matplotlib-users Mailing List for matplotlib (Page 3)

On 2005年7月27日, John Hunter wrote:
> Rich clearly has a very old examples dir if all his demos call
> matplotlib.matlab, but the zip file on the web *is* fairly current. I
> usually update the site with every release (0.82 was the last time -- I
> fell behind on 0.83) and the zip file is automatically populated with the
> current examples, as is
> http://matplotlib.sf.net/examples
 Thanks, John. Now I see the problem.
 The two examples I picked, test.py and wx_demo.py, are old. On the
above-referenced web page directory, I see this:
 test.py 28-Sep-2004 09:36
 wx_demo.py 20-Nov-2003 20:43
and they both call matplotlib.matlab.
 I'll try some others instead ... those with a date of 15 June 2005. These
old files are also in the zip file -0.82 on the main page of the site.
Rich
-- 
Dr. Richard B. Shepard, President | Author of "Quantifying Environmental
Applied Ecosystem Services, Inc. (TM) | Impact Assessments Using Fuzzy Logic"
<http://www.appl-ecosys.com> Voice: 503-667-4517 Fax: 503-667-8863

>>>>> "Robert" == Robert Kern <rk...@uc...> writes:
 >> Hmmm-m-m. All the demos call it. So, I'll just change them to
 >> call pylab and go on from there.
 Robert> Not in 0.83.1 they don't. Ignore the zip file on the front
 Robert> page of the website. It's old; the examples included with
 Robert> the source distribution are up-to-date.
Rich clearly has a very old examples dir if all his demos call
matplotlib.matlab, but the zip file on the web *is* fairly current. I
usually update the site with every release (0.82 was the last time --
I fell behind on 0.83) and the zip file is automatically populated
with the current examples, as is
 http://matplotlib.sf.net/examples
Also, it may help to point out that changes like this (moving from
matplotlib.matlab to matplotlib.pylab or simply pylab) are always
listed both at
 http://matplotlib.sf.net/whats_new.html and
 http://matplotlib.sf.net/API_CHANGES
JDH

On 2005年7月27日, Robert Kern wrote:
> Not in 0.83.1 they don't. Ignore the zip file on the front page of the
> website. It's old; the examples included with the source distribution are
> up-to-date.
 Ah! It was the .zip file I was using.
Thanks again,
Rich
-- 
Dr. Richard B. Shepard, President | Author of "Quantifying Environmental
Applied Ecosystem Services, Inc. (TM) | Impact Assessments Using Fuzzy Logic"
<http://www.appl-ecosys.com> Voice: 503-667-4517 Fax: 503-667-8863

Rich Shepard wrote:
> On 2005年7月27日, Robert Kern wrote:
> 
>> That module has been renamed pylab. That demo must be out of date; 
>> since I
>> do not see it in CVS, I presume it's gone now, and you don't have to 
>> worry
>> about it.
> 
> Hmmm-m-m. All the demos call it. So, I'll just change them to call pylab
> and go on from there.
Not in 0.83.1 they don't. Ignore the zip file on the front page of the 
website. It's old; the examples included with the source distribution 
are up-to-date.
-- 
Robert Kern
rk...@uc...
"In the fields of hell where the grass grows high
 Are the graves of dreams allowed to die."
 -- Richard Harter

On 2005年7月27日, Robert Kern wrote:
> That module has been renamed pylab. That demo must be out of date; since I
> do not see it in CVS, I presume it's gone now, and you don't have to worry
> about it.
 Hmmm-m-m. All the demos call it. So, I'll just change them to call pylab
and go on from there.
Many thanks,
Rich
-- 
Dr. Richard B. Shepard, President | Author of "Quantifying Environmental
Applied Ecosystem Services, Inc. (TM) | Impact Assessments Using Fuzzy Logic"
<http://www.appl-ecosys.com> Voice: 503-667-4517 Fax: 503-667-8863

