Plans for matplotlib py3k/final release (and impact on Debian)

Development matplotlib-devel
1

Hello,
as you may be aware of, in 1 month (more or less) Debian will freeze,
that means that no new upstream releases will be allowed in the
upcoming release, only fixex for important bugs.

Currently in Debian archive we have 1.1.1rc1 . Recent mails have
mentioned that the next release will be py3k enabled. Well, what are
you're plans to release such version? :slight_smile:

It would be really awesome to have a python3 matplotlib in Debian, and
i'd be happy to test any new RC you'd like to release.

Cheers,

···

--
Sandro Tosi (aka morph, morpheus, matrixhasu)
My website: http://matrixhasu.altervista.org/
Me at Debian: http://wiki.debian.org/SandroTosi

2

Just to mention: I've set up a daily builds PPA for matplotlib, and
it's been happily producing Python 3 builds for a while, so it looks
like it should be fairly painless:

Thanks,
Thomas

···

On 12 May 2012 10:56, Sandro Tosi <morph@...12...> wrote:

It would be really awesome to have a python3 matplotlib in Debian, and
i'd be happy to test any new RC you'd like to release.

3

My original offer (made several months ago) to help test this on Windows still stands :slight_smile:

···

On 12/05/2012 11:01, Thomas Kluyver wrote:

On 12 May 2012 10:56, Sandro Tosi<morph@...12...> wrote:

It would be really awesome to have a python3 matplotlib in Debian, and
i'd be happy to test any new RC you'd like to release.

Just to mention: I've set up a daily builds PPA for matplotlib, and
it's been happily producing Python 3 builds for a while, so it looks
like it should be fairly painless:
Matplotlib daily : Thomas Kluyver

Thanks,
Thomas

------------------------------------------------------------------------------
Live Security Virtual Conference
Exclusive live event will cover all the ways today's security and
threat landscape has changed and how IT managers can respond. Discussions
will include endpoint security, mobile security and the latest in malware
threats. http://www.accelacomm.com/jaw/sfrnl04242012/114/50122263/

--
Cheers.

Mark Lawrence.

4

Thanks. Does git master build and pass the unit tests on Windows?

Mike

···

On 05/12/2012 07:21 AM, Mark Lawrence wrote:

My original offer (made several months ago) to help test this on Windows
still stands :slight_smile:

5

git master builds and tests OK on win32-py2.7.

With the attached patch git master builds and works (in practice) OK on win-amd64-py3.2 but there are many test errors of type "RuntimeError: Could not open facefile X:\Python32\...\ttf\Vera.ttf; Cannot_Open_Resource". I do delete the ~\.matplotlib folder before running the tests and can verify that FT2Font(r"X:\Python32\lib\site-packages\matplotlib\mpl-data\fonts\ttf\Vera.ttf") works.

Christoph

mpl-master.diff (1.32 KB)

···

On 5/12/2012 6:16 AM, Michael Droettboom wrote:

On 05/12/2012 07:21 AM, Mark Lawrence wrote:

My original offer (made several months ago) to help test this on Windows
still stands :slight_smile:

Thanks. Does git master build and pass the unit tests on Windows?

Mike

6

That looks like the same issue we were having on the 1.1.x branch –
that it’s running out of file handles – that I thought was fixed by
this PR:

https://github.com/matplotlib/matplotlib/pull/798

Or are we seeing something else here?

Mike
···

On 05/12/2012 01:33 PM, Christoph Gohlke wrote:

  On 5/12/2012 6:16 AM, Michael Droettboom wrote:

    On 05/12/2012 07:21 AM, Mark Lawrence > > wrote:

      My original offer (made several months ago) to help test this

on Windows

      still stands :)

    Thanks.  Does git master build and pass the unit tests on

Windows?

    Mike

  git master builds and tests OK on win32-py2.7.




  With the attached patch git master builds and works (in practice)

OK on win-amd64-py3.2 but there are many test errors of type
“RuntimeError: Could not open facefile
X:\Python32.…\ttf\Vera.ttf; Cannot_Open_Resource”. I do delete
the ~.matplotlib folder before running the tests and can verify
that
FT2Font(r"X:\Python32\lib\site-packages\matplotlib\mpl-data\fonts\ttf\Vera.ttf")
works.

