matplotlib-users Mailing List for matplotlib (Page 3)

Jouni K. Seppänen <jk...@ik...> writes:
>> I think there's clearly a bug in that review request. Does
>> _backend_gdk.c not get compiled in tests?
>
> I have a proposed fix in branch fix-gdk-strides of
>
> https://github.com/jkseppan/matplotlib.git
>
> but I don't seem to have any systems with the prerequisites to compiling
> the gdk backend. Could someone who uses that backend try this out? It
> seems that drawing images or mathtext would exercise the code.
This was recently merged to master, together with another fix from
Michael Droettboom.
-- 
Jouni K. Seppänen
http://www.iki.fi/jks

Hello,
We are pleased to announce the release of matplotlib 1.4.1.
This is a bug-fix release for the 1.4 series.
 - reverts the changes to interactive plotting so ion will work as
before in all cases fixed boxplot regressions
 - fixes for finding freetype and libpng
 - sundry unicode fixes (looking up user folders, importing
seaborn/pandas/networkx with macosx backend)
 - nbagg works with python 3 + new font awesome
 - fixed saving dialogue in QT5
Source tarballs are available on sourceforge, github, and pypi. In
addition wheels for both mac and windows are available on both pypi
and source forge and windows executable installers are available on
source forge.
http://matplotlib.org/downloads.html
https://pypi.python.org/pypi/matplotlib/1.4.1
The matplotlib dev team

Hi
Does anybody knows how to freeze (cx_freeze) a script including basemap ?
Thanks
-- 
jose m allegue
jma...@gm...

Is it possible to tell matplotlib to only plot the African continent?
http://tommycarstensen.com/python2_matplotlib_basemap_merc_bluemarble_hresolution.jpg
I can do this in gnuplot, but I can't figure out how to do it with
matplotlib/basemap.
Thanks,
Tommy

Jouni K. Seppänen <jk...@ik...> writes:
> Thomas Caswell <tca...@gm...>
> writes:
>
>> What OS and what branch are you using?
>>
>> Can you roll back to one commit before the most recent merge?
>> https://github.com/matplotlib/matplotlib/pull/3547 updated the numpy
>> api. If that fixes it, please make a new issue on github reporting
>> the problem.
>
> I think there's clearly a bug in that review request. Does
> _backend_gdk.c not get compiled in tests?
I have a proposed fix in branch fix-gdk-strides of
https://github.com/jkseppan/matplotlib.git
but I don't seem to have any systems with the prerequisites to compiling
the gdk backend. Could someone who uses that backend try this out? It
seems that drawing images or mathtext would exercise the code.
-- 
Jouni K. Seppänen
http://www.iki.fi/jks

Thomas Caswell <tca...@gm...>
writes:
> What OS and what branch are you using?
>
> Can you roll back to one commit before the most recent merge?
> https://github.com/matplotlib/matplotlib/pull/3547 updated the numpy
> api. If that fixes it, please make a new issue on github reporting
> the problem.
I think there's clearly a bug in that review request. Does
_backend_gdk.c not get compiled in tests?
-- 
Jouni K. Seppänen
http://www.iki.fi/jks

