matplotlib-users Mailing List for matplotlib (Page 3)

Hi,
On Thu, Feb 28, 2008 at 1:47 PM, John Hunter <jd...@gm...> wrote:
>
> If I understand what you are trying to do, you should be using the
> sharex and sharey settings for ax1
Mmm, I don't think so. The extent of one image and other are in different
units, so I need to transform whatever value it is I have in one image to
the corresponding window in the other image. I don't know whether there's an
explicit way to let matplotlib know of a transformation function for this.
Not sure (it should) but I would use fig.canvas.draw() here in any case.
>
Your suggestion works fine! Thanks!
Jose
-- 
Centre for Terrestrial Carbon Dynamics
Department of Geography, University College London
Gower Street, London WC1E 6BT, UK

On Wed, Feb 27, 2008 at 12:04 PM, chombee <ch...@ne...> wrote:
> Is there any way to stop some of my text labels from overlapping?
For the ones that are tightly clustered, you could use the annotate
command to move the text farther away from the associated point.
Annotate has support for drawing an arrow from the text label to the
annotated point. But it won't do auto-layout -- you'll still have to
choose the coordinates of the text label so they don't overlap.
JDH

On Thu, Feb 28, 2008 at 6:55 AM, Jose Gomez-Dans <jgo...@gm...> wrote:
> Hi!
> I have two datasets with different data coordinates, which I want to plot on
> different subplots, all showing the same extent. I want to be able to zoom
> into one of the subplots, and have the other follow (through some conversion
> of the former's extent).
> Initially, I just want one plot to follow the other (but I could easily
> extend that to both plots if needs be). The code works, but I cannot update
> my plots automatically, although I call pylab.draw(). What's the way to
> have my plots respond to my events without dragging the window borders,
> resizing the window, etc? I am using matplotlib 0.90.1 and GTKAgg backend on
> Linux. Here's some code snippet of what I am doing:
If I understand what you are trying to do, you should be using the
sharex and sharey settings for ax1
 ax1 = fig.add_subplot(211)
 ax2 = fig.add_subplot(212, sharex=ax1, sharey=ax1)
then the xlim and ylim will be shared, so panning/zooming navigation
in either will affect both.
> pylab.draw() #Why doesn't the window refresh when it gets here? I
Not sure (it should) but I would use fig.canvas.draw() here in any case.
JDH

> How do I get rid of the redundant ticks on the top and right edges?
pylab.gca().get_xaxis().set_ticks_postion('bottom')
same for yaxis
> Is there any way to stop some of my text labels from overlapping?
don't know
> Why is there such a big gap between the plot itself and the axes ticks?
did you try pylab.axis('tight')
Bernhard

Hi!
I have two datasets with different data coordinates, which I want to plot on
different subplots, all showing the same extent. I want to be able to zoom
into one of the subplots, and have the other follow (through some conversion
of the former's extent).
Initially, I just want one plot to follow the other (but I could easily
extend that to both plots if needs be). The code works, but I cannot update
my plots automatically, although I call pylab.draw(). What's the way to
have my plots respond to my events without dragging the window borders,
resizing the window, etc? I am using matplotlib 0.90.1 and GTKAgg backend on
Linux. Here's some code snippet of what I am doing:
fig=pylab.figure()
ax =fig.add_subplot(211)
xtent1=(1111950.520,3335851.559,-2223901.039,-1111950.520)
xtent2=(0,2477,0,2319 )
im1=ax.imshow( image1,extent=xtent1,picker=True)
ax2=fig.add_subplot(212)
im2=ax2.imshow( image2)
ax2.set_xlim (1073, 1797)
ax2.set_ylim (1944, 1588)
pylab.draw()
def pick_new_extent(event):
 artist = event.artist
 (xlow,xhigh) = artist.get_axes().get_xlim()
 (ylow,yhigh) = artist.get_axes().get_ylim()
 [xmin, xmax, ymin, ymax] = CalculateWindowCoordinates(xlow, xhigh,
ylow, yhigh) #This does the coordinate transformation
 ax2.set_xlim(xmin,xmax)
 ax2.set_ylim(ymax, ymin)
 pylab.draw() #Why doesn't the window refresh when it gets here? I
need to change the window size to get it to refresh!!
fig.canvas.mpl_connect('pick_event', pick_new_exent)
Thanks!
Jose
-- 
Centre for Terrestrial Carbon Dynamics
Department of Geography, University College London
Gower Street, London WC1E 6BT, UK

On Thu, Feb 28, 2008 at 9:33 AM, darckense <dar...@fr...> wrote:
> Hello all,
>
> When you create a graph with a legend using legend(), is there a way to remove
> the black border drawn around the legend ?
I asked the same question some time ago. The trick is setting the
linewidth to zero:
from pylab import *
....
lg = legend()
lg.get_frame().set_linewidth(0)
Maybe there are other ways, but this one works for me.
Cheers,
YVES
> -------------------------------------------------------------------------
> This SF.net email is sponsored by: Microsoft
> Defy all challenges. Microsoft(R) Visual Studio 2008.
> http://clk.atdmt.com/MRT/go/vse0120000070mrt/direct/01/
> _______________________________________________
> Matplotlib-users mailing list
> Mat...@li...
> https://lists.sourceforge.net/lists/listinfo/matplotlib-users
>

Jeff,
yes it works, the unique constraint is to put the image when we create the 
basemap instance.
Le jeudi 28 février 2008, Jeff Whitaker a écrit :
> Lionel Roubeyrie wrote:
> > Hi Jeff,
> > yes I see this method, but set_xlim and set_ylim use axes coordinates,
> > and I would like to use geographic coordinates.
> > It would be great if we could put an image in a map directly by
> > specifying its geographic coordinates
>
> Lionel: There's no way to do that - but it's pretty easy to convert to
> geographic to map (axes) coordinates using the Basemap instance __call__
> method. So, if m is the Basemap instance, you can do something like
>
> x1,y1 = m(lon1,lon1)
> x2,y2 = m(lon2,lat2)
> ax.set_ylim(y1,y2)
> ax.set_xlim(x1,x2)
>
> -Jeff
>
> > Le mardi 26 février 2008, Jeff Whitaker a écrit :
> >> Lionel Roubeyrie wrote:
> >>> Hi all,
> >>> I'm sure it's a trivial question, but can't find any valid answer in
> >>> the basemap examples directory or with google : I have a georeferenced
> >>> TIFF file in 'lcc' projection, representing a little portion of France,
> >>> and I need to put it on a map, resets map limits to a closed portion of
> >>> the map, and put contours on it. I see how to initialise Basemap with
> >>> the limits of the raster, but not how to "make a zoom" in the map by
> >>> specifing coordinates. Thanks for your help
> >>
> >> Lionel: You can use the set_xlim and set_ylim axes methods to manually
> >> 'zoom' the plot. It has to be done after all the other plotting though,
> >> since all the Basemap methods (contour, imshow etc) will reset the axes
> >> limits to show the entire map region.
> >>
> >> -Jeff
-- 
Lionel Roubeyrie - lro...@li...
Chargé d'études et de maintenance
LIMAIR - la Surveillance de l'Air en Limousin
http://www.limair.asso.fr

Hello all,
When you create a graph with a legend using legend(), is there a way to remove 
the black border drawn around the legend ?
Thanks.

