matplotlib-users Mailing List for matplotlib

I'm developing an application for personal use on my Mac, using the 
GTAgg backend, which renders a series of plots in a figure in a pygtk 
scrolled window. I'm currently using matplotlib 0.87.3 installed via 
Fink. Matplotlib is performing well.
Today I had occasion to log into my Mac over ssh from work. I noted 
that the application was extremely slow to display the plots. 
Running locally on my Mac, the application takes about 5 seconds to 
display the plots when it opens. Remotely, it took around 5 minutes 
(repeatable). On a whim, I tried changing to the GTK backend - now 
it displayed the plots in about 20 seconds, which is quite acceptable 
considering it is remote access over an ssh tunnel.
I checked the CPU usage on both the Mac, and my work PC - the CPU 
usage was very, very low on both machines, so I'm not sure where the 
bottleneck lies. My install rights on the work PC are very 
restricted, so I'm using CygWin via XLiveCD, which I have copied onto 
the hard drive to speed it up.
http://xlivecd.indiana.edu/
This poor performance when using GTKAgg isn't really a problem - I'll 
code a command line switch to change to the GTK backend for the few 
times I need remote access. I'm reporting it just in case one of the 
developers thinks it could be useful to troubleshoot, in case it 
highlights a problem in the code. I also want to note that the GTK 
backend is useful to some people, and hope that it will continue to 
be supported (I seem to recall one mention that it was little used).
Thanks for matplotlib.
Kevin Horton
Ottawa, Canada

I had the same problem today, and solved it by using
matplotlib-0.87.2.win32-py2.4.exe
On Wed, 2006年09月06日 at 15:06 -0500, Ryan Krauss wrote:
> I am trying to install matplotlib on my office Windows computer. I
> have installed the following:
> python-2.4.3.msi
> numpy-1.0b5.win32-py2.4.exe
> scipy-0.5.1.win32-py2.4.exe
> matplotlib-0.87.5.win32-py2.4.exe
> 
> on a completely fresh install.
> 
> I am getting the following message:
> H:\>python
> Python 2.4.3 (#69, Mar 29 2006, 17:35:34) [MSC v.1310 32 bit (Intel)] on win32
> Type "help", "copyright", "credits" or "license" for more information.
> >>> import pylab
> Traceback (most recent call last):
> File "<stdin>", line 1, in ?
> File "C:\Python24\Lib\site-packages\pylab.py", line 1, in ?
> from matplotlib.pylab import *
> ImportError: No module named matplotlib.pylab
> >>> import matplotlib
> Traceback (most recent call last):
> File "<stdin>", line 1, in ?
> ImportError: No module named matplotlib
> 
> What am I doing wrong?
> 
> Thanks,
> 
> Ryan
> 
> -------------------------------------------------------------------------
> Using Tomcat but need to do more? Need to support web services, security?
> Get stuff done quickly with pre-integrated technology to make your job easier
> Download IBM WebSphere Application Server v.1.0.1 based on Apache Geronimo
> http://sel.as-us.falkag.net/sel?cmd=lnk&kid=120709&bid=263057&dat=121642
> _______________________________________________
> Matplotlib-users mailing list
> Mat...@li...
> https://lists.sourceforge.net/lists/listinfo/matplotlib-users
> 

Hi,
I installed matplotlib for winXP, and when I try to import pylab I get
this error:
>>> from pylab import *
Traceback (most recent call last):
 File "<pyshell#0>", line 1, in -toplevel-
 from pylab import *
 File "C:\Python24\Lib\site-packages\pylab.py", line 1, in -toplevel-
 from matplotlib.pylab import *
 File "C:\Python24\Lib\site-packages\matplotlib\pylab.py", line 196, in
-toplevel-
 import cm
 File "C:\Python24\Lib\site-packages\matplotlib\cm.py", line 5, in
-toplevel-
 import colors
 File "C:\Python24\Lib\site-packages\matplotlib\colors.py", line 33, in
-toplevel-
 from numerix import array, arange, take, put, Float, Int, where, \
 File "C:\Python24\Lib\site-packages\matplotlib\numerix\__init__.py",
line 60, in -toplevel-
 from Numeric import *
I'm using matplotlib-0.87.2.win32-py2.4.exe
Does anyone know what I'm doing wrong?
Thanks!

Charlie Moad wrote:
> Try the ones I added to the link above. I already had a mpl build on
> sf. Numpy is also at the link now.
Thanks Charlie. These seem to work, on the simplest of scripts, anyway. 
Did you build the wxPython accelerator into it?
-Chris
-- 
Christopher Barker, Ph.D.
Oceanographer
 		
NOAA/OR&R/HAZMAT (206) 526-6959 voice
7600 Sand Point Way NE (206) 526-6329 fax
Seattle, WA 98115 (206) 526-6317 main reception
Chr...@no...

