matplotlib-devel Mailing List for matplotlib (Page 2)

On Sun, Feb 26, 2006 at 01:00:07AM -0600, Albert Chin wrote:
> On Fri, Feb 24, 2006 at 12:15:29AM -0600, Albert Chin wrote:
> > On Wed, Feb 22, 2006 at 09:24:38PM -0600, Albert Chin wrote:
> > > 
> > > The failure occurs when displaying the data. I've also tried
> > > anim_tk.py to determine if it was a problem with the GTK+ backend but
> > > anim_tk.py fails in the same way.
> > 
> > I did some more digging and the "terminate called after throwing an
> > instance ..." error is coming from G++. Maybe something funk with C++
> > exceptions on HP/PA that isn't triggered on HP/IA.
> 
> The throw Py::RuntimeError is coming from src/ft2font.cpp. There's a
> global that holds information for FreeType libraries:
> FT_Library _ft2Library;
> 
> This is initialized when Python loads the module. That works fine.
> However, sometime after, _ft2Library is reinitialized to 0. This
> doesn't occur the _first_ time matplotlib is run. Only on subsequent
> runs when it reads in the font cache file does this occur. Odd.
Ok, I think I found the problem. src/ft2font.cpp is used by
_nc_backend_agg.so and ft2font.so. They _both_ have a global
_ft2Library. This isn't good. On HP-UX/PA, it seems the _ft2Library
from ft2font.so is used when matplotlib programs are run first.
However, after the cache file is generated, it seems to start using
the _ft2Library from ft2font.so but then when 'FT2Font()' is called,
the _ft2Library from _nc_backend_agg.so is used.
What we really want is just one _ft2Library, the one from
src/ft2font.cpp. src/_nc_backend_agg.cpp does need the functions in
src/ft2font.cpp but I think _nc_backend_agg.so should depend on
ft2font.so, not load a private copy of src/ft2font.cpp and expect
cross-platform behavior wrt a duplicate global _ft2Library.
Another alternative is to have anything requiring src/ft2font.cpp to
simply 'import ft2font'. For the moment, I've remove src/ft2font.cpp
from all modules except ft2font.so. This seems to work. Patch
attached. Should I submit patches to PyImport_ImportModule("ft2font")
as well?
-- 
albert chin (ch...@th...)

John Hunter <jdh...@ac...> 
writes:
> There have been a lot of changes to the rc file of late, and I fear
> these may be getting lost on users. It is not uncommon for me to see
> someone launch mpl and get a host of rc warnings.
It would probably help if the warnings did not merely say
"Unrecognized key bar in group foo" but something like "Option foo.bar
has been replaced by baz.quux, see release notes of version 0.87" so
the user would have some idea how to fix the problem. This wouldn't
probably require very much effort, if you are going to record the
changes somewhere in any case.
Ted Drain suggested having a version number in the rc file. How about
showing the user a brief list of release notes for all versions since
the one mentioned in the rc file?
-- 
Jouni