Rich Shepard wrote:
> I just became aware of matplotlib and have installed it (-0.83.1) on my
> Slackware-10.1 workstations. When I tried to run some of the examples I see
> this error:
> 
> [rshepard@salmo ~/development/python/matplotlib/examples]$ python 
> wx_demo.py Traceback (most recent call last):
> File "wx_demo.py", line 3, in ?
> from matplotlib.matlab import *
> ImportError: No module named matlab
> 
> Sure enough, there's no matlab module. Now, I do have
> /usr/lib/python2.4/site-packages/matplotlib/mlab.py
> 
> Is this the current version of matlab?
> 
> I know I'll have questions, but I'm going to RTF(ine)M and tutorial 
> first.
That module has been renamed pylab. That demo must be out of date; since 
I do not see it in CVS, I presume it's gone now, and you don't have to 
worry about it.
-- 
Robert Kern
rk...@uc...
"In the fields of hell where the grass grows high
 Are the graves of dreams allowed to die."
 -- Richard Harter

 I just became aware of matplotlib and have installed it (-0.83.1) on my
Slackware-10.1 workstations. When I tried to run some of the examples I see
this error:
[rshepard@salmo ~/development/python/matplotlib/examples]$ python wx_demo.py 
Traceback (most recent call last):
 File "wx_demo.py", line 3, in ?
 from matplotlib.matlab import *
ImportError: No module named matlab
 Sure enough, there's no matlab module. Now, I do have
/usr/lib/python2.4/site-packages/matplotlib/mlab.py
 Is this the current version of matlab?
 I know I'll have questions, but I'm going to RTF(ine)M and tutorial first.
TIA,
Rich
-- 
Dr. Richard B. Shepard, President | Author of "Quantifying Environmental
Applied Ecosystem Services, Inc. (TM) | Impact Assessments Using Fuzzy Logic"
<http://www.appl-ecosys.com> Voice: 503-667-4517 Fax: 503-667-8863