7

Hello,
as you may be aware of, in 1 month (more or less) Debian will freeze,
that means that no new upstream releases will be allowed in the
upcoming release, only fixex for important bugs.

Currently in Debian archive we have 1.1.1rc1 . Recent mails have
mentioned that the next release will be py3k enabled. Well, what are
you're plans to release such version? :slight_smile:

I don't think we have a timeframe on that yet. Getting out a solid, stable 1.1.1 (without Py3k support) I think is a higher priority right now.

It would be really awesome to have a python3 matplotlib in Debian, and
i'd be happy to test any new RC you'd like to release.

To get this going, perhaps I'll go ahead and make a branch off of master for 1.2.x and we could start testing it on all the platforms and see how far it feels from releasable state. There are few PRs in the queue that are simple yet critical that should get in there first.

Mike

···

On 05/12/2012 05:56 AM, Sandro Tosi wrote:

8

PR798 works for Python 2.7. But under Python 3.2 the tests are still running out of file handles. Manually increasing the open files limit helps (only one test fails).

Christoph

···

On 5/14/2012 7:43 AM, Michael Droettboom wrote:

On 05/12/2012 01:33 PM, Christoph Gohlke wrote:

On 5/12/2012 6:16 AM, Michael Droettboom wrote:

On 05/12/2012 07:21 AM, Mark Lawrence wrote:

My original offer (made several months ago) to help test this on
Windows
still stands :slight_smile:

Thanks. Does git master build and pass the unit tests on Windows?

Mike

git master builds and tests OK on win32-py2.7.

With the attached patch git master builds and works (in practice) OK
on win-amd64-py3.2 but there are many test errors of type
"RuntimeError: Could not open facefile X:\Python32\...\ttf\Vera.ttf;
Cannot_Open_Resource". I do delete the ~\.matplotlib folder before
running the tests and can verify that
FT2Font(r"X:\Python32\lib\site-packages\matplotlib\mpl-data\fonts\ttf\Vera.ttf")
works.

That looks like the same issue we were having on the 1.1.x branch --
that it's running out of file handles -- that I thought was fixed by
this PR:

use class level rather than instance level caching of fontd by jdh2358 · Pull Request #798 · matplotlib/matplotlib · GitHub

Or are we seeing something else here?

Mike

9

Ah, I guess that makes sense given how destructors are handled differently on Python 3. I have some thoughts on this that I might put into a PR. I don't have access to a Windows box at the moment, so I may need some help testing.

Mike

···

On 05/14/2012 12:58 PM, Christoph Gohlke wrote:

On 5/14/2012 7:43 AM, Michael Droettboom wrote:

On 05/12/2012 01:33 PM, Christoph Gohlke wrote:

On 5/12/2012 6:16 AM, Michael Droettboom wrote:

On 05/12/2012 07:21 AM, Mark Lawrence wrote:

My original offer (made several months ago) to help test this on
Windows
still stands :slight_smile:

Thanks. Does git master build and pass the unit tests on Windows?

Mike

git master builds and tests OK on win32-py2.7.

With the attached patch git master builds and works (in practice) OK
on win-amd64-py3.2 but there are many test errors of type
"RuntimeError: Could not open facefile X:\Python32\...\ttf\Vera.ttf;
Cannot_Open_Resource". I do delete the ~\.matplotlib folder before
running the tests and can verify that
FT2Font(r"X:\Python32\lib\site-packages\matplotlib\mpl-data\fonts\ttf\Vera.ttf")
works.

That looks like the same issue we were having on the 1.1.x branch --
that it's running out of file handles -- that I thought was fixed by
this PR:

use class level rather than instance level caching of fontd by jdh2358 · Pull Request #798 · matplotlib/matplotlib · GitHub

PR798 works for Python 2.7. But under Python 3.2 the tests are still
running out of file handles. Manually increasing the open files limit
helps (only one test fails).

10

Hello,

···

On Sat, May 12, 2012 at 11:56 AM, Sandro Tosi <morph@...12...> wrote:

Hello,
as you may be aware of, in 1 month (more or less) Debian will freeze,
that means that no new upstream releases will be allowed in the
upcoming release, only fixex for important bugs.