Hmm - I just ran your test and everything seems to work fine. I'm using 
MPL 0.85 and Qt 3.3.5. I did a diff of backend_qt.py and backend_qtagg.py 
from 0.85 to the current CVS repository and they're identical. So I'm not 
sure what could be going on...
Ted
At 10:30 AM 2/26/2006, Darren Dale wrote:
>On Sunday 26 February 2006 1:06 pm, Ted Drain wrote:
> > Darren,
> > I'll have someone take a look at it soon. We're right in the middle of a
> > big delivery so we don't really have a lot of time to spend on this right
> > now...
>
>Thanks, Ted. I'm not in a rush.
>
>I'm about to start my first python-based gui project, and want to use Qt. 
>I'll
>also continue to poke around at these problems myself as I find time.
>
>Darren
>
>
> > At 08:20 AM 2/26/2006, Darren Dale wrote:
> > >Respectfully: Is the QtAgg backend supported?
> > >
> > >On Saturday 18 February 2006 6:01 pm, Darren Dale wrote:
> > > > I'm having two problems with the QtAgg backend. If I do:
> > > > >>> from pylab import *
> > > > >>> plot([1,2])
> > > >
> > > > [<matplotlib.lines.Line2D instance at 0x2aaab1ead830>]
> > > >
> > > > >>> show()
> > > >
> > > > My figure window should be 8in x 6in, but instead it is 6in tall and as
> > > > wide as my screen will allow. So thats the first issue. The second is
> > > > this: if I close that window, and then do
> > > >
> > > > >>> plot([1,2])
> > > >
> > > > [<matplotlib.lines.Line2D instance at 0x2aaab1ebbcf8>]
> > > >
> > > > >>> show()
> > > >
> > > > I get the following traceback:
> > > >
> > > > Traceback (most recent call last):
> > > > File "<stdin>", line 1, in ?
> > > > File
> > > > "/usr/lib64/python2.4/site-packages/matplotlib/backends/backend_qt.py",
> > > > line 47, in show
> > > > manager.window.show()
> > > > RuntimeError: underlying C/C++ object has been deleted
> > > >
> > > > I'm using qt version 3.3.4, with cvs mpl 0.86.2, on a gentoo linux
> > > > system. The only nondefault rc setting is the choice of backend.
> > > >
> > > > Darren
> > > >
> > > >
> > > > -------------------------------------------------------
> > > > This SF.net email is sponsored by: Splunk Inc. Do you grep through log
> > > > files for problems? Stop! Download the new AJAX search engine that
> > > > makes searching your log files as easy as surfing the web. DOWNLOAD
> > > > SPLUNK!
> > > >
> > > > _______________________________________________
> > > > Matplotlib-devel mailing list
> > > > Mat...@li...
> > > > https://lists.sourceforge.net/lists/listinfo/matplotlib-devel
> > >
> > >--
> > >Darren S. Dale, Ph.D.
> > >dd...@co...
> > >
> > >
> > >-------------------------------------------------------
> > >This SF.Net email is sponsored by xPML, a groundbreaking scripting
> > > language that extends applications into web and mobile media. Attend the
> > > live webcast and join the prime developer group breaking into this new
> > > coding territory! _______________________________________________
> > >Matplotlib-devel mailing list
> > >Mat...@li...
> > >https://lists.sourceforge.net/lists/listinfo/matplotlib-devel
> >
> > -------------------------------------------------------
> > This SF.Net email is sponsored by xPML, a groundbreaking scripting language
> > that extends applications into web and mobile media. Attend the live
> > webcast and join the prime developer group breaking into this new coding
> > territory!
> > http://sel.as-us.falkag.net/sel?cmd=lnk&kid=110944&bid=241720&dat=121642
> > _______________________________________________
> > Matplotlib-devel mailing list
> > Mat...@li...
> > https://lists.sourceforge.net/lists/listinfo/matplotlib-devel
>
>--
>Darren S. Dale, Ph.D.
>dd...@co...
>
>
>-------------------------------------------------------
>This SF.Net email is sponsored by xPML, a groundbreaking scripting language
>that extends applications into web and mobile media. Attend the live webcast
>and join the prime developer group breaking into this new coding territory!
>http://sel.as-us.falkag.net/sel?cmd=lnk&kid=110944&bid=241720&dat=121642
>_______________________________________________
>Matplotlib-devel mailing list
>Mat...@li...
>https://lists.sourceforge.net/lists/listinfo/matplotlib-devel

Something else that might be useful is to put a RC file version number in 
the RC file itself (i.e. a version on the set of RC inputs that are 
there). This would allow MPL to check the version number of the RC file 
the user is inputting against the current number in the software. At the 
least, this allows for an error/warning message telling someone that there 
input is out of date. If you're more ambitious, you can write RC file 
version converters so that older versions are auto-converted (and maybe 
even auto-updated) to the current version when used.
Ted
At 09:49 AM 2/26/2006, Darren Dale wrote:
>On Friday 24 February 2006 14:57, John Hunter wrote:
> > There have been a lot of changes to the rc file of late, and I fear
> > these may be getting lost on users. It is not uncommon for me to see
> > someone launch mpl and get a host of rc warnings.
> >
> > I think we should put these in API_CHANGES and announce them a little
> > more prominently in the release notes and on the user's list.
>
>I'm probably the guilty party here, although I have tried to make visible
>announcements on the users list and clear explanations in the CHANGELOG.
>
>
>-------------------------------------------------------
>This SF.Net email is sponsored by xPML, a groundbreaking scripting language
>that extends applications into web and mobile media. Attend the live webcast
>and join the prime developer group breaking into this new coding territory!
>http://sel.as-us.falkag.net/sel?cmd=lnk&kid=110944&bid=241720&dat=121642
>_______________________________________________
>Matplotlib-devel mailing list
>Mat...@li...
>https://lists.sourceforge.net/lists/listinfo/matplotlib-devel