Hi Guys,
 I found the font that was causing the problem by adding 
'verbose.report('CreateFontDict: %s' % (fpath), 'debug')' just
inside the 'for fpath in fontfiles:' inside the 
font_manager.createFontDict function. Here's the last part of the output:
CreateFontDict: C:\WINDOWS\Fonts\VIVALDII.TTF
CreateFontDict: c:\windows\fonts\lfax.ttf
CreateFontDict: C:\WINDOWS\Fonts\PALSCRI.TTF
CreateFontDict: c:\windows\fonts\orlando.ttf
Assertion failed: ob_refcnt == 0, file CXX\cxx_extensions.cxx, line 1031
This application has requested the Runtime to terminate it in an unusual 
way.
Please contact the application's support team for more information.
----------------------------------------------------------------------------
When I remove orlando.tff from c:\windows\fonts,円 matplotlib works fine. 
Thanks again for your help guys!
-Brandon King
Brandon King wrote:
> John Hunter wrote:
>
>>>>>>> "Carl" == Carl Dr Kleffner <cmk...@gm...> writes:
>>>>>>> 
>>>>>>
>>
>> Carl> This is not a mystery error. This error pops up on some bad
>> Carl> fonts when the FontManager creates the dictionary of TTF
>> Carl> based fonts. An example is the CSD Font that installs with
>> Carl> jgrasp (jgrasp.org). 
>> That is very helpful -- if one of you can send me a problematic font
>> file offlist I will try and replicate and fix this bug. By running in
>> --verbose-debug, you should be able to ascertain which file is causing
>> the grief.
>> 
>>
> Here's the verbose-debug output.
>
> C:\proj\compclust\win32\compclustshell>C:\Python23\python.exe 
> matplotlibtest.py
> --verbose-debug
> matplotlib data path C:\Python23\share\matplotlib
> $HOME=C:\Documents and Settings\King
> CONFIGDIR=C:\Documents and Settings\King\.matplotlib
> loaded rc file C:\Python23\share\matplotlib\matplotlibrc
> matplotlib version 0.83.1
> verbose.level debug
> interactive is False
> platform is win32
> loaded modules: ['pylab', '__future__', 'copy_reg', 'sre_compile', 
> 'distutils',
> 'locale', '_sre', '__main__', 'site', '__builtin__', 'datetime', 
> 'matplotlib.tem
> pfile', 'encodings', 'os.path', 'encodings.encodings', 
> 'sre_constants', 'distuti
> ls.string', 'dateutil', 'matplotlib.datetime', 'strop', '_random', 
> 'tempfile', '
> errno', 'matplotlib.warnings', 'encodings.codecs', 'matplotlib.sys', 
> 're', 'ntpa
> th', 'pytz.sys', 'UserDict', 'distutils.sysconfig', 
> 'encodings.exceptions', 'nt'
> , 'pytz.sets', 'math', 'stat', 'zipimport', 'string', 'warnings', 
> 'encodings.typ
> es', '_codecs', 'sets', 'distutils.os', 'matplotlib', 
> 'encodings.cp1252', 'sys',
> 'pytz.tzinfo', 'pytz', 'pytz.datetime', 'matplotlib.__future__', 
> 'codecs', 'dis
> tutils.re', 'matplotlib.pytz', 'types', 'matplotlib.dateutil', 
> '_locale', 'matpl
> otlib.os', 'thread', 'sre', 'bisect', 'matplotlib.distutils', 
> 'signal', 'distuti
> ls.errors', 'random', 'linecache', 'itertools', 'time', 'exceptions', 
> 'sre_parse
> ', 'pytz.bisect', 'distutils.sys', 'os']
> numerix Numeric 23.8
> font search path ['C:\\Python23\\share\\matplotlib']
> trying fontname C:\Python23\share\matplotlib\cmex10.ttf
> trying fontname C:\Python23\share\matplotlib\cmmi10.ttf
> trying fontname C:\Python23\share\matplotlib\cmr10.ttf
> trying fontname C:\Python23\share\matplotlib\VeraMono.ttf
> trying fontname C:\Python23\share\matplotlib\cmsy10.ttf
> trying fontname C:\Python23\share\matplotlib\VeraSeBd.ttf
> trying fontname C:\Python23\share\matplotlib\cmtt10.ttf
> trying fontname C:\Python23\share\matplotlib\Vera.ttf
> Assertion failed: ob_refcnt == 0, file CXX\cxx_extensions.cxx, line 1031
>
> This application has requested the Runtime to terminate it in an 
> unusual way.
> Please contact the application's support team for more information.
>
> -Brandon King
>
> P.S. Thanks guys for helping with this!
>
>
>
>
> -------------------------------------------------------
> SF.Net email is Sponsored by the Better Software Conference & EXPO 
> September
> 19-22, 2005 * San Francisco, CA * Development Lifecycle Practices
> Agile & Plan-Driven Development * Managing Projects & Teams * Testing 
> & QA
> Security * Process Improvement & Measurement * http://www.sqe.com/bsce5sf
> _______________________________________________
> Matplotlib-users mailing list
> Mat...@li...
> https://lists.sourceforge.net/lists/listinfo/matplotlib-users
>
>