On 10/15/14 3:19 PM, Russell Owen wrote:
> I have a polar graph that works fine on linux but is blank on MacOS
> 10.9. Here is a minimal working example (using the API because it is
> part of a larger Tkinter application):
>
> #!/usr/bin/env python2
> import Tkinter
> import matplotlib
> from matplotlib.backends.backend_tkagg import FigureCanvasTkAgg
>
> root = Tkinter.Tk()
> root["bg"] = "green" # the green should be hidden by the plot
>
> plotFig = matplotlib.figure.Figure(figsize=(5, 5), frameon=False)
> figCanvas = FigureCanvasTkAgg(plotFig, root)
> figCanvas.get_tk_widget().pack()
> axis = plotFig.add_subplot(1, 1, 1, polar=True, autoscale_on=True)
>
> r = (0.1, 0.4, 0.6, 0.8)
> theta = (0, 1.5, 3.0, 4.5) # radians, 0 to right. pi/2 up
> axis.plot(theta, r, marker="o", markeredgecolor="black", markersize=3)
> figCanvas.draw()
>
> root.mainloop()
>
>
> On MacOS 10.9 when I run this from Terminal I see a green window (rather
> than the plot I expect). When I click on the plot window to bring it to
> the front, the window turns white with a thin black border (presumably
> the plot, but with no content).
>
> Details:
> - matplotlib 1.3.1
> - Tcl/Tk 8.5.11 (for good reasons, unfortunately)
> - python.org python 2.7.6, running in 32-bit mode (due to a known issue
> with MacOS 10.9 and this old version of Tcl/Tk)
>
> Any ideas how to get the plot to show?
Looks like this is an incompatibility between Tcl/Tk 8.5.11 and MacOS 
10.9. Tcl/Tk 8.5.15 does not show the problem. Time to try to build 
8.5.17rc1 from source, I guess.
-- Russell

Please try with commit 7f71233d55683b1c1bda1b365b61f912f8a154df
What version of numpy are you using, we only support 1.6+
Tom
On Fri, Oct 17, 2014 at 1:35 PM, Nils Wagner <ni...@go...> wrote:
> I am using
>
> commit 0f4a68d3b75ae5ef192b1c90d876e18ff9de86c5
> Merge: 7f71233 b279c15
> Author: Thomas A Caswell <tca...@gm...>
> Date: Fri Oct 17 10:45:39 2014 -0400
>
> on opensuse12.3
>
>
> On Fri, Oct 17, 2014 at 7:14 PM, Thomas Caswell <tca...@gm...> wrote:
>>
>> What OS and what branch are you using?
>>
>> Can you roll back to one commit before the most recent merge?
>> https://github.com/matplotlib/matplotlib/pull/3547 updated the numpy
>> api. If that fixes it, please make a new issue on github reporting
>> the problem.
>>
>>
>>
>> Tom
>>
>> On Fri, Oct 17, 2014 at 12:42 PM, Nils Wagner <ni...@go...>
>> wrote:
>> > Hi all,
>> >
>> > I cannot build matplotlib from git
>> >
>> > src/_backend_gdk.c: In function ‘pixbuf_get_pixels_array’:
>> > src/_backend_gdk.c:43:10: error: ‘PyArrayObject’ has no member named
>> > ‘strides’
>> > src/_backend_gdk.c:46:10: error: ‘PyArrayObject’ has no member named
>> > ‘base’
>> > error: command '/usr/bin/gcc' failed with exit status 1
>> >
>> >
>> >
>> > ------------------------------------------------------------------------------
>> > Comprehensive Server Monitoring with Site24x7.
>> > Monitor 10 servers for 9ドル/Month.
>> > Get alerted through email, SMS, voice calls or mobile push
>> > notifications.
>> > Take corrective actions from your mobile device.
>> > http://p.sf.net/sfu/Zoho
>> > _______________________________________________
>> > Matplotlib-users mailing list
>> > Mat...@li...
>> > https://lists.sourceforge.net/lists/listinfo/matplotlib-users
>> >
>>
>>
>>
>> --
>> Thomas Caswell
>> tca...@gm...
>
>
-- 
Thomas Caswell
tca...@gm...