On Sunday 26 February 2006 1:06 pm, Ted Drain wrote:
> Darren,
> I'll have someone take a look at it soon. We're right in the middle of a
> big delivery so we don't really have a lot of time to spend on this right
> now...
Thanks, Ted. I'm not in a rush. 
I'm about to start my first python-based gui project, and want to use Qt. I'll 
also continue to poke around at these problems myself as I find time.
Darren
> At 08:20 AM 2/26/2006, Darren Dale wrote:
> >Respectfully: Is the QtAgg backend supported?
> >
> >On Saturday 18 February 2006 6:01 pm, Darren Dale wrote:
> > > I'm having two problems with the QtAgg backend. If I do:
> > > >>> from pylab import *
> > > >>> plot([1,2])
> > >
> > > [<matplotlib.lines.Line2D instance at 0x2aaab1ead830>]
> > >
> > > >>> show()
> > >
> > > My figure window should be 8in x 6in, but instead it is 6in tall and as
> > > wide as my screen will allow. So thats the first issue. The second is
> > > this: if I close that window, and then do
> > >
> > > >>> plot([1,2])
> > >
> > > [<matplotlib.lines.Line2D instance at 0x2aaab1ebbcf8>]
> > >
> > > >>> show()
> > >
> > > I get the following traceback:
> > >
> > > Traceback (most recent call last):
> > > File "<stdin>", line 1, in ?
> > > File
> > > "/usr/lib64/python2.4/site-packages/matplotlib/backends/backend_qt.py",
> > > line 47, in show
> > > manager.window.show()
> > > RuntimeError: underlying C/C++ object has been deleted
> > >
> > > I'm using qt version 3.3.4, with cvs mpl 0.86.2, on a gentoo linux
> > > system. The only nondefault rc setting is the choice of backend.
> > >
> > > Darren
> > >
> > >
> > > -------------------------------------------------------
> > > This SF.net email is sponsored by: Splunk Inc. Do you grep through log
> > > files for problems? Stop! Download the new AJAX search engine that
> > > makes searching your log files as easy as surfing the web. DOWNLOAD
> > > SPLUNK!
> > >
> > > _______________________________________________
> > > Matplotlib-devel mailing list
> > > Mat...@li...
> > > https://lists.sourceforge.net/lists/listinfo/matplotlib-devel
> >
> >--
> >Darren S. Dale, Ph.D.
> >dd...@co...
> >
> >
> >-------------------------------------------------------
> >This SF.Net email is sponsored by xPML, a groundbreaking scripting
> > language that extends applications into web and mobile media. Attend the
> > live webcast and join the prime developer group breaking into this new
> > coding territory! _______________________________________________
> >Matplotlib-devel mailing list
> >Mat...@li...
> >https://lists.sourceforge.net/lists/listinfo/matplotlib-devel
>
> -------------------------------------------------------
> This SF.Net email is sponsored by xPML, a groundbreaking scripting language
> that extends applications into web and mobile media. Attend the live
> webcast and join the prime developer group breaking into this new coding
> territory!
> http://sel.as-us.falkag.net/sel?cmd=lnk&kid=110944&bid=241720&dat=121642
> _______________________________________________
> Matplotlib-devel mailing list
> Mat...@li...
> https://lists.sourceforge.net/lists/listinfo/matplotlib-devel
-- 
Darren S. Dale, Ph.D.
dd...@co...

Darren,
I'll have someone take a look at it soon. We're right in the middle of a 
big delivery so we don't really have a lot of time to spend on this right 
now...
Ted
At 08:20 AM 2/26/2006, Darren Dale wrote:
>Respectfully: Is the QtAgg backend supported?
>
>On Saturday 18 February 2006 6:01 pm, Darren Dale wrote:
> > I'm having two problems with the QtAgg backend. If I do:
> > >>> from pylab import *
> > >>> plot([1,2])
> >
> > [<matplotlib.lines.Line2D instance at 0x2aaab1ead830>]
> >
> > >>> show()
> >
> > My figure window should be 8in x 6in, but instead it is 6in tall and as
> > wide as my screen will allow. So thats the first issue. The second is this:
> > if I close that window, and then do
> >
> > >>> plot([1,2])
> >
> > [<matplotlib.lines.Line2D instance at 0x2aaab1ebbcf8>]
> >
> > >>> show()
> >
> > I get the following traceback:
> >
> > Traceback (most recent call last):
> > File "<stdin>", line 1, in ?
> > File
> > "/usr/lib64/python2.4/site-packages/matplotlib/backends/backend_qt.py",
> > line 47, in show
> > manager.window.show()
> > RuntimeError: underlying C/C++ object has been deleted
> >
> > I'm using qt version 3.3.4, with cvs mpl 0.86.2, on a gentoo linux system.
> > The only nondefault rc setting is the choice of backend.
> >
> > Darren
> >
> >
> > -------------------------------------------------------
> > This SF.net email is sponsored by: Splunk Inc. Do you grep through log
> > files for problems? Stop! Download the new AJAX search engine that makes
> > searching your log files as easy as surfing the web. DOWNLOAD SPLUNK!
> >
> > _______________________________________________
> > Matplotlib-devel mailing list
> > Mat...@li...
> > https://lists.sourceforge.net/lists/listinfo/matplotlib-devel
>
>--
>Darren S. Dale, Ph.D.
>dd...@co...
>
>
>-------------------------------------------------------
>This SF.Net email is sponsored by xPML, a groundbreaking scripting language
>that extends applications into web and mobile media. Attend the live webcast
>and join the prime developer group breaking into this new coding territory!
>_______________________________________________
>Matplotlib-devel mailing list
>Mat...@li...
>https://lists.sourceforge.net/lists/listinfo/matplotlib-devel