Sorry to keep revamping this, but Debian freeze will be in about 2
weeks: do you have in mind a date for a matplotlib release?

Cheers.
--
Sandro Tosi (aka morph, morpheus, matrixhasu)
My website: http://matrixhasu.altervista.org/
Me at Debian: http://wiki.debian.org/SandroTosi

11

I did not see any replies to Sandro's reasonable question.

Wasn't there at least a plan for a 1.1.x release soon?

There are quite a few open pull requests for master--not many for 1.1.x--but I don't know that any are blockers.

Eric

···

On 06/02/2012 10:38 AM, Sandro Tosi wrote:

Hello,

On Sat, May 12, 2012 at 11:56 AM, Sandro Tosi<morph@...12...> wrote:

Hello,
as you may be aware of, in 1 month (more or less) Debian will freeze,
that means that no new upstream releases will be allowed in the
upcoming release, only fixex for important bugs.

Sorry to keep revamping this, but Debian freeze will be in about 2
weeks: do you have in mind a date for a matplotlib release?

Cheers.

12

There are no open issues tagged "release_candidate" so I'll go ahead
and make the 1.1.x release if there are no objections.

JDH

···

On Thu, Jun 7, 2012 at 1:55 PM, Eric Firing <efiring@...229...> wrote:

I did not see any replies to Sandro's reasonable question.

Wasn't there at least a plan for a 1.1.x release soon?

There are quite a few open pull requests for master--not many for
1.1.x--but I don't know that any are blockers.

13

Is there time for another release candidate?

With the latest github branch on win-amd64-py2.7 I get both, a segfault and a test failure in "matplotlib.tests.test_text.test_font_styles.test".

Christoph

···

On 6/7/2012 12:04 PM, John Hunter wrote:

On Thu, Jun 7, 2012 at 1:55 PM, Eric Firing <efiring@...229...> wrote:

I did not see any replies to Sandro's reasonable question.

Wasn't there at least a plan for a 1.1.x release soon?

There are quite a few open pull requests for master--not many for
1.1.x--but I don't know that any are blockers.

There are no open issues tagged "release_candidate" so I'll go ahead
and make the 1.1.x release if there are no objections.

JDH

14

If you are getting a segfault, is there on the latest branch, would it
make sense to do an RC? It seems like we would want to fix the
segfault first. Can you give us more details on the segfault, eg
backend, conditions to reproduce, etc?

···

On Thu, Jun 7, 2012 at 2:57 PM, Christoph Gohlke <cgohlke@...244...> wrote:

Is there time for another release candidate?

With the latest github branch on win-amd64-py2.7 I get both, a segfault
and a test failure in "matplotlib.tests.test_text.test_font_styles.test".

15

Is there time for another release candidate?

With the latest github branch on win-amd64-py2.7 I get both, a segfault
and a test failure in "matplotlib.tests.test_text.test_font_styles.test".

If you are getting a segfault, is there on the latest branch, would it
make sense to do an RC? It seems like we would want to fix the
segfault first. Can you give us more details on the segfault, eg
backend, conditions to reproduce, etc?

There have been many changes since the last rc. The crash is probably related to very recent changes. Ideally the segfault should be fixed before rc2.

The crash on win-amd64-py2.7 is with the PDF backend:

Python 2.7.3 (default, Apr 10 2012, 23:24:47) [MSC v.1500 64 bit (AMD64)] on win32

import matplotlib as mpl
mpl.use("agg")
mpl.test(verbosity=2)

<snip>
matplotlib.tests.test_mathtext.mathtext_stixsans_65_test.test ... ok
matplotlib.tests.test_mathtext.mathtext_stixsans_65_test.test ... KNOWNFAIL: Cannot compare svg files on this system
matplotlib.tests.test_mathtext.test_fontinfo ... ok
matplotlib.tests.test_text.test_antialiasing.test ... ok
matplotlib.tests.test_text.test_font_styles.test ... ok
matplotlib.tests.test_text.test_font_styles.test ...
crash

On win32-py2.7 the test fails but doesn't crash. The failure is that bold font styles are not rendered bold:

Python 2.7.3 (default, Apr 10 2012, 23:31:26) [MSC v.1500 32 bit (Intel)] on win32
<snip>