What OS and what branch are you using?
Can you roll back to one commit before the most recent merge?
https://github.com/matplotlib/matplotlib/pull/3547 updated the numpy
api. If that fixes it, please make a new issue on github reporting
the problem.
Tom
On Fri, Oct 17, 2014 at 12:42 PM, Nils Wagner <ni...@go...> wrote:
> Hi all,
>
> I cannot build matplotlib from git
>
> src/_backend_gdk.c: In function ‘pixbuf_get_pixels_array’:
> src/_backend_gdk.c:43:10: error: ‘PyArrayObject’ has no member named
> ‘strides’
> src/_backend_gdk.c:46:10: error: ‘PyArrayObject’ has no member named ‘base’
> error: command '/usr/bin/gcc' failed with exit status 1
>
>
> ------------------------------------------------------------------------------
> Comprehensive Server Monitoring with Site24x7.
> Monitor 10 servers for 9ドル/Month.
> Get alerted through email, SMS, voice calls or mobile push notifications.
> Take corrective actions from your mobile device.
> http://p.sf.net/sfu/Zoho
> _______________________________________________
> Matplotlib-users mailing list
> Mat...@li...
> https://lists.sourceforge.net/lists/listinfo/matplotlib-users
>
-- 
Thomas Caswell
tca...@gm...

Hi all,
I cannot build matplotlib from git
src/_backend_gdk.c: In function ‘pixbuf_get_pixels_array’:
src/_backend_gdk.c:43:10: error: ‘PyArrayObject’ has no member named
‘strides’
src/_backend_gdk.c:46:10: error: ‘PyArrayObject’ has no member named ‘base’
error: command '/usr/bin/gcc' failed with exit status 1

On 10/17/2014 13:27, Tommy Carstensen wrote:
> To matplotlib-users,
>
> basemap bluemarble() requires PIL, which is not available for Python3.
Have you tried using pillow? https://pypi.python.org/pypi/Pillow/2.6.1
It might just a drop in replacement, but that depends on how bluemarble 
uses PIL.
http://pillow.readthedocs.org/porting-pil-to-pillow.html
Werner

Tom,
I believe the "pillow" package is the actively maintained fork of PIL.
Could be wrong on that though.
-p
On Fri, Oct 17, 2014 at 4:27 AM, Tommy Carstensen <
tom...@gm...> wrote:
> To matplotlib-users,
>
> basemap bluemarble() requires PIL, which is not available for Python3.
> What is the usual workaround, when using Python and wanting to do
> bluemarble()?
>
> Thanks,
> Tommy
>
>
> ------------------------------------------------------------------------------
> Comprehensive Server Monitoring with Site24x7.
> Monitor 10 servers for 9ドル/Month.
> Get alerted through email, SMS, voice calls or mobile push notifications.
> Take corrective actions from your mobile device.
> http://p.sf.net/sfu/Zoho
> _______________________________________________
> Matplotlib-users mailing list
> Mat...@li...
> https://lists.sourceforge.net/lists/listinfo/matplotlib-users
>

To matplotlib-users,
basemap bluemarble() requires PIL, which is not available for Python3.
What is the usual workaround, when using Python and wanting to do
bluemarble()?
Thanks,
Tommy

To matplotlib-users,
I ran the installation guide here on Mac OS:
http://matplotlib.org/basemap/users/installing.html
Except I did brew install geos and subsequently did:
export DYLD_LIBRARY_PATH=/Users/tc9/homebrew/lib
export GEOS_LIBRARY_PATH=/Users/tc9/homebrew/Cellar/geos/3.4.2
export GEOS_DIR=/Users/tc9/homebrew/Cellar/geos/3.4.2
And instead of python setup.py install I did:
/usr/local/bin/python3 setup.py install --user
I got this warning:
/Library/Frameworks/Python.framework/Versions/3.4/lib/python3.4/distutils/dist.py:260:
UserWarning: Unknown distribution option: 'namespace_packages'
 warnings.warn(msg)
When I try to run simpletest.py in the folder examples with
/usr/local/bin/python3 I get this error:
Assertion failed: (0), function query, file AbstractSTRtree.cpp, line 285.
Abort trap: 6
What am I doing wrong?
Thanks,
Tommy