Thanks for you reply Eric.
Since I am a beginner in matplotlib, I don't want to blame the soft 
first. I mean it could be a bug, but I have the feeling my code for 
masking the missing data is useless.
My real question is : How to mask data that do not exist ??? In the 
masked_demo.py example, it is show how to mask data from a set like "My 
data are
x=[1,2,3,4,5,6,7,8,9,10]
y = sin(x) or anyelse continuous fonction and I want to mask 1,2,3 so I 
write : ym = ma.masked_where(y <3, y)
But in my case, it is more like :
x = [1,2,3,4,7,8,9,15,16,17]
y = discrete values (no continuous fonction !) like 
[5,2,20,17,3,12,9,18,18,9]
What I want to do with these data :
1/plot graph with no line connecting the point (x,y) = (4,17) and the 
point (7,3), etc.
2/fill the space under the plot but do not fill the space where there is 
no data (eg between the same 2 points).
The 1st thing seems to be easy = just follow the example masked_demo
But in my case, how can I mask data that don't exist ? I can't write ym 
= ma.masked_where(y <3, y)
Hum, I hope I was clearer....
Michael
Eric Firing a écrit :
> You have run into a bug in the combination of poly_between and 
> fill--maybe only the former, which not taking masked arrays into 
> account. I have not looked at it enough to know whether it will be 
> easy or hard to fix, but it certainly should be fixed. I can't look 
> at it more right now, unfortunately.
>
> Eric
>
> Michaël Douchin wrote:
>> Hi list
>>
>> I searched the list and google, but couldn't find a way to solve my pbm.
>>
>> I have data stored in a list (from an sql query) , with these "columns":
>> x = time serie in hours
>> y = some level value
>>
>> There are some missing values : eg between 08:33 and 08:40.
>> Here is my code :
>> sqla="SELECT * FROM import_parcelle a WHERE dat_loc='" + 
>> date_traite + "' AND code_uc='" + code_uc + "' ORDER BY a.heu_loc ;"
>> resa=db.query(sqla)
>> data=resa.dictresult() #x = time serie
>> x= [ datetime.datetime(*time.strptime(a["dat_loc"]+" 
>> "+a["heu_loc"],'%Y-%m-%d %H:%M:%S')[0:6] ) for a in data]
>> #y5 = tank level
>> y5= [float(a["niv_cuv"]) for a in data] # --> extraction de la 
>> colonne y1
>> figure()
>> # the plot command with no mask
>> 
>> plot_date(x,y5,color='b',linestyle='None',marker='',xdate=True,ydate=False) 
>>
>> #the mask
>> #ym5 = ma.masked_where(y5 <300, y5)
>> #plot_date(x,ym5,color='r',linestyle='-',xdate=True,ydate=False)
>> #the filling under the curve
>> xs, ys = poly_between(x, 0, y5)
>> fill(xs,ys)
>>
>>
>> Here is the result:
>> http://michaeldouchin.free.fr/17_2007年06月07日_10B_vitesse.png
>> As you see, I commented the lines with the mask, because it did not 
>> change anything
>>
>> To see what I am looking for, here is the result under R (a 
>> statistical tool)
>> http://michaeldouchin.free.fr/17_2007年06月07日_10B_vitesse.jpg
>> As you see, between 08:48 and 08:50 (for example), there is a gap, 
>> showing we have no data for this interval.
>>
>> As I want to automatically draw this graph for different set of data, 
>> I can't look each set in detail.
>> I tried to folow the example masked_demo.py, but I could not adapt it 
>> to my case....
>>
>> Any hint ?
>> Thanks very much in advance
>>
>> Michael
>>
>> ------------------------------------------------------------------------- 
>>
>> This SF.net email is sponsored by: Microsoft
>> Defy all challenges. Microsoft(R) Visual Studio 2008.
>> http://clk.atdmt.com/MRT/go/vse0120000070mrt/direct/01/
>> _______________________________________________
>> Matplotlib-users mailing list
>> Mat...@li...
>> https://lists.sourceforge.net/lists/listinfo/matplotlib-users
>
>
>

Out of the country. I am cc'ing the matplotlib list.
On Wed, Feb 27, 2008 at 3:08 PM, Luis Carlos Garelli <
lga...@us...> wrote:
>
> Message body follows:
>
> Greetings, first, sorry to bother you... I am working using
> the matplotlib, and i am having some strange problems with
> my application. When i execute in some pcs, there is no
> problem at all, but in others pcs, i receive this message:
>
> File "pylab.pyc", line 1 in ?
> File "matplotlib\pylab.pyc", line 203 in ?
> File "matplotlib\axes.pyc", line 14 in ?
> File "matplotlib\agg.pyc", line 5 in ?
> File "matplotlib\_agg.pyc", line 12 in ?
> File "matplotlib\_agg.pyc", line 10 in ?
> ImportError: Dll Load Failed
>
> All the pc's have Windows Xp, most of them Pentium 4, some
> are Pentium D. I don't know what may be the issue. All those
> pcs have the same software, just different hardware, i hope
> you could help me with this little problem. Thanks in advance...
>
> --
> This message has been sent to you, a registered SourceForge.net user,
> by another site user, through the SourceForge.net site. This message
> has been delivered to your SourceForge.net mail alias. You may reply
> to this message using the "Reply" feature of your email client, or
> using the messaging facility of SourceForge.net at:
> https://sourceforge.net/sendmessage.php?touser=2021607
>
>