John Hunter wrote:
>>>>>>"Carl" == Carl Dr Kleffner <cmk...@gm...> writes:
>>>>>> 
>>>>>>
>
> Carl> This is not a mystery error. This error pops up on some bad
> Carl> fonts when the FontManager creates the dictionary of TTF
> Carl> based fonts. An example is the CSD Font that installs with
> Carl> jgrasp (jgrasp.org). 
>
>That is very helpful -- if one of you can send me a problematic font
>file offlist I will try and replicate and fix this bug. By running in
>--verbose-debug, you should be able to ascertain which file is causing
>the grief.
> 
>
Here's the verbose-debug output.
C:\proj\compclust\win32\compclustshell>C:\Python23\python.exe 
matplotlibtest.py
--verbose-debug
matplotlib data path C:\Python23\share\matplotlib
$HOME=C:\Documents and Settings\King
CONFIGDIR=C:\Documents and Settings\King\.matplotlib
loaded rc file C:\Python23\share\matplotlib\matplotlibrc
matplotlib version 0.83.1
verbose.level debug
interactive is False
platform is win32
loaded modules: ['pylab', '__future__', 'copy_reg', 'sre_compile', 
'distutils',
'locale', '_sre', '__main__', 'site', '__builtin__', 'datetime', 
'matplotlib.tem
pfile', 'encodings', 'os.path', 'encodings.encodings', 'sre_constants', 
'distuti
ls.string', 'dateutil', 'matplotlib.datetime', 'strop', '_random', 
'tempfile', '
errno', 'matplotlib.warnings', 'encodings.codecs', 'matplotlib.sys', 
're', 'ntpa
th', 'pytz.sys', 'UserDict', 'distutils.sysconfig', 
'encodings.exceptions', 'nt'
, 'pytz.sets', 'math', 'stat', 'zipimport', 'string', 'warnings', 
'encodings.typ
es', '_codecs', 'sets', 'distutils.os', 'matplotlib', 
'encodings.cp1252', 'sys',
 'pytz.tzinfo', 'pytz', 'pytz.datetime', 'matplotlib.__future__', 
'codecs', 'dis
tutils.re', 'matplotlib.pytz', 'types', 'matplotlib.dateutil', 
'_locale', 'matpl
otlib.os', 'thread', 'sre', 'bisect', 'matplotlib.distutils', 'signal', 
'distuti
ls.errors', 'random', 'linecache', 'itertools', 'time', 'exceptions', 
'sre_parse
', 'pytz.bisect', 'distutils.sys', 'os']
numerix Numeric 23.8
font search path ['C:\\Python23\\share\\matplotlib']
trying fontname C:\Python23\share\matplotlib\cmex10.ttf
trying fontname C:\Python23\share\matplotlib\cmmi10.ttf
trying fontname C:\Python23\share\matplotlib\cmr10.ttf
trying fontname C:\Python23\share\matplotlib\VeraMono.ttf
trying fontname C:\Python23\share\matplotlib\cmsy10.ttf
trying fontname C:\Python23\share\matplotlib\VeraSeBd.ttf
trying fontname C:\Python23\share\matplotlib\cmtt10.ttf
trying fontname C:\Python23\share\matplotlib\Vera.ttf
Assertion failed: ob_refcnt == 0, file CXX\cxx_extensions.cxx, line 1031
This application has requested the Runtime to terminate it in an unusual 
way.
Please contact the application's support team for more information.
-Brandon King
P.S. Thanks guys for helping with this!

Hello Steve,
Steve Schmerler schrieb:
> Hi
>
> I run into problems using axhline() (mpl 0.81) when setting the 'xmin'
> and 'xmax' keywords. The horizontal line to be drawn doesn't show up or
> just a part of it. What's up (see attached script and .eps files)?
>
you misunderstood the meaning of the 'xmin' and 'xmax' keyword 
arguments. If you look at the documtation for axhline, you will see 
that a value of 0.0 refers to the left side of the figure (well, 
actually the axes), and 1.0 refers to the very right. So the default 
setting xmin=0, xmax=1 will draw a horizontal line from the very left to 
the very right. 
With this information, your other three examples should be clear.
Best regards,
Niklas.

Hello Rameshwari!
Vedpathak, Rameshwari IN BLR SISL schrieb:
> Hello,
>
> The signals in matplotlib functionality are plotted by joining 2 points.
> So if the signal values are say
> y-axis: 0,1,2,3,4...
> X-axis (time): 0, 10, 20, 30, 40....
>
> then it draws a line which joins these points and the signal that we 
> see is a straight line ramping in upward direction.
>
> But I want to draw the signal such that it should hold the signal 
> value till the next point so that the signal now seen is like a step 
> ramp signal.
>
> So for the above example till time = 10 the value should be 0 and 
> again from time = 10 to time =20 value should be 1.
>
> Does anyone have an idea how to draw such signals with matplotlib 
> functionality)
>
Matplotlib indeed has this functionality :-). Try:
x = [0,10,20,30,40]
y = [0,1,2,3,4]
plot( x, y, linestyle="steps" )
Best regards,
Niklas.