Please use the link on the bottom of every message sent to this mailing
list:
https://lists.sourceforge.net/lists/listinfo/matplotlib-users
On that page, there is an option for un-subscribing.
Best,
Ray
On Thu, Oct 16, 2014 at 4:22 PM, Almerindo Paiva <alo...@gm...> wrote:
> Dear Sirs,
>
> I would like to ask you to unsubscripte my account from
> mat...@li....
>
> Kind regards,
>
> Almerindo Paiva
>
>
> ------------------------------------------------------------------------------
> Comprehensive Server Monitoring with Site24x7.
> Monitor 10 servers for 9ドル/Month.
> Get alerted through email, SMS, voice calls or mobile push notifications.
> Take corrective actions from your mobile device.
> http://p.sf.net/sfu/Zoho
> _______________________________________________
> Matplotlib-users mailing list
> Mat...@li...
> https://lists.sourceforge.net/lists/listinfo/matplotlib-users
>
>

Dear Sirs,
I would like to ask you to unsubscripte my account from
mat...@li....
Kind regards,
Almerindo Paiva

Tobais,
In the future please send all messages like this to the user mailing list (cc'd)
Try using the 1.4.1rc1 release candidate, there are windows installers
on sourceforge. If that does not work, please open a new github
issue.
Tom
On Thu, Oct 16, 2014 at 2:43 PM, Tobias Landwehr <Dol...@gm...> wrote:
> Greetings Thomas,
>
> I am sorry to disturb you, but I found on github, that you could've already
> provided a solution to my problem almost three weeks ago. I couldn't really
> understand the way how to solve the prob, which is the reason I decided to
> write you.
>
> I am currently working with Python 2.7.2 and want to import the matplotlib.
> Problems arise obviously since my username-folder contains a non-ascii sign
> (german Ä). I would prefer to not setup a second user and the user's folder
> shouldn't/can't be renamed.
>
> If you could find some time to deal with the problem over the weekend or
> give me a hint, how to solve it, I would be really grateful. Below you'll
> find the fail report.
>
> Many thanks in advance, Tobias
>
> Traceback (most recent call last):
> File "C:\Python27\workspace\ITT\src\itt\statistics\block01\LandCover.py",
> line 7, in <module>
> import matplotlib.pyplot as plt
> File "C:\Python27\lib\site-packages\matplotlib\__init__.py", line 1048, in
> <module>
> rcParams = rc_params()
> File "C:\Python27\lib\site-packages\matplotlib\__init__.py", line 897, in
> rc_params
> fname = matplotlib_fname()
> File "C:\Python27\lib\site-packages\matplotlib\__init__.py", line 759, in
> matplotlib_fname
> configdir = _get_configdir()
> File "C:\Python27\lib\site-packages\matplotlib\__init__.py", line 630, in
> _get_configdir
> return _get_config_or_cache_dir(_get_xdg_config_dir())
> File "C:\Python27\lib\site-packages\matplotlib\__init__.py", line 555, in
> _get_xdg_config_dir
> path = get_home()
> File "C:\Python27\lib\site-packages\matplotlib\__init__.py", line 323, in
> wrapper
> ret = func(*args, **kwargs)
> File "C:\Python27\lib\site-packages\matplotlib\__init__.py", line 509, in
> _get_home
> path = os.path.expanduser("~")
> File "C:\Python27\lib\ntpath.py", line 310, in expanduser
> return userhome + path[i:]
> UnicodeDecodeError: 'ascii' codec can't decode byte 0x84 in position 16:
> ordinal not in range(128)
>
-- 
Thomas Caswell
tca...@gm...