>>>>> "Darren" == Darren Dale <dd...@co...> writes:
 Darren> I'm probably the guilty party here, although I have tried
 Darren> to make visible announcements on the users list and clear
 Darren> explanations in the CHANGELOG.
You're not alone -- I am guilty too, eg splitting out the xtick and
ytick settings... I just notice more and more that people in my group
are getting these kinds of warnings, so I thought it was worht
mentioning.
JDH

On Friday 24 February 2006 14:57, John Hunter wrote:
> There have been a lot of changes to the rc file of late, and I fear
> these may be getting lost on users. It is not uncommon for me to see
> someone launch mpl and get a host of rc warnings.
>
> I think we should put these in API_CHANGES and announce them a little
> more prominently in the release notes and on the user's list.
I'm probably the guilty party here, although I have tried to make visible 
announcements on the users list and clear explanations in the CHANGELOG.

>>>>> "Eric" == Eric Firing <ef...@ha...> writes:
 Eric> John, I updated from CVS, made a one-line change to
 Eric> contour.py to add alpha support, and tested using
 Eric> contour_demo.py from bash shell--and found that with linux
 Eric> default gtk-agg backend, the gtk.main() doesn't return when
 Eric> the plot window is closed. I have to Ctl-C to get out.
Steve's recent change should fix your problem -- it was my bad.
JDH

Respectfully: Is the QtAgg backend supported?
On Saturday 18 February 2006 6:01 pm, Darren Dale wrote:
> I'm having two problems with the QtAgg backend. If I do:
> >>> from pylab import *
> >>> plot([1,2])
>
> [<matplotlib.lines.Line2D instance at 0x2aaab1ead830>]
>
> >>> show()
>
> My figure window should be 8in x 6in, but instead it is 6in tall and as
> wide as my screen will allow. So thats the first issue. The second is thi=
s:
> if I close that window, and then do
>
> >>> plot([1,2])
>
> [<matplotlib.lines.Line2D instance at 0x2aaab1ebbcf8>]
>
> >>> show()
>
> I get the following traceback:
>
> Traceback (most recent call last):
> =A0 File "<stdin>", line 1, in ?
> =A0 File
> "/usr/lib64/python2.4/site-packages/matplotlib/backends/backend_qt.py",
> line 47, in show
> =A0 =A0 manager.window.show()
> RuntimeError: underlying C/C++ object has been deleted
>
> I'm using qt version 3.3.4, with cvs mpl 0.86.2, on a gentoo linux system.
> The only nondefault rc setting is the choice of backend.
>
> Darren
>
>
> -------------------------------------------------------
> This SF.net email is sponsored by: Splunk Inc. Do you grep through log
> files for problems? Stop! Download the new AJAX search engine that makes
> searching your log files as easy as surfing the web. DOWNLOAD SPLUNK!
> http://sel.as-us.falkag.net/sel?cmd=3Dlnk&kid=103432&bid#0486&dat=121642
> _______________________________________________
> Matplotlib-devel mailing list
> Mat...@li...
> https://lists.sourceforge.net/lists/listinfo/matplotlib-devel
=2D-=20
Darren S. Dale, Ph.D.
dd...@co...

On Saturday 25 February 2006 8:01 pm, Andrew Straw wrote:
> Hi All,
>
> It looks like SourceForge has (finally) adopted site-wide subversion
> repositories: http://sourceforge.net/docs/E09/
>
> What do people feel about migrating from CVS to svn? SourceForge
> apparently even has an automated tool to do so. They also state that
> there will be no more delays between developer access and anonymous access.
>
> Personally I'm +1 on switching -- matplotlib is the only project I keep
> up with that still uses CVS.
+1 for switching.