Hello Dave,
Dave schrieb:
> I was trying to track down a ylim related freeze problem in pylab 
> 0.81. Not much luck so far, but I did run across another issue. Can 
> someone try this and see if it displays sensible results in the latest 
> versions?
> 
> import pylab as p
> p.cla()
> p.plot([0,1.6], [10000, -0.05])
> p.plot([0,2], [-.050, 10000])
> p.xlim((-.1, 1.8))
> p.ylim((-.1, 0.1))
> p.show()
> 
> For me using TkAgg / python 2.3 I see a large filled triangle from the 
> second plot command and nothing from the first. Admittedly this is 
> usually not a practial issue but I'm wondering if it's somehow related 
> to the freeze I sometimes see when plotting data with a few points far 
> outside of the present axis scale.
> 
> -- David
Regarding the freeze: I had a similar problem when I implemented my own 
zoom method about a week ago. I don't really know the reason for this 
behaviour -- my computer would freeze completely once I zoomed in to 
much (a very stable Slackware machine... sniff) .
I tried out the new matplotlib 0.83.1 and now everything is fine again.
Regards,
Niklas.

Hello Philippe,
phi...@ho... schrieb:
> Hi list,
>
> I'm using matplotlib oo to monitor figure.
> I successfully add a legen with :
> self.fig = Figure(figsize=(self.hSize,self.vSize), dpi=self.res)
> self.a = self.fig.add_subplot(111)
> self.a.legend(self.labels,loc='best')
>
> How can i delete this legend?
>
A quick glance at the corresponding source code (method 'legend' in 
'axes.py') shows that the legend is set as axes.legend_.
To remove a legend, you therefore only need to do
self.a.legend_ = None
and redraw the figure.
Regards,
Niklas.

Hi list,
I'm using matplotlib oo to monitor figure.
I successfully add a legen with :
self.fig = Figure(figsize=(self.hSize,self.vSize), dpi=self.res)
self.a = self.fig.add_subplot(111)
self.a.legend(self.labels,loc='best')
How can i delete this legend?
Thanks a lot,
Philippe Collet

Hi
I run into problems using axhline() (mpl 0.81) when setting the 'xmin'
and 'xmax' keywords. The horizontal line to be drawn doesn't show up or
just a part of it. What's up (see attached script and .eps files)?
cheers,
steve

>>>>> "Carl" == Carl Dr Kleffner <cmk...@gm...> writes:
 Carl> This is not a mystery error. This error pops up on some bad
 Carl> fonts when the FontManager creates the dictionary of TTF
 Carl> based fonts. An example is the CSD Font that installs with
 Carl> jgrasp (jgrasp.org). 
That is very helpful -- if one of you can send me a problematic font
file offlist I will try and replicate and fix this bug. By running in
--verbose-debug, you should be able to ascertain which file is causing
the grief.
 Carl> BTW: As the bakoma fonts are free now, is this ttf cache
 Carl> needed by default anymore? The CTAN bakoma.zip includes now
 Carl> 140 TTF fonts.
I don't see how license of the bakoma fonts and the ttf cache are
related. The cache is used to store a dictionary of which fonts were
found on your system so that it doesn't have to search the system
every time. I do not want to configure mpl to use only the bakoma
fonts since I haven't been too impressed with the cm*10*.ttf fonts I
used with mathtext. For example, last I checked they did not appear
to have kerning information, and the letter stroke thicknesses appear
uneven to me in some cases. Perhaps the others are better, or I am
doing something wrong.
But in any case, thanks for the heads up on the license change. I
have update mpl CVS with the new fonts and license info.
JDH