Thanks Ben. I'll take a look through the archive.
On 10/16/14, 9:43 AM, Benjamin Root wrote:
> What you are looking for is called "rasterization". Matplotlib does this
> deep within the codebase and is not accessible (believe me, I tried).
> However, there have been other discussions in this mailing list about
> how to use GDAL to rasterize a set of polygons (represented as paths),
> including some links to stack-overflow questions. It isn't a complete
> end-to-end solution, but the pieces are there.
>
> Cheers!
> Ben Root
>
>
> On Wed, Oct 15, 2014 at 2:09 PM, David Hoese <dh...@gm...
> <mailto:dh...@gm...>> wrote:
>
> For the first question, if I save the figure (as a PNG I'm guessing,
> unless you can save into a more array-like format), I'd have to make
> sure that there were no labels or ticks and that the axes fit the whole
> figure. I'd also have to get the dpi and size information correct, but I
> suppose it would be possible that way. I was hoping for something a
> little easier and in-memory. This might be the simplest answer the more
> I think about it.
>
> If I'm starting from the paths, I'd still have to write them to the
> "grid" array. I thought maybe the backend could do that and I magically
> get the image.
>
> I'll look in to using savefig and getting the data out.
>
> -Dave
>
> On 10/15/14, 12:42 PM, Joy merwin monteiro wrote:
> > pardon the query if it seems dumb, but why don't you do a savefig()
> > after plotting the data and then convert it to any format you like?
> >
> > alternatively, contour() and contourf() both create paths that can
> > be accessed:
> >
> > cf = contourf(.....)
> >
> > output = cf.collections.pop()
> > paths = output.get_paths()[i] # for the various contours
> >
> > the x,y coordinates can then be accessed as
> >
> > xcoords = paths.vertices.transpose()[0]
> > ycoords = paths.vertices.transpose()[1]
> >
> > you can then do whatever you wish with them.
> >
> > Joy
> >
> >
> > On Wed, Oct 15, 2014 at 9:11 PM, David Hoese <dh...@gm... <mailto:dh...@gm...>
> > <mailto:dh...@gm... <mailto:dh...@gm...>>> wrote:
> >
> > I've been searching and reading through source code and
> google searches
> > to see if this is possible, but no luck so far. I'm basically
> trying to
> > map some data using Basemap, use contourf to map it to an
> image, and
> > then put that image in a geotiff (or other format) for use in
> other GIS
> > programs. I have other tools for remapping data and creating
> geotiffs,
> > but the contour image looks better. All I would need to get
> this to work
> > would be an array representing the image inside the axes of a
> contourf
> > plot. I found a lot of geotiff -> Basemap png results, but I
> would like
> > the reverse.
> >
> > Since the plots are made using paths/patches I'm guessing I
> would have
> > to have a backend render the image and then extract the image
> somehow.
> > Does anyone have some tips or tricks to do something like
> this? Or am I
> > thinking about this completely wrong?
> >
> > Thanks for any help and if you could CC me in any replies it
> would be
> > much appreciated.
> >
> > -Dave
> >
> >
> ------------------------------------------------------------------------------
> > Comprehensive Server Monitoring with Site24x7.
> > Monitor 10 servers for 9ドル/Month.
> > Get alerted through email, SMS, voice calls or mobile push
> > notifications.
> > Take corrective actions from your mobile device.
> > http://p.sf.net/sfu/Zoho
> > _______________________________________________
> > Matplotlib-users mailing list
> > Mat...@li...
> <mailto:Mat...@li...>
> > <mailto:Mat...@li...
> <mailto:Mat...@li...>>
> > https://lists.sourceforge.net/lists/listinfo/matplotlib-users
> >
> >
> >
> >
> > --
> > The best ruler, when he finishes his
> > tasks and completes his affairs,
> > the people say
> > "It all happened naturally"
> >
> > - Te Tao Ch'ing
>
> ------------------------------------------------------------------------------
> Comprehensive Server Monitoring with Site24x7.
> Monitor 10 servers for 9ドル/Month.
> Get alerted through email, SMS, voice calls or mobile push
> notifications.
> Take corrective actions from your mobile device.
> http://p.sf.net/sfu/Zoho
> _______________________________________________
> Matplotlib-users mailing list
> Mat...@li...
> <mailto:Mat...@li...>
> https://lists.sourceforge.net/lists/listinfo/matplotlib-users
>
>