Lionel Roubeyrie wrote:
> Hi Jeff,
> yes I see this method, but set_xlim and set_ylim use axes coordinates, and I 
> would like to use geographic coordinates.
> It would be great if we could put an image in a map directly by specifying its 
> geographic coordinates
> 
Lionel: There's no way to do that - but it's pretty easy to convert to 
geographic to map (axes) coordinates using the Basemap instance __call__ 
method. So, if m is the Basemap instance, you can do something like
x1,y1 = m(lon1,lon1)
x2,y2 = m(lon2,lat2)
ax.set_ylim(y1,y2)
ax.set_xlim(x1,x2)
-Jeff
> Le mardi 26 février 2008, Jeff Whitaker a écrit :
> 
>> Lionel Roubeyrie wrote:
>> 
>>> Hi all,
>>> I'm sure it's a trivial question, but can't find any valid answer in the
>>> basemap examples directory or with google : I have a georeferenced TIFF
>>> file in 'lcc' projection, representing a little portion of France, and I
>>> need to put it on a map, resets map limits to a closed portion of the
>>> map, and put contours on it. I see how to initialise Basemap with the
>>> limits of the raster, but not how to "make a zoom" in the map by
>>> specifing coordinates. Thanks for your help
>>> 
>> Lionel: You can use the set_xlim and set_ylim axes methods to manually
>> 'zoom' the plot. It has to be done after all the other plotting though,
>> since all the Basemap methods (contour, imshow etc) will reset the axes
>> limits to show the entire map region.
>>
>> -Jeff
>> 
>
>
>
> 
-- 
Jeffrey S. Whitaker Phone : (303)497-6313
NOAA/OAR/CDC R/PSD1 FAX : (303)497-6449
325 Broadway Boulder, CO, USA 80305-3328

All: I'm using the text() function to place city labels on a map 
with Basemap. However, many of these cities are so close together 
that the labels overlap when they're placed on the map.
Is there some way to determine the bounding box for a given text 
object _before_ placing it on the map, so that I can write an 
algorithm to detect overlaps?
Alternately, if someone else has some clever way of placing non- 
conflicting labels on maps, I'd love to hear about it!
Thanks,
Mike 