I was trying to track down a ylim related freeze problem in pylab 0.81. Not=
=20
much luck so far, but I did run across another issue. Can someone try this=
=20
and see if it displays sensible results in the latest versions?
 import pylab as p
p.cla()
p.plot([0,1.6], [10000, -0.05])
p.plot([0,2], [-.050, 10000])
p.xlim((-.1, 1.8))
p.ylim((-.1, 0.1))
p.show()=20
For me using TkAgg / python 2.3 I see a large filled triangle from the=20
second plot command and nothing from the first. Admittedly this is usually=
=20
not a practial issue but I'm wondering if it's somehow related to the freez=
e=20
I sometimes see when plotting data with a few points far outside of the=20
present axis scale.
 -- David

On 2005年7月27日, Rameshwari IN SISL Vedpathak apparently wrote:=20
> y-axis: 0,1,2,3,4... X-axis (time): 0, 10, 20, 30,=20
> 40....=20
> So for the above example till time =3D 10 the value should=20
> be 0 and again from time =3D 10 to time =3D20 value should be=20
> 1.=20
> Does anyone have an idea how to draw such signals with matplotlib=20
> functionality=20
Someone else may tell you this functionality is already=20
built into Matplotlib. If not, you can just repeat values.
I do not know what the best way to do this is, but the
following should work. (You'll probably want to fiddle with=20
the axes.)
Alan Isaac
def step_pts(x,y):
=09pts=3Dzip(x,y)
=09inter =3D zip(x[1:],y[:-1])
=09z =3D zip(*[j for i in map(None,pts,inter) for j in i][:-1])
=09return z[0],z[1]
x =3D [0.,10.,20.,30.]
y =3D [1.,2.,3.,4.]
from pylab import *
plot(*step_pts(x,y))
show()

Hi
I run into problems using axhline() (mpl 0.81) when setting the 'xmin'
and 'xmax' keywords. The horizontal line to be drawn doesn't show up or
just a part of it. What's up (see attached script and .eps files)?
cheers,
steve

Hello,
The signals in matplotlib functionality are plotted by joining 2 points.
So if the signal values are say=20
y-axis: 0,1,2,3,4...
X-axis (time): 0, 10, 20, 30, 40....
then it draws a line which joins these points and the signal that we see
is a straight line ramping in upward direction.
But I want to draw the signal such that it should hold the signal value
till the next point so that the signal now seen is like a step ramp
signal.
So for the above example till time =3D 10 the value should be 0 and =
again
from time =3D 10 to time =3D20 value should be 1.
Does anyone have an idea how to draw such signals with matplotlib
functionality
Best Regards,
Rameshwari

> --- Ursprüngliche Nachricht ---
> Von: John Hunter <jdh...@ac...>
> An: Brandon King <ki...@ca...>
> Kopie: mat...@li...
> Betreff: Re: [Matplotlib-users] Return of the mystery error.
> Datum: 2005年7月26日 21:10:45 -0500
> 
> >>>>> "Brandon" == Brandon King <ki...@ca...> writes:
> 
> Brandon> Hi John, Thanks again for the quick responce. Sorry about
> Brandon> not doing the simple test earlier. I installed the latest
> Brandon> version this time and used the script you suggested (.py
> Brandon> file is attached). Here's the output:
> 
> OK, damn, this is indeed the mystery error. It has reared its ugly
> head on various (three, as far as I can see) win32 machines and not
> others and noone has figured out why. Unfortunately, I can't
> reproduce it and don't know how to fix it. Please post as much
> information about your system (exact OS, which service pack,
> motherboard/cpu, where you got python from, etc) as possible because
> it may be helpful down the road.
> 
> I think you may be failing in the font load section. Could you rerun
> your test script one more time with --verbose-debug and report that in
> addition to the information above?
> 
> Thanks,
> JDH
> 
This is not a mystery error. This error pops up on some bad fonts
when the FontManager creates the dictionary of TTF based fonts.
An example is the CSD Font that installs with jgrasp (jgrasp.org).
BTW: As the bakoma fonts are free now, is this ttf cache needed 
by default anymore? The CTAN bakoma.zip includes now 140 TTF fonts.
Regards
Carl
-- 
GMX DSL = Maximale Leistung zum minimalen Preis!
2000 MB nur 2,99, Flatrate ab 4,99 Euro/Monat: http://www.gmx.net/de/go/dsl