What you are looking for is called "rasterization". Matplotlib does this
deep within the codebase and is not accessible (believe me, I tried).
However, there have been other discussions in this mailing list about how
to use GDAL to rasterize a set of polygons (represented as paths),
including some links to stack-overflow questions. It isn't a complete
end-to-end solution, but the pieces are there.
Cheers!
Ben Root
On Wed, Oct 15, 2014 at 2:09 PM, David Hoese <dh...@gm...> wrote:
> For the first question, if I save the figure (as a PNG I'm guessing,
> unless you can save into a more array-like format), I'd have to make
> sure that there were no labels or ticks and that the axes fit the whole
> figure. I'd also have to get the dpi and size information correct, but I
> suppose it would be possible that way. I was hoping for something a
> little easier and in-memory. This might be the simplest answer the more
> I think about it.
>
> If I'm starting from the paths, I'd still have to write them to the
> "grid" array. I thought maybe the backend could do that and I magically
> get the image.
>
> I'll look in to using savefig and getting the data out.
>
> -Dave
>
> On 10/15/14, 12:42 PM, Joy merwin monteiro wrote:
> > pardon the query if it seems dumb, but why don't you do a savefig()
> > after plotting the data and then convert it to any format you like?
> >
> > alternatively, contour() and contourf() both create paths that can
> > be accessed:
> >
> > cf = contourf(.....)
> >
> > output = cf.collections.pop()
> > paths = output.get_paths()[i] # for the various contours
> >
> > the x,y coordinates can then be accessed as
> >
> > xcoords = paths.vertices.transpose()[0]
> > ycoords = paths.vertices.transpose()[1]
> >
> > you can then do whatever you wish with them.
> >
> > Joy
> >
> >
> > On Wed, Oct 15, 2014 at 9:11 PM, David Hoese <dh...@gm...
> > <mailto:dh...@gm...>> wrote:
> >
> > I've been searching and reading through source code and google
> searches
> > to see if this is possible, but no luck so far. I'm basically trying
> to
> > map some data using Basemap, use contourf to map it to an image, and
> > then put that image in a geotiff (or other format) for use in other
> GIS
> > programs. I have other tools for remapping data and creating
> geotiffs,
> > but the contour image looks better. All I would need to get this to
> work
> > would be an array representing the image inside the axes of a
> contourf
> > plot. I found a lot of geotiff -> Basemap png results, but I would
> like
> > the reverse.
> >
> > Since the plots are made using paths/patches I'm guessing I would
> have
> > to have a backend render the image and then extract the image
> somehow.
> > Does anyone have some tips or tricks to do something like this? Or
> am I
> > thinking about this completely wrong?
> >
> > Thanks for any help and if you could CC me in any replies it would be
> > much appreciated.
> >
> > -Dave
> >
> >
> ------------------------------------------------------------------------------
> > Comprehensive Server Monitoring with Site24x7.
> > Monitor 10 servers for 9ドル/Month.
> > Get alerted through email, SMS, voice calls or mobile push
> > notifications.
> > Take corrective actions from your mobile device.
> > http://p.sf.net/sfu/Zoho
> > _______________________________________________
> > Matplotlib-users mailing list
> > Mat...@li...
> > <mailto:Mat...@li...>
> > https://lists.sourceforge.net/lists/listinfo/matplotlib-users
> >
> >
> >
> >
> > --
> > The best ruler, when he finishes his
> > tasks and completes his affairs,
> > the people say
> > "It all happened naturally"
> >
> > - Te Tao Ch'ing
>
>
> ------------------------------------------------------------------------------
> Comprehensive Server Monitoring with Site24x7.
> Monitor 10 servers for 9ドル/Month.
> Get alerted through email, SMS, voice calls or mobile push notifications.
> Take corrective actions from your mobile device.
> http://p.sf.net/sfu/Zoho
> _______________________________________________
> Matplotlib-users mailing list
> Mat...@li...
> https://lists.sourceforge.net/lists/listinfo/matplotlib-users
>