···

On 6/7/2012 1:01 PM, John Hunter wrote:

On Thu, Jun 7, 2012 at 2:57 PM, Christoph Gohlke <cgohlke@...244...> wrote:

======================================================================
FAIL: matplotlib.tests.test_text.test_font_styles.test
----------------------------------------------------------------------
Traceback (most recent call last):
   File "X:\Python27\lib\site-packages\nose\case.py", line 197, in runTest
     self.test(*self.arg)
   File "X:\Python27\lib\site-packages\matplotlib\testing\decorators.py", line 36, in failer
     result = f(*args, **kwargs)
   File "X:\Python27\lib\site-packages\matplotlib\testing\decorators.py", line 140, in do_test
     '(RMS %(rms).3f)'%err)
ImageComparisonFailure: images not close: test\result_images\test_text\font_styles.png vs.
  test\result_images\test_text\expected-font_styles.png (RMS 47.138)

======================================================================
FAIL: matplotlib.tests.test_text.test_font_styles.test
----------------------------------------------------------------------
Traceback (most recent call last):
   File "X:\Python27\lib\site-packages\nose\case.py", line 197, in runTest
     self.test(*self.arg)
   File "X:\Python27\lib\site-packages\matplotlib\testing\decorators.py", line 36, in failer
     result = f(*args, **kwargs)
   File "X:\Python27\lib\site-packages\matplotlib\testing\decorators.py", line 140, in do_test
     '(RMS %(rms).3f)'%err)
ImageComparisonFailure: images not close: test\result_images\test_text\font_styles_pdf.png vs.
  test\result_images\test_text\expected-font_styles_pdf.png (RMS 23.409)

----------------------------------------------------------------------
Ran 1068 tests in 231.978s

FAILED (KNOWNFAIL=268, failures=2)

I'll file a ticket if I can't fix it using a debug build.

Christoph

16

Point of information: running matplotlib/tests.py on linux with 1.1.x, I get 1 image comparison failure:
failed-diff-mathfont_stix_14.png

It is all a matter of subtle differences in horizontal placement, and to my eye, the spacing looks better with what I am getting than with the expected image.

Is anyone else getting this failure? If so, maybe updating the expected image would take care of it. Otherwise, maybe the tolerance needs to be increased.

···

On 06/07/2012 09:04 AM, John Hunter wrote:

On Thu, Jun 7, 2012 at 1:55 PM, Eric Firing<efiring@...229...> wrote:

I did not see any replies to Sandro's reasonable question.

Wasn't there at least a plan for a 1.1.x release soon?

There are quite a few open pull requests for master--not many for
1.1.x--but I don't know that any are blockers.

There are no open issues tagged "release_candidate" so I'll go ahead
and make the 1.1.x release if there are no objections.

JDH

17

Christoph

···

On 6/7/2012 1:34 PM, Christoph Gohlke wrote:

On 6/7/2012 1:01 PM, John Hunter wrote:

On Thu, Jun 7, 2012 at 2:57 PM, Christoph Gohlke <cgohlke@...244...> wrote:

Is there time for another release candidate?

With the latest github branch on win-amd64-py2.7 I get both, a segfault
and a test failure in "matplotlib.tests.test_text.test_font_styles.test".

If you are getting a segfault, is there on the latest branch, would it
make sense to do an RC? It seems like we would want to fix the
segfault first. Can you give us more details on the segfault, eg
backend, conditions to reproduce, etc?

There have been many changes since the last rc. The crash is probably
related to very recent changes. Ideally the segfault should be fixed
before rc2.

The crash on win-amd64-py2.7 is with the PDF backend:

Python 2.7.3 (default, Apr 10 2012, 23:24:47) [MSC v.1500 64 bit
(AMD64)] on win32

import matplotlib as mpl
mpl.use("agg")
mpl.test(verbosity=2)

<snip>
matplotlib.tests.test_mathtext.mathtext_stixsans_65_test.test ... ok
matplotlib.tests.test_mathtext.mathtext_stixsans_65_test.test ...
KNOWNFAIL: Cannot compare svg files on this system
matplotlib.tests.test_mathtext.test_fontinfo ... ok
matplotlib.tests.test_text.test_antialiasing.test ... ok
matplotlib.tests.test_text.test_font_styles.test ... ok
matplotlib.tests.test_text.test_font_styles.test ...
crash

