list
> Matplotlib-devel at python.org
> https://mail.python.org/mailman/listinfo/matplotlib-devel
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/matplotlib-devel/attachments/20180131/707162af/attachment-0001.html>

From jklymak at uvic.ca  Wed Jan 31 18:15:47 2018
From: jklymak at uvic.ca (Jody Klymak)
Date: Wed, 31 Jan 2018 15:15:47 -0800
Subject: [Matplotlib-devel] Call notes 29 Jan
In-Reply-To: <CAGRr6BGdjzzF_6JFA8hrMc7Gf5i0_jw2XVYMos0OjrrDrwffQQ@mail.gmail.com>
References: <9159D15C-00C2-4857-BE57-CD71BE149966@uvic.ca>
 <4EB5E55B-9FCD-41A1-9776-C3909472B583@uvic.ca>
 <CAGRr6BGdjzzF_6JFA8hrMc7Gf5i0_jw2XVYMos0OjrrDrwffQQ@mail.gmail.com>
Message-ID: <EBD53ECC-EF41-453E-A30B-0E91023FE06F@uvic.ca>



> On 31 Jan 2018, at 15:07, Antony Lee <antony.lee at berkeley.edu> wrote:
> 
> 2. any specific points about "python authoring less than desirable" for rst?  what functionalities would be desirable?

I think all that was meant was that you can?t easily syntax highlight python in rst in most editors (not to start a vi/emacs thread ;-))

Personal opinion:  I can?t write/compile an rst snippet in a reasonable timeframe:  I?m not smart enough to write anything correctly the first time, so I prefer to hack.  My computer is pretty good, but it takes 12 minutes to compile the docs, even if I make one small change.  If there were a way to compile one file at a time, or if the doc build were smart enough to keep track of what has changed and what hasn?t it?d make editing in rst a lot easier.  

Cheers,  Jody

From antony.lee at berkeley.edu  Wed Jan 31 18:33:57 2018
From: antony.lee at berkeley.edu (Antony Lee)
Date: Wed, 31 Jan 2018 15:33:57 -0800
Subject: [Matplotlib-devel] Call notes 29 Jan
In-Reply-To: <EBD53ECC-EF41-453E-A30B-0E91023FE06F@uvic.ca>
References: <9159D15C-00C2-4857-BE57-CD71BE149966@uvic.ca>
 <4EB5E55B-9FCD-41A1-9776-C3909472B583@uvic.ca>
 <CAGRr6BGdjzzF_6JFA8hrMc7Gf5i0_jw2XVYMos0OjrrDrwffQQ@mail.gmail.com>
 <EBD53ECC-EF41-453E-A30B-0E91023FE06F@uvic.ca>
Message-ID: <CAGRr6BEvjMadZCKZ6wD3V2DmQ2np7qXHa-aX_wffsiCN3-i6Ew@mail.gmail.com>

Screenshot just for the record: left is "out-of-the-box" vim with no
customizations (vim -u NONE + :syntax on; which already has syntax
highlighting out of the box for python blocks); right is my setup (using
vim-restructuredtext HEAD).
I agree that the very slow edit-compile cycle for sphinx is less than
optimal.  If you are just working on the rst part of a single file, you can
sort-of get away with rst2html (ships with docutils, bound to `:make` on my
vim setup), which will spew out tons of errors because it doesn't know any
sphinx constructs, but will still let you check the basic markup part.  You
can also check docstrings that way by just processing the docstring in a
separate file with rst2html.
Antony

2018-01-31 15:15 GMT-08:00 Jody Klymak <jklymak at uvic.ca>:

>
>
> > On 31 Jan 2018, at 15:07, Antony Lee <antony.lee at berkeley.edu> wrote:
> >
> > 2. any specific points about "python authoring less than desirable" for
> rst?  what functionalities would be desirable?
>
> I think all that was meant was that you can?t easily syntax highlight
> python in rst in most editors (not to start a vi/emacs thread ;-))
>
> Personal opinion:  I can?t write/compile an rst snippet in a reasonable
> timeframe:  I?m not smart enough to write anything correctly the first
> time, so I prefer to hack.  My computer is pretty good, but it takes 12
> minutes to compile the docs, even if I make one small change.  If there
> were a way to compile one file at a time, or if the doc build were smart
> enough to keep track of what has changed and what hasn?t it?d make editing
> in rst a lot easier.
>
> Cheers,  Jody
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/matplotlib-devel/attachments/20180131/48bc1c0f/attachment-0001.html>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: Screenshot_20180131_152655.png
Type: image/png
Size: 590193 bytes
Desc: not available
URL: <http://mail.python.org/pipermail/matplotlib-devel/attachments/20180131/48bc1c0f/attachment-0001.png>

From Nicolas.Rougier at inria.fr  Wed Jan 31 22:18:37 2018
From: Nicolas.Rougier at inria.fr (Nicolas Rougier)
Date: Thu, 1 Feb 2018 04:18:37 +0100
Subject: [Matplotlib-devel] Call notes 29 Jan
In-Reply-To: <CAE-UAvRJ7pYYR5M9O54f9E-VBxBZpw4wFoFj45P4pRn1GNC+AQ@mail.gmail.com>
References: <9159D15C-00C2-4857-BE57-CD71BE149966@uvic.ca>
 <CAE-UAvRJ7pYYR5M9O54f9E-VBxBZpw4wFoFj45P4pRn1GNC+AQ@mail.gmail.com>
Message-ID: <03E005EC-7E4D-419E-9B0B-D75CD2F1574E@inria.fr>



> On 31 Jan 2018, at 22:54, Nelle Varoquaux <nelle.varoquaux at gmail.com> wrote:
> 
>>  - Jody noted that beyond the book, a shorter ?getting started? guide on the webpage would still be desirable:
> 
> There was an attempt at integrating Nicolas Rougier's excellent
> matplotlib tutorial to the doc. I believe he had given permission to
> do so, and that the amount of work required to merge the tutorial was
> extremely small.

Yes, and I can help with the integration as well (the tutorial needs an update anyway).

Nicolas


> 
> Cheers,
> N
> 
>> 
>>        ACTION: Jody to co-ordinate a couple of outline documents that could start outlining both of these projects before any meeting.
>> 
>> There was also some discussion of online documentation technology:
>>  - Notebooks (via nbsphinx: https://nbsphinx.readthedocs.io)
>>        - hard to make PRs with.  Diffs are a mess in GitHub.
>>      - Can?t use emacs (easily)
>>  - RST-based, with plot directive:
>>        - python authoring less than desirable
>>  - Sphinx-gallery based:
>>        - rst authoring less than desirable
>>        - sphinx gallery has very restricted ordering methods (file size and alphabetical)
>> 
>> ## Release 2.2
>> 
>> Various Release-Critical PRs were looked at.
>> 
>> 
>> --
>> Jody Klymak
>> http://web.uvic.ca/~jklymak/
>> 
>> 
>> 
>> 
>> 
>> _______________________________________________
>> Matplotlib-devel mailing list
>> Matplotlib-devel at python.org
>> https://mail.python.org/mailman/listinfo/matplotlib-devel
> _______________________________________________
> Matplotlib-devel mailing list
> Matplotlib-devel at python.org
> https://mail.python.org/mailman/listinfo/matplotlib-devel