> I have had a similar problem here with python-2.5.1, numpy-1.0.4 and
> mpl-0.91.2 all installed on my linux (from scratch) system. I reported
> this last year but ending in a dead end with no one being able to help.
Okay, that's not very encouraging.
I would have thought that a segmentation fault would have attracted more
attention and more answer. besides, i didn't see any issue in the bug
tracker about that. Maybe I should add one report if I find the time.
My --verbose-debug_annoying output seemed to indicate it had something
> to do with the loading of the fonts, like yours, but that's all the
> farther we all got. My work around was to install the prerequisites for
> GtkAgg and go with that. Everything works fine now using GtkAgg.
Unfortunately, I have to use Tkagg : I need to add some graphs in a Tk-based
application...
I even recently installed ATLAS using lapack so I could install numpy
> and scipy with their full functionality and was also successful doing
> that. I'm able to do curve fitting and find roots of polynomials.
Installing a full numpy-scipy-matplotlib environment is my second goal.
I had some pb with odr and ndimage, which caused the scipy test suite to
fail, but now it works.
installing matplotlib is the last barrier.
So I
> believe the TkAgg backend does have some incompatibilities but no one
> has a clue as to why. Otherwise, it's all smooth sailing doing it all
> from scratch.
Here are my last findings :
1) I runned my installations script on another computer, a linux 64 bit
with Red Hat linux 3, and I had the same pb.
So this bug seems reproductible and do not depend on the architecture or on
the distribution.
There's nothing really extraordinary in these script, but I can send you
them if you want.
2) I tried to compile all these programs in debug mode. I added the options
 --enable-symbols when building tcl, tk and python
 --with-pydebug when compiling python
 --debug when builkding matplotlib.
This doesn't give more information when running `python2.5
test.py--verbose-debug_annoying`, but the backtrace in gdb is a little
longer :
$ gdb python2.5
GNU gdb 6.7.1-debian
Copyright (C) 2007 Free Software Foundation, Inc.
License GPLv3+: GNU GPL version 3 or later <http://gnu.org/licenses/gpl.html
>
This is free software: you are free to change and redistribute it.
There is NO WARRANTY, to the extent permitted by law. Type "show copying"
and "show warranty" for details.
This GDB was configured as "i486-linux-gnu"...
Using host libthread_db library "/lib/i686/cmov/libthread_db.so.1".
(gdb) run
Starting program: /home/loic/tmp/bluelagoon/bin/python2.5
[Thread debugging using libthread_db enabled]
Python 2.5.2 (r252:60911, Feb 27 2008, 20:12:29)
[GCC 4.2.3] on linux2
Type "help", "copyright", "credits" or "license" for more information.
[New Thread 0xb7bd4b60 (LWP 6052)]
>>> from pylab import *
[New Thread 0xb67fbb90 (LWP 6055)]
[378309 refs]
>>> plot([1,2,3])
[<matplotlib.lines.Line2D instance at 0x8bd6cac>]
[384105 refs]
>>> show()
Program received signal SIGSEGV, Segmentation fault.
[Switching to Thread 0xb7bd4b60 (LWP 6052)]
0xb5fe49fc in Bbox::ll_api (this=0x7ffffffb) at src/_transforms.h:362
362 Point* ll_api() {return _ll;}
Current language: auto; currently c++
(gdb) bt
#0 0xb5fe49fc in Bbox::ll_api (this=0x7ffffffb) at src/_transforms.h:362
#1 0xb5fe441b in PyAggImagePhoto (clientdata=0x0, interp=0x8b0ab58, argc=5,
argv=0x8d6e028) at src/_tkagg.cpp:91
#2 0xb696cb76 in TclInvokeStringCommand (clientData=0x8d20fe0,
interp=0x8b0ab58, objc=5, objv=0xbfe677d4)
 at
/home/loic/tmp/bluelagoon/src/tcl8.5.1/unix/../generic/tclBasic.c:2105
#3 0xb696e915 in TclEvalObjvInternal (interp=0x8b0ab58, objc=5,
objv=0xbfe677d4, command=0x0, length=0, flags=262144)
 at
/home/loic/tmp/bluelagoon/src/tcl8.5.1/unix/../generic/tclBasic.c:3647
#4 0xb696eeca in Tcl_EvalObjv (interp=0x8b0ab58, objc=5, objv=0xbfe677d4,
flags=262144)
 at
/home/loic/tmp/bluelagoon/src/tcl8.5.1/unix/../generic/tclBasic.c:3842
#5 0xb6cd32be in Tkapp_Call (selfptr=0x8bc7348, args=0x8b588ec) at
/home/loic/tmp/bluelagoon/src/Python-2.5.2/Modules/_tkinter.c:1327
#6 0xb7df6932 in PyCFunction_Call (func=0x8bcd424, arg=0x8b588ec, kw=0x0)
at Objects/methodobject.c:108
#7 0xb7e70fb3 in call_function (pp_stack=0xbfe67e50, oparg=5) at
Python/ceval.c:3573
#8 0xb7e6c1c4 in PyEval_EvalFrameEx (f=0x8e1bd1c, throwflag=0) at
Python/ceval.c:2272
#9 0xb7e6e880 in PyEval_EvalCodeEx (co=0x8b43e68, globals=0x8b817b4,
locals=0x0, args=0x8debbd4, argcount=2, kws=0x8debbdc, kwcount=1,
 defs=0x87a39e8, defcount=2, closure=0x0) at Python/ceval.c:2836
#10 0xb7e7154d in fast_function (func=0x8b82c14, pp_stack=0xbfe68530, n=4,
na=2, nk=1) at Python/ceval.c:3669
#11 0xb7e71171 in call_function (pp_stack=0xbfe68530, oparg=258) at
Python/ceval.c:3594
#12 0xb7e6c1c4 in PyEval_EvalFrameEx (f=0x8deba8c, throwflag=0) at
Python/ceval.c:2272
#13 0xb7e6e880 in PyEval_EvalCodeEx (co=0x8ac8f28, globals=0x8a5fdf4,
locals=0x0, args=0x8deb570, argcount=1, kws=0x8deb574, kwcount=0,
 defs=0x0, defcount=0, closure=0x0) at Python/ceval.c:2836
#14 0xb7e7154d in fast_function (func=0x8bc3214, pp_stack=0xbfe68c10, n=1,
na=1, nk=0) at Python/ceval.c:3669
#15 0xb7e71171 in call_function (pp_stack=0xbfe68c10, oparg=0) at
Python/ceval.c:3594
#16 0xb7e6c1c4 in PyEval_EvalFrameEx (f=0x8deb414, throwflag=0) at
Python/ceval.c:2272
#17 0xb7e6e880 in PyEval_EvalCodeEx (co=0x8ac8ec8, globals=0x8a5fdf4,
locals=0x0, args=0x879ba90, argcount=2, kws=0x0, kwcount=0,
 defs=0x0, defcount=0, closure=0x0) at Python/ceval.c:2836
#18 0xb7ddaac5 in function_call (func=0x8bc31c4, arg=0x879ba7c, kw=0x0) at
Objects/funcobject.c:517
#19 0xb7db2dd3 in PyObject_Call (func=0x8bc31c4, arg=0x879ba7c, kw=0x0) at
Objects/abstract.c:1861
#20 0xb7e71f24 in ext_do_call (func=0x8bc31c4, pp_stack=0xbfe692ec, flags=1,
na=1, nk=0) at Python/ceval.c:3853
#21 0xb7e6c424 in PyEval_EvalFrameEx (f=0x8deaa5c, throwflag=0) at
Python/ceval.c:2312
#22 0xb7e6e880 in PyEval_EvalCodeEx (co=0x8af4c88, globals=0x8ae8ad4,
locals=0x0, args=0xb7b9a948, argcount=20, kws=0x0, kwcount=0,
 defs=0x0, defcount=0, closure=0x0) at Python/ceval.c:2836
#23 0xb7ddaac5 in function_call (func=0x8b77444, arg=0xb7b9a934, kw=0x0) at
Objects/funcobject.c:517
#24 0xb7db2dd3 in PyObject_Call (func=0x8b77444, arg=0xb7b9a934, kw=0x0) at