Awesome, thanks!
On Thu, Oct 16, 2014 at 8:14 AM, Sandro Tosi <mo...@de...> wrote:
> Hi!
> I just uploaded mpl 1.4.1-rc1 in Debian, so it can geta bit of
> exposure even on the weird HW we still support.
>
> Cheers,
> Sandro
>
> On Tue, Oct 14, 2014 at 5:40 PM, Thomas Caswell <tca...@gm...> wrote:
>> I am happy to announce that I have tagged a release candidate for
>> 1.4.1. This is a bug-fix release which fixes most of the bug that
>> popped up in 1.4.0 including:
>>
>> - setup.py does not die when freetype is not installed
>> - reverts the changes to interactive plotting so `ion` will work as expected
>> - sundry unicode fixes (looking up user folders, importing
>> seaborn/pandas/networkx with macosx backend
>> - fixed boxplot regressions
>>
>> The tarball is available from github, sourceforge and can be install via
>>
>> pip install matplotlib==1.4.1rc1
>>
>> Tom
>>
>>
>>
>> --
>> Thomas Caswell
>> tca...@gm...
>>
>> ------------------------------------------------------------------------------
>> Comprehensive Server Monitoring with Site24x7.
>> Monitor 10 servers for 9ドル/Month.
>> Get alerted through email, SMS, voice calls or mobile push notifications.
>> Take corrective actions from your mobile device.
>> http://p.sf.net/sfu/Zoho
>> _______________________________________________
>> Matplotlib-devel mailing list
>> Mat...@li...
>> https://lists.sourceforge.net/lists/listinfo/matplotlib-devel
>
>
>
> --
> Sandro Tosi (aka morph, morpheus, matrixhasu)
> My website: http://matrixhasu.altervista.org/
> Me at Debian: http://wiki.debian.org/SandroTosi
-- 
Thomas Caswell
tca...@gm...

Hi!
I just uploaded mpl 1.4.1-rc1 in Debian, so it can geta bit of
exposure even on the weird HW we still support.
Cheers,
Sandro
On Tue, Oct 14, 2014 at 5:40 PM, Thomas Caswell <tca...@gm...> wrote:
> I am happy to announce that I have tagged a release candidate for
> 1.4.1. This is a bug-fix release which fixes most of the bug that
> popped up in 1.4.0 including:
>
> - setup.py does not die when freetype is not installed
> - reverts the changes to interactive plotting so `ion` will work as expected
> - sundry unicode fixes (looking up user folders, importing
> seaborn/pandas/networkx with macosx backend
> - fixed boxplot regressions
>
> The tarball is available from github, sourceforge and can be install via
>
> pip install matplotlib==1.4.1rc1
>
> Tom
>
>
>
> --
> Thomas Caswell
> tca...@gm...
>
> ------------------------------------------------------------------------------
> Comprehensive Server Monitoring with Site24x7.
> Monitor 10 servers for 9ドル/Month.
> Get alerted through email, SMS, voice calls or mobile push notifications.
> Take corrective actions from your mobile device.
> http://p.sf.net/sfu/Zoho
> _______________________________________________
> Matplotlib-devel mailing list
> Mat...@li...
> https://lists.sourceforge.net/lists/listinfo/matplotlib-devel
-- 
Sandro Tosi (aka morph, morpheus, matrixhasu)
My website: http://matrixhasu.altervista.org/
Me at Debian: http://wiki.debian.org/SandroTosi

Thanks for the reiteration/clarification... Do you see any reason why a fail-safe check similar to what I proposed would be ill advised?
MJ
-----Original Message-----
From: Matthew Brett [mailto:mat...@gm...] 
Sent: Wednesday, October 15, 2014 3:44 PM
To: Mark Janikas
Cc: Thomas Caswell; Phil Elson; Benjamin Root; Matplotlib Users
Subject: Re: [Matplotlib-users] FW: traceback when import matplotlib.pyplot twice
Hi,
On Wed, Oct 15, 2014 at 2:56 PM, Mark Janikas <mja...@es...> wrote:
>
> When I replaced the file I got the 1st error below. As you had pointed out earlier... this is strange. It only occurs when you run it more than once... strange indeed... it is like a manager is being created and deleted but perhaps the reference hasn’t
Just to echo what Tom said, the error suggests that ``gc.collect()`` is being called after the gc module is most of the way through being torn down, so that the ``collect`` function has already been deleted...
Cheers,
Matthew