On Fri, Feb 24, 2006 at 12:15:29AM -0600, Albert Chin wrote:
> On Wed, Feb 22, 2006 at 09:24:38PM -0600, Albert Chin wrote:
> > 
> > The failure occurs when displaying the data. I've also tried
> > anim_tk.py to determine if it was a problem with the GTK+ backend but
> > anim_tk.py fails in the same way.
> 
> I did some more digging and the "terminate called after throwing an
> instance ..." error is coming from G++. Maybe something funk with C++
> exceptions on HP/PA that isn't triggered on HP/IA.
The throw Py::RuntimeError is coming from src/ft2font.cpp. There's a
global that holds information for FreeType libraries:
 FT_Library _ft2Library;
This is initialized when Python loads the module. That works fine.
However, sometime after, _ft2Library is reinitialized to 0. This
doesn't occur the _first_ time matplotlib is run. Only on subsequent
runs when it reads in the font cache file does this occur. Odd.
-- 
albert chin (ch...@th...)

I can't speak about switchers using the SourceForge site, since it's
only been recently they've offered subversion. Here are various projects
I know about and their histories:
Python - switched to svn from cvs some time ago, (I think) dropping
SourceForge hosting of their repository in the process. I'm not close to
the development process, but as an occasional source-tree grabber and
c.l.py reader, I didn't hear of any issues.
numpy/scipy - switched to svn from cvs several months ago (September?).
there were some initial problems converting the history with the cvs2svn
script, but I think they were worked out, perhaps with the help of the
subversion developers. as far as I know, there have been no issues with
svn itself once the history was converted.
ipython - I think they are very pleased with subversion. They're
maintaining one branch and actively developing next generation ipython
(what will hopefully become v1) in another.
Personally, the biggest win for me with subversion is that I can do "svn
diff" and it's nearly instant -- no network access is needed. It would
also be nice to finally be rid of the asynchronous developer vs.
anonymous access that sourceforge's cvs solution entails.
Cheers!
Andrew
Eric Firing wrote:
> Andrew Straw wrote:
>
>> Hi All,
>>
>> It looks like SourceForge has (finally) adopted site-wide subversion
>> repositories: http://sourceforge.net/docs/E09/
>>
>> What do people feel about migrating from CVS to svn? SourceForge
>> apparently even has an automated tool to do so. They also state that
>> there will be no more delays between developer access and anonymous
>> access.
>>
>> Personally I'm +1 on switching -- matplotlib is the only project I keep
>> up with that still uses CVS.
>>
>> Cheers!
>> Andrew
>
>
> Andrew,
>
> I don't mind switching; I installed svn on my main machine some time
> ago so I could grab numpy, so I presume that using it for mpl will not
> be difficult. Do you know what the experience has been in other
> projects that have switched? Everything smooth? Or a week or two of
> chaos?
>
> Eric

Andrew Straw wrote:
> Hi All,
> 
> It looks like SourceForge has (finally) adopted site-wide subversion
> repositories: http://sourceforge.net/docs/E09/
> 
> What do people feel about migrating from CVS to svn? SourceForge
> apparently even has an automated tool to do so. They also state that
> there will be no more delays between developer access and anonymous access.
> 
> Personally I'm +1 on switching -- matplotlib is the only project I keep
> up with that still uses CVS.
> 
> Cheers!
> Andrew
Andrew,
I don't mind switching; I installed svn on my main machine some time ago 
so I could grab numpy, so I presume that using it for mpl will not be 
difficult. Do you know what the experience has been in other projects 
that have switched? Everything smooth? Or a week or two of chaos?
Eric

Eric Firing wrote:
> John,
> 
> I updated from CVS, made a one-line change to contour.py to add alpha 
> support, and tested using contour_demo.py from bash shell--and found 
> that with linux default gtk-agg backend, the gtk.main() doesn't return 
> when the plot window is closed. I have to Ctl-C to get out.
> 
> Eric
John,
The following patch fixes the problem for me, and makes sense to me:
Index: backend_gtk.py
===================================================================
RCS file: 
/cvsroot/matplotlib/matplotlib/lib/matplotlib/backends/backend_gtk.py,v
retrieving revision 1.123
diff -u -r1.123 backend_gtk.py
--- backend_gtk.py 24 Feb 2006 20:00:28 -0000 1.123
+++ backend_gtk.py 26 Feb 2006 02:30:42 -0000
@@ -76,8 +76,8 @@
 manager.window.show()
 if mainloop and gtk.main_level() == 0:
- gtk.main()
 show._mainloop = True
+ gtk.main()
 show._mainloop = False
 def new_figure_manager(num, *args, **kwargs):
You have show._mainloop = True after the call to gtk.main(), but that 
doesn't return; when the windows are deleted, the mainloop is still 
running but the flag has not been set, so deleting the last window does 
not end the loop. So the obvious solution (above) is to set the flag 
before calling gtk.main()
I was about to simply make the change, but decided to leave it to you, 
since I don't understand the bug you were trying to fix in the first 
place. Maybe some sort of locking is actually needed to make sure 
nothing unexpected can happen in another thread between the setting of 
the flag and the call to gtk.main().
Eric