Objects/abstract.c:1861
#25 0xb7dbedfb in instancemethod_call (func=0x8b77444, arg=0xb7b9a934,
kw=0x0) at Objects/classobject.c:2519
#26 0xb7db2dd3 in PyObject_Call (func=0x8513374, arg=0x8a6b434, kw=0x0) at
Objects/abstract.c:1861
#27 0xb7e70724 in PyEval_CallObjectWithKeywords (func=0x8513374,
arg=0x8a6b434, kw=0x0) at Python/ceval.c:3442
#28 0xb6cd5561 in PythonCmd (clientData=0x84f2538, interp=0x8b0ab58,
argc=20, argv=0x8d6dfcc)
 at /home/loic/tmp/bluelagoon/src/Python-2.5.2/Modules/_tkinter.c:2014
#29 0xb696cb76 in TclInvokeStringCommand (clientData=0x8d86ba8,
interp=0x8b0ab58, objc=20, objv=0x8d6df74)
 at
/home/loic/tmp/bluelagoon/src/tcl8.5.1/unix/../generic/tclBasic.c:2105
#30 0xb696e915 in TclEvalObjvInternal (interp=0x8b0ab58, objc=20,
objv=0x8d6df74, command=0xffffffff <Address 0xffffffff out of bounds>,
 length=-1, flags=0) at
/home/loic/tmp/bluelagoon/src/tcl8.5.1/unix/../generic/tclBasic.c:3647
#31 0xb69c5a80 in TclExecuteByteCode (interp=0x8b0ab58, codePtr=0x8db0670)
 at
/home/loic/tmp/bluelagoon/src/tcl8.5.1/unix/../generic/tclExecute.c:2255
#32 0xb69c3763 in Tcl_ExprObj (interp=0x8b0ab58, objPtr=0x8d94aa8,
resultPtrPtr=0xbfe6ad1c)
 at
/home/loic/tmp/bluelagoon/src/tcl8.5.1/unix/../generic/tclExecute.c:1275
#33 0xb697091c in Tcl_ExprBooleanObj (interp=0x8b0ab58, objPtr=0xb7ef5300,
ptr=0xbfe6ad6c)
 at
/home/loic/tmp/bluelagoon/src/tcl8.5.1/unix/../generic/tclBasic.c:5053
#34 0xb697cd24 in Tcl_IfObjCmd (dummy=0x0, interp=0x8b0ab58, objc=3,
objv=0x8d6de44)
 at /home/loic/tmp/bluelagoon/src/tcl8.5.1/unix/../generic/tclCmdIL.c:233
#35 0xb696e915 in TclEvalObjvInternal (interp=0x8b0ab58, objc=3,
objv=0x8d6de44,
 command=0xbfe6b0b8 "if {\"[139539316resize 385 ?? ?? 490 ?? ?? ?? 650 0
0 ?? 0 ?? ?? .146360228 22 ?? ?? ??]\" == \"break\"} break\n",
---Type <return> to continue, or q <return> to quit---
 length=107, flags=0) at
/home/loic/tmp/bluelagoon/src/tcl8.5.1/unix/../generic/tclBasic.c:3647
#36 0xb696f945 in TclEvalEx (interp=0x8b0ab58,
 script=0xbfe6b0b8 "if {\"[139539316resize 385 ?? ?? 490 ?? ?? ?? 650 0 0
?? 0 ?? ?? .146360228 22 ?? ?? ??]\" == \"break\"} break\n",
 numBytes=107, flags=131072, line=1) at
/home/loic/tmp/bluelagoon/src/tcl8.5.1/unix/../generic/tclBasic.c:4294
#37 0xb696f097 in Tcl_EvalEx (interp=0x8b0ab58,
 script=0xbfe6b0b8 "if {\"[139539316resize 385 ?? ?? 490 ?? ?? ?? 650 0 0
?? 0 ?? ?? .146360228 22 ?? ?? ??]\" == \"break\"} break\n",
 numBytes=107, flags=131072) at
/home/loic/tmp/bluelagoon/src/tcl8.5.1/unix/../generic/tclBasic.c:4000
#38 0xb6a897e4 in Tk_BindEvent (bindingTable=0x8c3b880, eventPtr=0xbfe6b314,
tkwin=0x8c5c508, numObjects=0, objectPtr=0xbfe6b248)
 at /home/loic/tmp/bluelagoon/src/tk8.5.1/unix/../generic/tkBind.c:1773
#39 0xb6a90a88 in TkBindEventProc (winPtr=0x8c5c508, eventPtr=0xbfe6b314)
 at /home/loic/tmp/bluelagoon/src/tk8.5.1/unix/../generic/tkCmds.c:284
#40 0xb6a9af04 in Tk_HandleEvent (eventPtr=0xbfe6b314) at
/home/loic/tmp/bluelagoon/src/tk8.5.1/unix/../generic/tkEvent.c:1423
#41 0xb6abfc2e in TkDoConfigureNotify (winPtr=0x8c5c508) at
/home/loic/tmp/bluelagoon/src/tk8.5.1/unix/../generic/tkWindow.c:2216
#42 0xb6abf208 in Tk_MakeWindowExist (tkwin=0x8c5c508) at
/home/loic/tmp/bluelagoon/src/tk8.5.1/unix/../generic/tkWindow.c:1768
#43 0xb6abeef4 in Tk_MapWindow (tkwin=0x8c5c508) at
/home/loic/tmp/bluelagoon/src/tk8.5.1/unix/../generic/tkWindow.c:1641
#44 0xb6ab35ad in ArrangePacking (clientData=0x8c223d0) at
/home/loic/tmp/bluelagoon/src/tk8.5.1/unix/../generic/tkPack.c:820
#45 0xb6a2d483 in TclServiceIdle () at
/home/loic/tmp/bluelagoon/src/tcl8.5.1/unix/../generic/tclTimer.c:738
#46 0xb6a0aa69 in Tcl_DoOneEvent (flags=-3) at
/home/loic/tmp/bluelagoon/src/tcl8.5.1/unix/../generic/tclNotify.c:991
#47 0xb6cd6f1a in Tkapp_MainLoop (selfptr=0x8bc7348, args=0x8256dc4) at
/home/loic/tmp/bluelagoon/src/Python-2.5.2/Modules/_tkinter.c:2547
#48 0xb7df67c1 in PyCFunction_Call (func=0x8b943ec, arg=0x8256dc4, kw=0x0)
at Objects/methodobject.c:73
#49 0xb7e70fb3 in call_function (pp_stack=0xbfe6bb10, oparg=1) at
Python/ceval.c:3573
#50 0xb7e6c1c4 in PyEval_EvalFrameEx (f=0x8de9ffc, throwflag=0) at
Python/ceval.c:2272
#51 0xb7e6e880 in PyEval_EvalCodeEx (co=0x8adda48, globals=0x8ae8ad4,
locals=0x0, args=0x8de9288, argcount=0, kws=0x8de9288, kwcount=0,
 defs=0x8b6e4a8, defcount=1, closure=0x0) at Python/ceval.c:2836
#52 0xb7e7154d in fast_function (func=0x8b6f034, pp_stack=0xbfe6c1f0, n=0,
na=0, nk=0) at Python/ceval.c:3669
#53 0xb7e71171 in call_function (pp_stack=0xbfe6c1f0, oparg=0) at
Python/ceval.c:3594
#54 0xb7e6c1c4 in PyEval_EvalFrameEx (f=0x8de913c, throwflag=0) at
Python/ceval.c:2272
#55 0xb7e6e880 in PyEval_EvalCodeEx (co=0x8ac8ce8, globals=0x8a5fdf4,
locals=0x0, args=0x8de9118, argcount=0, kws=0x8de9118, kwcount=0,
 defs=0x0, defcount=0, closure=0x0) at Python/ceval.c:2836
#56 0xb7e7154d in fast_function (func=0x8bc3084, pp_stack=0xbfe6c8d0, n=0,
na=0, nk=0) at Python/ceval.c:3669
#57 0xb7e71171 in call_function (pp_stack=0xbfe6c8d0, oparg=0) at
Python/ceval.c:3594
#58 0xb7e6c1c4 in PyEval_EvalFrameEx (f=0x8de8fd4, throwflag=0) at
Python/ceval.c:2272
#59 0xb7e6e880 in PyEval_EvalCodeEx (co=0xb7bcc268, globals=0xb7bb9214,
locals=0xb7bb9214, args=0x0, argcount=0, kws=0x0, kwcount=0,
 defs=0x0, defcount=0, closure=0x0) at Python/ceval.c:2836
#60 0xb7e6352c in PyEval_EvalCode (co=0xb7bcc268, globals=0xb7bb9214,
locals=0xb7bb9214) at Python/ceval.c:494
#61 0xb7e9c10b in run_mod (mod=0x80b80e0, filename=0xb7ee4b33 "<stdin>",
globals=0xb7bb9214, locals=0xb7bb9214, flags=0xbfe6cc04,
 arena=0x80589a8) at Python/pythonrun.c:1273
#62 0xb7e9a621 in PyRun_InteractiveOneFlags (fp=0xb7d1e440,
filename=0xb7ee4b33 "<stdin>", flags=0xbfe6cc04) at Python/pythonrun.c:792
#63 0xb7e9a24a in PyRun_InteractiveLoopFlags (fp=0xb7d1e440,
filename=0xb7ee4b33 "<stdin>", flags=0xbfe6cc04) at Python/pythonrun.c:723
#64 0xb7e9a076 in PyRun_AnyFileExFlags (fp=0xb7d1e440, filename=0xb7ee4b33
"<stdin>", closeit=0, flags=0xbfe6cc04)
 at Python/pythonrun.c:692
#65 0xb7eaa63b in Py_Main (argc=1, argv=0xbfe6cd14) at Modules/main.c:523
(gdb)
I'm not used to debugging C code or to cope with segmentation faults.
Do you see anything which could give me a clue for fixing this pb ?
Thanks for the reply,
-- 
LB