On Tuesday 26 July 2005 10:14 pm, John Hunter wrote:
> >>>>> "Darren" == Darren Dale <dd...@co...> writes:
>
> Darren> On Tuesday 26 July 2005 01:11 pm, Mark Bakker wrote:
> >> I have the same experience that eps files seem to be much
> >> larger than you expect, or than produced by other programs for
> >> similar plots. Is it maybe because fonts are embedded in the
> >> eps file? Anybody know?
>
> Darren> It is because the fonts are embedded in the eps file.
>
> As Darren notes,, it is the font embedding. If this bothers you, you
> have two options. Set one of the following to True in your matplotlib
> rc file
>
> # use of afm fonts -- breaks mathtext but results in small files
> ps.useafm : False
>
> # Experimental: use ghostscript to distill ps output - may yield smaller
> files ps.usedistiller : False
>
> For the latter you will need a recent version of ghostscript.
Another option you might try, which is not built in to matplotlib, is to 
create a pdf using ps2pdf (included with ghostscript), and then convert back 
using pdftops (included with xpdf). I know it seems dumb, but if you use 
mpl's tex/latex support, this is the best method for cleaning the files up so 
they can be embedded in another document.
-- 
Darren

>>>>> "Darren" == Darren Dale <dd...@co...> writes:
 Darren> On Tuesday 26 July 2005 01:11 pm, Mark Bakker wrote:
 >> I have the same experience that eps files seem to be much
 >> larger than you expect, or than produced by other programs for
 >> similar plots. Is it maybe because fonts are embedded in the
 >> eps file? Anybody know?
 Darren> It is because the fonts are embedded in the eps file.
As Darren notes,, it is the font embedding. If this bothers you, you
have two options. Set one of the following to True in your matplotlib
rc file
 # use of afm fonts -- breaks mathtext but results in small files
 ps.useafm : False 
 
 # Experimental: use ghostscript to distill ps output - may yield smaller files
 ps.usedistiller : False 
For the latter you will need a recent version of ghostscript.
Hope this helps,
JDH

>>>>> "Brandon" == Brandon King <ki...@ca...> writes:
 Brandon> Hi John, Thanks again for the quick responce. Sorry about
 Brandon> not doing the simple test earlier. I installed the latest
 Brandon> version this time and used the script you suggested (.py
 Brandon> file is attached). Here's the output:
OK, damn, this is indeed the mystery error. It has reared its ugly
head on various (three, as far as I can see) win32 machines and not
others and noone has figured out why. Unfortunately, I can't
reproduce it and don't know how to fix it. Please post as much
information about your system (exact OS, which service pack,
motherboard/cpu, where you got python from, etc) as possible because
it may be helpful down the road.
I think you may be failing in the font load section. Could you rerun
your test script one more time with --verbose-debug and report that in
addition to the information above?
Thanks,
JDH

OK, so I modified the toolbar (by hacking the toolbar2 class in the TK back=
end).
But I can just make that a new class and call it MyToolbar.
How can I now conveniently have matplotlib use this toolbar?
In the rc file you can define 'classic' or 'toolbar2', but not
'customtoolbar' or something like that.
I looked at the 'embedded_in_tk' example. That would work, but then I
lose a lot of other functionality. Is it possible to just set a flag
and have matplotlib read my toolbar (asking too much maybe?).
Thanks,
Mark