Hi All,
It looks like SourceForge has (finally) adopted site-wide subversion
repositories: http://sourceforge.net/docs/E09/
What do people feel about migrating from CVS to svn? SourceForge
apparently even has an automated tool to do so. They also state that
there will be no more delays between developer access and anonymous access.
Personally I'm +1 on switching -- matplotlib is the only project I keep
up with that still uses CVS.
Cheers!
Andrew

John,
I updated from CVS, made a one-line change to contour.py to add alpha 
support, and tested using contour_demo.py from bash shell--and found 
that with linux default gtk-agg backend, the gtk.main() doesn't return 
when the plot window is closed. I have to Ctl-C to get out.
Eric

>>>>> "David" == David TREMOUILLES <dav...@gm...> writes:
 >>> 2006年2月24日, David Huard <dav...@gm... >: What if I
 >>> want it to be non-interactive ? I want to save graphics
 >>> directly to a file from a script, without them being
 >>> displayed. It's weird that in this case close() crashes the
 >>> whole thing.
 David> Yes your wright the script have to work in non interactive
 David> mode. (I was just trying to give an explanation to Emmanuel
 David> problem). The following script should work (but it crashes
 David> like Emmanuel reported):
 David> #!/usr/bin/python import matplotlib import pylab #Turn off
 David> interactive mode matplotlib.interactive(False) #plot figure
 David> 1 pylab.figure() pylab.plot([1,2,3],[1,2,3])
 David> pylab.savefig("test_pylab1.png") pylab.close() #plot figure
 David> 2 pylab.figure() pylab.plot([1,2,3],[1,2,3])
 David> pylab.savefig("test_pylab2.png") pylab.close()
This is a bug and I fixed it in CVS (Steve take a look and see if you
agree with the show._mainloop approach) but basically you don't want
to use matplotlib this way.
If you do not want the figure to pop-up, that is you only want to
create a PNG and move on, just use the Agg backend rather than GTK*.
Eg
python myscript.py -dAgg
The problem you experienced arose because you closed all the figures
before you called show in non-interactive mode. For a GUI this makes
no sense, because you are telling the GUI to close all the windows
before it shows any of them. 
I fixed the GTK backend to not try and quit gtk before starting it :-)
In summary, I recommend:
 * in non-interactive mode with a GUI backend, always end your script
 with show. This starts the GUI mainloop. Do not close all your
 figures before you call show.
 * if you don't want an image to pop up, use non-interactive mode,
 and a non-GUI backend, and show is not required but it doesn't
 hurt either.
JDH

There have been a lot of changes to the rc file of late, and I fear
these may be getting lost on users. It is not uncommon for me to see
someone launch mpl and get a host of rc warnings.
I think we should put these in API_CHANGES and announce them a little
more prominently in the release notes and on the user's list.
Cheers,
JDH

Jeff,
I am copying this to John and Charlie so they will know that all known 
imshow-related problems have been resolved (correct?). (Also to the 
devel list, just in case.) This and the previous message were not copied 
to either of the lists.
What is the difference between numpy.ma and the other masked array 
implementations that caused the exception? Is this something that 
should be fixed in numpy.ma?
Thanks.
Eric
Jeff Whitaker wrote:
> Andrew Jaffe wrote:
> 
>>
>>> Andrew: (val-vmin)/float(vmax-vmin) should be equivalent to the 
>>> original, since I believe vmin and vmax are always scalars.
>>>
>>> BTW: you can trigger the numpy.ma/imshow problem by uncommenting the 
>>> 'colorbar(im)' line in the image_masked.py example. With the original
>>>
>>> result = (1.0/(vmax-vmin))*(val-vmin)
>>> line in colors.py, it will throw an exception with numpy 0.9.5, but
>>>
>>> result = (val-vmin)/float(vmax-vmin)
>>>
>>> works.
>>>
>>> -Jeff
>>
>>
>> Jeff,
>>
>> Great -- thanks!
>>
>> Will you commit this change to CVS?
>>
>> Andrew
>>
> 
> Done.
> 
> -Jeff
> 