On win32-py2.7 the test fails but doesn't crash. The failure is that
bold font styles are not rendered bold:

Python 2.7.3 (default, Apr 10 2012, 23:31:26) [MSC v.1500 32 bit
(Intel)] on win32
<snip>

FAIL: matplotlib.tests.test_text.test_font_styles.test
----------------------------------------------------------------------
Traceback (most recent call last):
    File "X:\Python27\lib\site-packages\nose\case.py", line 197, in runTest
      self.test(*self.arg)
    File
"X:\Python27\lib\site-packages\matplotlib\testing\decorators.py", line
36, in failer
      result = f(*args, **kwargs)
    File
"X:\Python27\lib\site-packages\matplotlib\testing\decorators.py", line
140, in do_test
      '(RMS %(rms).3f)'%err)
ImageComparisonFailure: images not close:
test\result_images\test_text\font_styles.png vs.
   test\result_images\test_text\expected-font_styles.png (RMS 47.138)

======================================================================
FAIL: matplotlib.tests.test_text.test_font_styles.test
----------------------------------------------------------------------
Traceback (most recent call last):
    File "X:\Python27\lib\site-packages\nose\case.py", line 197, in runTest
      self.test(*self.arg)
    File
"X:\Python27\lib\site-packages\matplotlib\testing\decorators.py", line
36, in failer
      result = f(*args, **kwargs)
    File
"X:\Python27\lib\site-packages\matplotlib\testing\decorators.py", line
140, in do_test
      '(RMS %(rms).3f)'%err)
ImageComparisonFailure: images not close:
test\result_images\test_text\font_styles_pdf.png vs.
   test\result_images\test_text\expected-font_styles_pdf.png (RMS 23.409)

----------------------------------------------------------------------
Ran 1068 tests in 231.978s

FAILED (KNOWNFAIL=268, failures=2)

I'll file a ticket if I can't fix it using a debug build.

18

I'm not getting this error. Perhaps the tolerance should be increased or the acceptable range of freetype versions needs to be reduced.

Mike

···

On 06/07/2012 05:44 PM, Eric Firing wrote:

On 06/07/2012 09:04 AM, John Hunter wrote:

On Thu, Jun 7, 2012 at 1:55 PM, Eric Firing<efiring@...229...> wrote:

I did not see any replies to Sandro's reasonable question.

Wasn't there at least a plan for a 1.1.x release soon?

There are quite a few open pull requests for master--not many for
1.1.x--but I don't know that any are blockers.

There are no open issues tagged "release_candidate" so I'll go ahead
and make the 1.1.x release if there are no objections.

JDH

Point of information: running matplotlib/tests.py on linux with 1.1.x, I
get 1 image comparison failure:
failed-diff-mathfont_stix_14.png

It is all a matter of subtle differences in horizontal placement, and to
my eye, the spacing looks better with what I am getting than with the
expected image.

Is anyone else getting this failure? If so, maybe updating the expected
image would take care of it. Otherwise, maybe the tolerance needs to be
increased.

19

I also want to flag PR #934 as something that should go into the rc. It's currently against master but it's important enough/simple enough that it should be backported.

Mike

···

On 06/07/2012 03:04 PM, John Hunter wrote:

On Thu, Jun 7, 2012 at 1:55 PM, Eric Firing<efiring@...229...> wrote:

I did not see any replies to Sandro's reasonable question.

Wasn't there at least a plan for a 1.1.x release soon?

There are quite a few open pull requests for master--not many for
1.1.x--but I don't know that any are blockers.

There are no open issues tagged "release_candidate" so I'll go ahead
and make the 1.1.x release if there are no objections.

JDH

------------------------------------------------------------------------------
Live Security Virtual Conference
Exclusive live event will cover all the ways today's security and
threat landscape has changed and how IT managers can respond. Discussions
will include endpoint security, mobile security and the latest in malware
threats. http://www.accelacomm.com/jaw/sfrnl04242012/114/50122263/
_______________________________________________
Matplotlib-devel mailing list
Matplotlib-devel@lists.sourceforge.net
matplotlib-devel List Signup and Options