 
Hello,
 I've been working with the example file finance_work2.py on the
screenshots page. It seems to me that the author tried to get volume
bars to show under the candlestick chart on different axes but on the
same plot, however this does not show up in the screenshots, or when I
try to run the program. I've been working on this for several days to
try and get them to show up, but to no avail. Does anyone have any
experience with this?
 Also, I have a simple question. When I'm making several plots in the
same figure, the xaxis mouse-over data only shows up for the lowest
plot. Is this a limitation of matplotlib or is there a function I can
call so that the mouseover shows the x coordinates on the other plots as
well.
Many thanks in advance.
-Ryan Wagner
 
 

 Greetings... I am working using
the matplotlib version 0.90.1, and i am having some strange problems with
my application. When i execute in some pcs, there is no
problem at all, but in others pcs, i receive this message:
File "pylab.pyc", line 1 in ?
File "matplotlib\pylab.pyc", line 203 in ?
File "matplotlib\axes.pyc", line 14 in ?
File "matplotlib\agg.pyc", line 5 in ?
File "matplotlib\_agg.pyc", line 12 in ?
File "matplotlib\_agg.pyc", line 10 in ?
ImportError: Dll Load Failed
All the pc's have Windows Xp, most of them Pentium 4, some
are Pentium D. I don't know what may be the issue. All those
pcs have the same software, just different hardware, i hope
you could help me with this little problem. Thanks in advance...

On Wed, Feb 27, 2008 at 1:38 PM, Paul Milliken
<pau...@gm...> wrote:
> When I download the source code for matplotlib (the file is
> matplotlib-0.91.2.tar.gz) from
> http://sourceforge.net/project/showfiles.php?group_id=80706 I can't uzip it.
> Specifically, the command "tar -xzf matplotlib-0.91.2.tar.gz" returns:
>
> gzip: stdin: unexpected end of file
> tar: Skipping to next header
> tar: Child returned status 1
> tar: Error exit delayed from previous errors
Works for me -- perhaps it is your download client?
johnh@flag:tmp> wget
http://superb-east.dl.sourceforge.net/sourceforge/matplotlib/matplotlib-0.91.2.tar.gz
--13:44:01-- http://superb-east.dl.sourceforge.net/sourceforge/matplotlib/matplotlib-0.91.2.tar.gz
 => `matplotlib-0.91.2.tar.gz'
Resolving superb-east.dl.sourceforge.net... 209.160.66.130
Connecting to superb-east.dl.sourceforge.net[209.160.66.130]:80... connected.
HTTP request sent, awaiting response... 200 OK
Length: 6,457,944 [application/x-gzip]
100%[===============================================================>]
6,457,944 1.49M/s ETA 00:00
13:44:07 (1.11 MB/s) - `matplotlib-0.91.2.tar.gz' saved [6457944/6457944]
johnh@flag:tmp> tar tvfz matplotlib-0.91.2.tar.gz |head
drwxr-xr-x cmoad/staff 0 2008年01月06日 12:31:24 matplotlib-0.91.2/
-rw-r--r-- cmoad/staff 38340 2008年01月06日 12:24:31 matplotlib-0.91.2/API_CHANGES
-rw-r--r-- cmoad/staff 130839 2008年01月06日 12:25:41 matplotlib-0.91.2/CHANGELOG
drwxr-xr-x cmoad/staff 0 2008年01月06日 12:31:23 matplotlib-0.91.2/CXX/
-rw-r--r-- cmoad/staff 4114 2007年11月29日 20:36:50
matplotlib-0.91.2/CXX/Config.hxx
-rw-r--r-- cmoad/staff 7008 2007年11月29日 20:36:50
matplotlib-0.91.2/CXX/Exception.hxx
-rw-r--r-- cmoad/staff 26950 2007年11月29日 20:36:50
matplotlib-0.91.2/CXX/Extensions.hxx
-rw-r--r-- cmoad/staff 25172 2007年11月29日 20:36:50
matplotlib-0.91.2/CXX/IndirectPythonInterface.cxx
-rw-r--r-- cmoad/staff 5665 2007年11月29日 20:36:50
matplotlib-0.91.2/CXX/IndirectPythonInterface.hxx
-rw-r--r-- cmoad/staff 82717 2007年11月29日 20:36:50
matplotlib-0.91.2/CXX/Objects.hxx
johnh@flag:tmp>

When I download the source code for matplotlib (the file is
matplotlib-0.91.2.tar.gz)<http://sourceforge.net/project/showfiles.php?group_id=80706&package_id=82474&release_id=566411>from
http://sourceforge.net/project/showfiles.php?group_id=80706 I can't uzip
it. Specifically, the command "tar -xzf matplotlib-0.91.2.tar.gz" returns:
gzip: stdin: unexpected end of file
tar: Skipping to next header
tar: Child returned status 1
tar: Error exit delayed from previous errors
*
I also tried downloaded older versions from sourceforge but with the same
result. (The mirror that I downloaded from was optusnet.dl.sourceforge.net
.)*
Has anyone else had this problem?
Regards,
Paul Milliken