Hi,
What one can do is :
f =3D figure(1)
... some figure
f.clear()
f =3D figure(1)
... another figure
f.clear()
However, it is only a work around, not an actual solution...
Cheers,
David
2006年2月24日, David TREMOUILLES <dav...@gm...>:
>
> >>2006年2月24日, David Huard <dav...@gm... >:
> >>What if I want it to be non-interactive ? I want to save graphics
> >>directly to a file from a script, without them being displayed. It'=
s
> >>weird that in this case close() crashes the whole thing.
>
> Yes your wright the script have to work in non interactive mode. (I was
> just trying to give an explanation to Emmanuel problem).
> The following script should work (but it crashes like Emmanuel reported)=
:
>
> #!/usr/bin/python
> import matplotlib
> import pylab
> #Turn off interactive mode
> matplotlib.interactive(False)
> #plot figure 1
> pylab.figure()
> pylab.plot([1,2,3],[1,2,3])
> pylab.savefig("test_pylab1.png")
> pylab.close()
> #plot figure 2
> pylab.figure()
> pylab.plot([1,2,3],[1,2,3])
> pylab.savefig("test_pylab2.png")
> pylab.close()
>
> Second figure cannot be created due to the crash.
>
> We need the help of matplotlib Gurus !
>
> Regards,
>
> David
>
>
>
> 2006年2月24日, David Huard <dav...@gm...>:
> > What if I want it to be non-interactive ? I want to save graphics
> > directly to a file from a script, without them being displayed. It's
> > weird that in this case close() crashes the whole thing.
> >
> > David
> >
> > 2006年2月23日, David TREMOUILLES < dav...@gm...>:
> > > Could that be because interactive mode was set in your previous insta=
ll
> ?
> > > (in .matplotlibrc file)
> > > if I try:
> > >
> > > #!/usr/bin/python
> > > import matplotlib
> > > import pylab
> > > matplotlib.interactive(True)
> > > pylab.figure()
> > > pylab.close()
> > >
> > > Then it works...
> > > Hope this could help you...
> > >
> > > David
> > >
> > > 2006年2月23日, Emmanuel Pecontal <pec...@ob...>:
> > >
> > > > Hello,
> > > >
> > > > If I run the following script on my system (SuSE 10.0, matplotlib-0=
.87
> but
> > > > same behaviour with 0.86):
> > > >
> > > > #!/usr/bin/python
> > > >
> > > > import matplotlib
> > > > import pylab
> > > >
> > > > pylab.figure()
> > > > pylab.close ()
> > > >
> > > > It crashes with the following error message:
> > > >
> > > > Traceback (most recent call last):
> > > > File "./tmp.py", line 8, in ?
> > > > pylab.close()
> > > > File
> > > "/usr/lib/python2.4/site-packages/matplotlib/pylab.py",
> > > line 768, in
> > > > close
> > > > else: _pylab_helpers.Gcf.destroy(figManager.num)
> > > > File
> > >
> "/usr/lib/python2.4/site-packages/matplotlib/_pylab_helpers.py",
> > > line
> > > > 28, in destroy
> > > > figManager.destroy()
> > > > File
> > >
> "/usr/lib/python2.4/site-packages/matplotlib/backends/backend_gtk.py",
> > > > line 456, in destroy
> > > > gtk.main_quit()
> > > > RuntimeError: called outside of a mainloop
> > > >
> > > > Does someone has an idea of the problem?
> > > > I didn't have this kind of problem with my SuSE 9.3 and
> matplotlib-0.85
> > > >
> > > > --
> > > > Emmanuel P=E9contal
> > > >
> > > > CRAL - Observatoire de Lyon
> > > > 9, Av. Charles Andre
> > > > F-69561 Saint Genis Laval Cedex
> > > >
> > > > tel (33) (0)4.78.86.83.76 - fax (33) (0)4.78.86.83.86
> > > >
> > > > email : pec...@ob...
> > > >
> > >
> > >
> >
>
>