Hi,
On Wed, Oct 15, 2014 at 2:56 PM, Mark Janikas <mja...@es...> wrote:
>
> When I replaced the file I got the 1st error below. As you had pointed out earlier... this is strange. It only occurs when you run it more than once... strange indeed... it is like a manager is being created and deleted but perhaps the reference hasn’t
Just to echo what Tom said, the error suggests that ``gc.collect()``
is being called after the gc module is most of the way through being
torn down, so that the ``collect`` function has already been
deleted...
Cheers,
Matthew

I have a polar graph that works fine on linux but is blank on MacOS 
10.9. Here is a minimal working example (using the API because it is 
part of a larger Tkinter application):
#!/usr/bin/env python2
import Tkinter
import matplotlib
from matplotlib.backends.backend_tkagg import FigureCanvasTkAgg
root = Tkinter.Tk()
root["bg"] = "green" # the green should be hidden by the plot
plotFig = matplotlib.figure.Figure(figsize=(5, 5), frameon=False)
figCanvas = FigureCanvasTkAgg(plotFig, root)
figCanvas.get_tk_widget().pack()
axis = plotFig.add_subplot(1, 1, 1, polar=True, autoscale_on=True)
r = (0.1, 0.4, 0.6, 0.8)
theta = (0, 1.5, 3.0, 4.5) # radians, 0 to right. pi/2 up
axis.plot(theta, r, marker="o", markeredgecolor="black", markersize=3)
figCanvas.draw()
root.mainloop()
On MacOS 10.9 when I run this from Terminal I see a green window (rather 
than the plot I expect). When I click on the plot window to bring it to 
the front, the window turns white with a thin black border (presumably 
the plot, but with no content).
Details:
- matplotlib 1.3.1
- Tcl/Tk 8.5.11 (for good reasons, unfortunately)
- python.org python 2.7.6, running in 32-bit mode (due to a known issue 
with MacOS 10.9 and this old version of Tcl/Tk)
Any ideas how to get the plot to show?
-- Russell

When I replaced the file I got the 1st error below. As you had pointed out earlier... this is strange. It only occurs when you run it more than once... strange indeed... it is like a manager is being created and deleted but perhaps the reference hasn’t... so you get a no-op on the subsequent import.. but are then left with a manager that is None....?? Thanks for the tip BTW... I replaced all my None comparisons to "is" and "is not". I went ahead and added the comparisons to your changes and it works... but now you need to assure that you do not gc.collect unless you had at least one Non-None manager or else you get the 2nd traceback. The code that works on my end is the last image... not sure if it passes your regression tests..??..??
Thanks so much for all of your help... please let me know if there is anything else I can do to help.
MJ
[cid:image001.png@01CFE881.FB6CCAF0]
[cid:image002.png@01CFE888.2A514840]
[cid:image003.png@01CFE888.2A514840]
-----Original Message-----
From: Thomas Caswell [mailto:tca...@gm...]
Sent: Saturday, October 11, 2014 2:18 PM
To: Phil Elson
Cc: Mark Janikas; Benjamin Root; Matplotlib Users
Subject: Re: [Matplotlib-users] FW: traceback when import matplotlib.pyplot twice
See https://github.com/matplotlib/matplotlib/pull/3638
That might help to make sure that things tear them selves down in the right order.
Tom
On Sat, Oct 11, 2014 at 10:12 AM, Phil Elson <pel...@gm...> wrote:
>
> On 10 October 2014 19:10, Thomas Caswell <tca...@gm...> wrote:
>>
>> I suspect a better fix is to change all of the staticmethods ->
>> classmethods
>
>
>
> +1
--
Thomas Caswell
tca...@gm...