You have run into a bug in the combination of poly_between and 
fill--maybe only the former, which not taking masked arrays into 
account. I have not looked at it enough to know whether it will be easy 
or hard to fix, but it certainly should be fixed. I can't look at it 
more right now, unfortunately.
Eric
Michaël Douchin wrote:
> Hi list
> 
> I searched the list and google, but couldn't find a way to solve my pbm.
> 
> I have data stored in a list (from an sql query) , with these "columns":
> x = time serie in hours
> y = some level value
> 
> There are some missing values : eg between 08:33 and 08:40.
> Here is my code :
> sqla="SELECT * FROM import_parcelle a WHERE dat_loc='" + date_traite 
> + "' AND code_uc='" + code_uc + "' ORDER BY a.heu_loc ;"
> resa=db.query(sqla)
> data=resa.dictresult() 
> #x = time serie
> x= [ datetime.datetime(*time.strptime(a["dat_loc"]+" 
> "+a["heu_loc"],'%Y-%m-%d %H:%M:%S')[0:6] ) for a in data]
> #y5 = tank level
> y5= [float(a["niv_cuv"]) for a in data] # --> extraction de la 
> colonne y1
> figure()
> # the plot command with no mask
> 
> plot_date(x,y5,color='b',linestyle='None',marker='',xdate=True,ydate=False)
> #the mask
> #ym5 = ma.masked_where(y5 <300, y5)
> #plot_date(x,ym5,color='r',linestyle='-',xdate=True,ydate=False)
> #the filling under the curve
> xs, ys = poly_between(x, 0, y5)
> fill(xs,ys)
> 
> 
> Here is the result:
> http://michaeldouchin.free.fr/17_2007年06月07日_10B_vitesse.png
> As you see, I commented the lines with the mask, because it did not 
> change anything
> 
> To see what I am looking for, here is the result under R (a statistical 
> tool)
> http://michaeldouchin.free.fr/17_2007年06月07日_10B_vitesse.jpg
> As you see, between 08:48 and 08:50 (for example), there is a gap, 
> showing we have no data for this interval.
> 
> As I want to automatically draw this graph for different set of data, I 
> can't look each set in detail.
> I tried to folow the example masked_demo.py, but I could not adapt it to 
> my case....
> 
> Any hint ?
> Thanks very much in advance
> 
> Michael
> 
> -------------------------------------------------------------------------
> This SF.net email is sponsored by: Microsoft
> Defy all challenges. Microsoft(R) Visual Studio 2008.
> http://clk.atdmt.com/MRT/go/vse0120000070mrt/direct/01/
> _______________________________________________
> Matplotlib-users mailing list
> Mat...@li...
> https://lists.sourceforge.net/lists/listinfo/matplotlib-users

Okay I've learnt a bit more about this:
http://img229.imageshack.us/img229/6467/scatterplot5st6.png
I need to explicitly make a new figure as well as a new axis, and put
the rc calls before the figure calls, rc changes only seem to take
effect on figures created afterwards. Also I was stupidly using the plot
function (for plotting lines) for a scatter plot, I've now discovered
the scatter function.
Things are pretty much how I want them now, but I still have a couple of
questions:
How do I get rid of the redundant ticks on the top and right edges?
Is there any way to stop some of my text labels from overlapping?
Why is there such a big gap between the plot itself and the axes ticks?
The scatter function seems to do this, but I don't see any option to
control the size of the gap. I wonder if I've done something weird like
create an axes within an axes. I would like to get rid of the gap because
it makes the graphic less readable, in particular it gives a misleading
impression about where (0,0) is located.
Here's my new source code:
rc('xtick.major',size=8)
rc('xtick',direction='out')
rc('ytick.major',size=8)
rc('ytick',direction='out')
fig = figure(facecolor='white' )
axes(frameon=False)
title('Comparison of lower- and upper-bounds per function in week one
and week two')
l =
scatter(functions.lowerboundsweek1,functions.lowerboundsweek2,marker='s',color='g')
u =
scatter(functions.upperboundsweek1,functions.upperboundsweek2,marker='^',color='b')
legend((l,u),('Lower-bounds','Upper-bounds'),'best')
plot([0,1],[0,1],color='#999999',linestyle='--')
xlabel('Week 1')
ylabel('Week 2')
xticks(functions.upperboundsweek1+functions.lowerboundsweek1,'
'*len(functions.upperboundsweek1+functions.lowerboundsweek1))
yticks(functions.upperboundsweek2+functions.lowerboundsweek2,'
'*len(functions.upperboundsweek2+functions.lowerboundsweek2))
for function,x,y in
map(None,functions.shortnames,functions.upperboundsweek1,functions.upperboundsweek2):
 annotate(function,xy=(x,y),size=8,color='b') 
for function,x,y in
map(None,functions.shortnames,functions.lowerboundsweek1,functions.lowerboundsweek2):
 annotate(function,xy=(x,y),size=8,color='g') 
On Tue, 2008年02月26日 at 16:51 +0000, chombee wrote:
> I'm having a couple of problems drawing a basic relational scatter plot.
> (Specifically, it's called a dot-dash-plot in the book I have and is
> described as "framing the bivariate scatter with the marginal distribution
> of each variable.") The idea is that you have a bivariate scatter plot
> and use the two marginal frequency distributions of the data as the two
> axes.
> 
> 1. By setting the x and y tick length and using space characters as the
> tick labels I was able to get ticks only and no labels on the axes as I
> want. But these ticks appear along the top and right edges of the plot,
> as well as the left and bottom. Is there any way to get rid of the top
> and right ticks?
> 
> 2. I'm getting phantom numbers on my axes, the 0.2, 0.4, 0.6 etc. I
> don't know where these are coming from or what they measure (my data
> ranges from 0-7 on both axes, not 0-1).
> 
> 3. I want to label each data point with a string. I'm doing this with
> annotate but the strings sometimes overlap each other. Any idea how to
> avoid this?
> 
> Here's a PNG of my plot as it is:
> 
> http://img137.imageshack.us/img137/4015/scatterplotlk2.png
> 
> And here's the source:
> 
> # datax and datay are equal length lists of floats, the data I want to 
> # plot. labels is another equal length list containing the strings I 
> # want to label the data points with. All three lists are in the same
> # order.
> 
> axes(frameon=False)
> rc('xtick.major',size=8)
> rc('xtick',direction='out')
> xticks(datax,' '*len(datax))
> rc('ytick.major',size=8)
> rc('ytick',direction='out')
> yticks(datay,' '*len(datay))
> plot(datax,datay,'r.')
> plot([0,7],[0,7]) 
> xlabel('Proportion of total number of cards, week 1')
> ylabel('Proportion of total number of cards, week 2')
> title('Frequency of cards per function')
> for label,x,y in map(None,labels,datax,datay):
> annotate(function,xy=(x,y),size=8)
> 
> Thanks!
> 
> 
> 
> -------------------------------------------------------------------------
> This SF.net email is sponsored by: Microsoft
> Defy all challenges. Microsoft(R) Visual Studio 2008.
> http://clk.atdmt.com/MRT/go/vse0120000070mrt/direct/01/
> _______________________________________________
> Matplotlib-users mailing list
> Mat...@li...
> https://lists.sourceforge.net/lists/listinfo/matplotlib-users