 >>2006年2月24日, David Huard <dav...@gm... >:
 >>What if I want it to be non-interactive ? I want to save graphics
 >>directly to a file from a script, without them being displayed. It's
 >>weird that in this case close() crashes the whole thing.
Yes your wright the script have to work in non interactive mode. (I was jus=
t
trying to give an explanation to Emmanuel problem).
The following script should work (but it crashes like Emmanuel reported):
#!/usr/bin/python
import matplotlib
import pylab
#Turn off interactive mode
matplotlib.interactive(False)
#plot figure 1
pylab.figure()
pylab.plot([1,2,3],[1,2,3])
pylab.savefig("test_pylab1.png")
pylab.close()
#plot figure 2
pylab.figure()
pylab.plot([1,2,3],[1,2,3])
pylab.savefig("test_pylab2.png")
pylab.close()
Second figure cannot be created due to the crash.
We need the help of matplotlib Gurus !
Regards,
David
2006年2月24日, David Huard <dav...@gm...>:
>
> What if I want it to be non-interactive ? I want to save graphics
> directly to a file from a script, without them being displayed. It's
> weird that in this case close() crashes the whole thing.
>
> David
>
> 2006年2月23日, David TREMOUILLES < dav...@gm...>:
> > Could that be because interactive mode was set in your previous install
> ?
> > (in .matplotlibrc file)
> > if I try:
> >
> > #!/usr/bin/python
> > import matplotlib
> > import pylab
> > matplotlib.interactive(True)
> > pylab.figure()
> > pylab.close()
> >
> > Then it works...
> > Hope this could help you...
> >
> > David
> >
> > 2006年2月23日, Emmanuel Pecontal <pec...@ob...>:
> >
> > > Hello,
> > >
> > > If I run the following script on my system (SuSE 10.0, matplotlib-0.8=
7but
> > > same behaviour with 0.86):
> > >
> > > #!/usr/bin/python
> > >
> > > import matplotlib
> > > import pylab
> > >
> > > pylab.figure()
> > > pylab.close ()
> > >
> > > It crashes with the following error message:
> > >
> > > Traceback (most recent call last):
> > > File "./tmp.py", line 8, in ?
> > > pylab.close()
> > > File
> > "/usr/lib/python2.4/site-packages/matplotlib/pylab.py",
> > line 768, in
> > > close
> > > else: _pylab_helpers.Gcf.destroy(figManager.num)
> > > File
> > "/usr/lib/python2.4/site-packages/matplotlib/_pylab_helpers.py",
> > line
> > > 28, in destroy
> > > figManager.destroy()
> > > File
> > "/usr/lib/python2.4/site-packages/matplotlib/backends/backend_gtk.py",
> > > line 456, in destroy
> > > gtk.main_quit()
> > > RuntimeError: called outside of a mainloop
> > >
> > > Does someone has an idea of the problem?
> > > I didn't have this kind of problem with my SuSE 9.3 and
> matplotlib-0.85
> > >
> > > --
> > > Emmanuel P=E9contal
> > >
> > > CRAL - Observatoire de Lyon
> > > 9, Av. Charles Andre
> > > F-69561 Saint Genis Laval Cedex
> > >
> > > tel (33) (0)4.78.86.83.76 - fax (33) (0)4.78.86.83.86
> > >
> > > email : pec...@ob...
> > >
> >
> >
>

On Wed, Feb 22, 2006 at 09:24:38PM -0600, Albert Chin wrote:
> 
> The failure occurs when displaying the data. I've also tried
> anim_tk.py to determine if it was a problem with the GTK+ backend but
> anim_tk.py fails in the same way.
I did some more digging and the "terminate called after throwing an
instance ..." error is coming from G++. Maybe something funk with C++
exceptions on HP/PA that isn't triggered on HP/IA.
-- 
albert chin (ch...@th...)

On Wed, Feb 22, 2006 at 08:49:16PM -0600, John Hunter wrote:
> >>>>> "Albert" == Albert Chin <mat...@ml...> writes:
> 
> Albert> We've built 0.87 with Python 2.4.2/GCC 3.4.3 on Solaris,
> Albert> HP-UX, Tru64 UNIX, IRIX, and Redhat Linux. It works fine
> Albert> on all platforms except HP-UX/PA (HP-UX 11.23/IA64 works
> Albert> fine). Some of the examples, like two_scales.py, work fine
> Albert> if run once. However, with ~/.matplotlib/ttffont.cache
> Albert> existing, a second run of two_scales.py (or any other
> Albert> example), fails with the following: terminate called after
> Albert> throwing an instance of 'Py::RuntimeError'
> 
> Please clear the tex cache dir and then run the example both times
> with the --verbose-debug flag, and post the output along with the
> complete traceback.
~/tex.cache is empty.
 $ rm -rf ~/.matplotlib
 $ python two_scales.py --verbose-debug 2>&1 | tee /tmp/output.1
 $ python two_scales.py --verbose-debug 2>&1 | tee /tmp/output.2
The failure occurs when displaying the data. I've also tried
anim_tk.py to determine if it was a problem with the GTK+ backend but
anim_tk.py fails in the same way.
What do you mean by "traceback"? Stack trace from Python after the
error? None is given.
-- 
albert chin (ch...@th...)

>>>>> "Albert" == Albert Chin <mat...@ml...> writes:
 Albert> We've built 0.87 with Python 2.4.2/GCC 3.4.3 on Solaris,
 Albert> HP-UX, Tru64 UNIX, IRIX, and Redhat Linux. It works fine
 Albert> on all platforms except HP-UX/PA (HP-UX 11.23/IA64 works
 Albert> fine). Some of the examples, like two_scales.py, work fine
 Albert> if run once. However, with ~/.matplotlib/ttffont.cache
 Albert> existing, a second run of two_scales.py (or any other
 Albert> example), fails with the following: terminate called after
 Albert> throwing an instance of 'Py::RuntimeError'
Please clear the tex cache dir and then run the example both times
with the --verbose-debug flag, and post the output along with the
complete traceback.
JDH