On Wednesday 27 February 2008 12:21:18 pm Jochen Voss wrote:
> Hi Darren,
>
> On Wed, Feb 27, 2008 at 11:33:00AM -0500, Darren Dale wrote:
> > Thanks for the images, I see what you are saying.
> >
> > However, when I run your broken.py script I dont see any problems. The ps
> > and eps look like the png output.
>
> Hmmm, this is strange. I remember that things worked a bit earlier on
> my powerbook. Maybe it works on MacOSX (matplotlib installed via
> fink) and does not work on Debian (matplotlib installed from source)?
>
> > This is with mpl-0.91.2, I have the distiller
> > turned off and usetex is false. Have you changed any rc settings that
> > might have effected your output?
>
> I don't think I have changed rc settings outside the script: I deleted
> the ~/.matplotlib/ directory and had it recreated. I do not have an
> /etc/matplotlibrc file. Is there another place where leftover files
> could cause spurious settings to apply?
>
> I have the line
>
> rc('text', usetex=True)
>
> in the script, though. Surprise: if I comment out this line, both
> out.eps and out.ps suddenly look ok!!! Does usetex force this
> distilling stuff to kick in? What external dependencies does this
> use?
usetex has to use a distiller, either ghostscript or xpdf (xpdf is a bad name 
for it, it requires ps2pdf and pdftops, the latter is now supplied by 
poppler). The reason is that latex with the PSFrag package generates ps and 
eps files that cannot be embedded in another postscript file. So we pass it 
through a distiller. Trouble like you are seeing is almost always solved by 
upgrading your ghostscript. I am using gpl ghostscript-8.61.

Hi Darren,
On Wed, Feb 27, 2008 at 11:33:00AM -0500, Darren Dale wrote:
> Thanks for the images, I see what you are saying.
> 
> However, when I run your broken.py script I dont see any problems. The ps and 
> eps look like the png output.
Hmmm, this is strange. I remember that things worked a bit earlier on
my powerbook. Maybe it works on MacOSX (matplotlib installed via
fink) and does not work on Debian (matplotlib installed from source)?
> This is with mpl-0.91.2, I have the distiller 
> turned off and usetex is false. Have you changed any rc settings that might 
> have effected your output?
I don't think I have changed rc settings outside the script: I deleted
the ~/.matplotlib/ directory and had it recreated. I do not have an
/etc/matplotlibrc file. Is there another place where leftover files
could cause spurious settings to apply?
I have the line
 rc('text', usetex=True)
in the script, though. Surprise: if I comment out this line, both
out.eps and out.ps suddenly look ok!!! Does usetex force this
distilling stuff to kick in? What external dependencies does this
use?
Many thanks,
Jochen
-- 
http://seehuhn.de/

On Wednesday 27 February 2008 11:25:25 am Jochen Voss wrote:
> Hi Darren,
>
> On Wed, Feb 27, 2008 at 09:24:18AM -0500, Darren Dale wrote:
> > I don't see the problem here with either 0.91.2, the maintanance branch,
> > or the development branch. Maybe I dont know what I am looking for. Could
> > you please send me an eps and png example so I can compare my results
> > with yours?
>
> Thanks for having a look. I append a slightly modified version of the
> script and output files "out.ps", "out.eps" and "out.png". The files
> are created using the calls
>
> ./broken.py ps
> ./broken.py eps
> ./broken.py png
>
> out.ps - new failure mode: too small picture (shows only the pixel data)
> out.eps - shifted as before
> out.png - this looks ok.
Thanks for the images, I see what you are saying.
However, when I run your broken.py script I dont see any problems. The ps and 
eps look like the png output. This is with mpl-0.91.2, I have the distiller 
turned off and usetex is false. Have you changed any rc settings that might 
have effected your output?

Hi Darren,
On Wed, Feb 27, 2008 at 09:24:18AM -0500, Darren Dale wrote:
> I don't see the problem here with either 0.91.2, the maintanance branch, or 
> the development branch. Maybe I dont know what I am looking for. Could you 
> please send me an eps and png example so I can compare my results with yours? 
Thanks for having a look. I append a slightly modified version of the
script and output files "out.ps", "out.eps" and "out.png". The files
are created using the calls
 ./broken.py ps
 ./broken.py eps
 ./broken.py png
out.ps - new failure mode: too small picture (shows only the pixel data)
out.eps - shifted as before
out.png - this looks ok.
Many thanks,
Jochen
-- 
http://seehuhn.de/

Hi Jeff,
yes I see this method, but set_xlim and set_ylim use axes coordinates, and I 
would like to use geographic coordinates.
It would be great if we could put an image in a map directly by specifying its 
geographic coordinates
Le mardi 26 février 2008, Jeff Whitaker a écrit :
> Lionel Roubeyrie wrote:
> > Hi all,
> > I'm sure it's a trivial question, but can't find any valid answer in the
> > basemap examples directory or with google : I have a georeferenced TIFF
> > file in 'lcc' projection, representing a little portion of France, and I
> > need to put it on a map, resets map limits to a closed portion of the
> > map, and put contours on it. I see how to initialise Basemap with the
> > limits of the raster, but not how to "make a zoom" in the map by
> > specifing coordinates. Thanks for your help
>
> Lionel: You can use the set_xlim and set_ylim axes methods to manually
> 'zoom' the plot. It has to be done after all the other plotting though,
> since all the Basemap methods (contour, imshow etc) will reset the axes
> limits to show the entire map region.
>
> -Jeff
-- 
Lionel Roubeyrie - lro...@li...
Chargé d'études et de maintenance
LIMAIR - la Surveillance de l'Air en Limousin
http://www.limair.asso.fr

I have had a similar problem here with python-2.5.1, numpy-1.0.4 and 
mpl-0.91.2 all installed on my linux (from scratch) system. I reported 
this last year but ending in a dead end with no one being able to help. 
My --verbose-debug_annoying output seemed to indicate it had something 
to do with the loading of the fonts, like yours, but that's all the 
farther we all got. My work around was to install the prerequisites for 
GtkAgg and go with that. Everything works fine now using GtkAgg.
I even recently installed ATLAS using lapack so I could install numpy 
and scipy with their full functionality and was also successful doing 
that. I'm able to do curve fitting and find roots of polynomials. So I 
believe the TkAgg backend does have some incompatibilities but no one 
has a clue as to why. Otherwise, it's all smooth sailing doing it all 
from scratch.
Take Good Care,
Wayne
LB wrote:
> I'm still investigating for my segmentation fault pb with matplotlib 
> and Tkagg,
> but now, I have no more idea.
>