I just built the latest on my Linux box, and noticed that it didn't find 
 the wxPython headers. I figured out what to do but I have the 
following suggestions:
1) at the very least, make the message more useful. I've enclosed a diff 
 to setupext.py that adds some more text.
2) It wouldn't be too hard to make it a bit smarter about finding the 
right wx-config. wxPython does not install it to a dir on the PATH by 
default. In my case, I have it there from other wx development installs. 
 The right wx-config, on my machine, is in:
/usr/lib/wxPython-2.6.1.0-gtk2-unicode/bin/wx-config
by checking the wxPython version installed, you should be able to find 
it. Or you could use "locate", which is how I found it. I don't know how 
ubiquitous locate is on *nix systems, however.
locate "*wxPython-2.6.3.0*wx-config"
3) Re-do the wxAgg back-end to not need any compiled code! -- one of 
these days....
-Chris
-- 
Christopher Barker, Ph.D.
Oceanographer
 		
NOAA/OR&R/HAZMAT (206) 526-6959 voice
7600 Sand Point Way NE (206) 526-6329 fax
Seattle, WA 98115 (206) 526-6317 main reception
Chr...@no...

I am trying to install matplotlib on my office Windows computer. I
have installed the following:
python-2.4.3.msi
numpy-1.0b5.win32-py2.4.exe
scipy-0.5.1.win32-py2.4.exe
matplotlib-0.87.5.win32-py2.4.exe
on a completely fresh install.
I am getting the following message:
H:\>python
Python 2.4.3 (#69, Mar 29 2006, 17:35:34) [MSC v.1310 32 bit (Intel)] on win32
Type "help", "copyright", "credits" or "license" for more information.
>>> import pylab
Traceback (most recent call last):
 File "<stdin>", line 1, in ?
 File "C:\Python24\Lib\site-packages\pylab.py", line 1, in ?
 from matplotlib.pylab import *
ImportError: No module named matplotlib.pylab
>>> import matplotlib
Traceback (most recent call last):
 File "<stdin>", line 1, in ?
ImportError: No module named matplotlib
What am I doing wrong?
Thanks,
Ryan

Try the ones I added to the link above. I already had a mpl build on
sf. Numpy is also at the link now.
On 9/6/06, Christopher Barker <Chr...@no...> wrote:
> Charlie Moad wrote:
> > I just uploaded "working" (at least for me)
> > versions to sourceforge again.
>
> Charlie,
>
> What are your plans for an OS-X build? Is it time?
>
> We'd need a new numpy build too -- that I could do.
>
> -Chris
>
>
>
> --
> Christopher Barker, Ph.D.
> Oceanographer
>
> NOAA/OR&R/HAZMAT (206) 526-6959 voice
> 7600 Sand Point Way NE (206) 526-6329 fax
> Seattle, WA 98115 (206) 526-6317 main reception
>
> Chr...@no...
>
> -------------------------------------------------------------------------
> Using Tomcat but need to do more? Need to support web services, security?
> Get stuff done quickly with pre-integrated technology to make your job easier
> Download IBM WebSphere Application Server v.1.0.1 based on Apache Geronimo
> http://sel.as-us.falkag.net/sel?cmd=lnk&kid=120709&bid=263057&dat=121642
> _______________________________________________
> Matplotlib-users mailing list
> Mat...@li...
> https://lists.sourceforge.net/lists/listinfo/matplotlib-users
>

Charlie Moad wrote:
> I just uploaded "working" (at least for me)
> versions to sourceforge again.
Charlie,
What are your plans for an OS-X build? Is it time?
We'd need a new numpy build too -- that I could do.
-Chris
-- 
Christopher Barker, Ph.D.
Oceanographer
 		
NOAA/OR&R/HAZMAT (206) 526-6959 voice
7600 Sand Point Way NE (206) 526-6329 fax
Seattle, WA 98115 (206) 526-6317 main reception
Chr...@no...

Thanks alot, everything seems to be ok now! (with numpy 1.0b5)
-sven
Charlie Moad schrieb:
> Sorry all for the error. I just uploaded "working" (at least for me)
> versions to sourceforge again. For those who are sicking of playing
> with mirrors here is a direct download. Don't expect this link to be
> good for a long time though.
> 
> http://euclid.uits.iupui.edu/mplfiles/
> 
> - Charlie
> 
> On 9/6/06, Charlie Moad <cw...@gm...> wrote:
>> The source error must of propagated to those builds. I will post new
>> ones shortly.
>>
>> On 9/6/06, Sven Schreiber <sve...@gm...> wrote:
>> > Well the thread on the devel list that I referred to explicitly has the
>> > win32 exe in its title ("Missing __init__.py in
>> > matplotlib-0.87.5.win32-py2.4.exe ?") . The starting post there pretty
>> > much says it all.
>> >
>> > There's also a recent post on the numpy list that sounds like it's
>> maybe
>> > the same problem (quote):
>> > """
>> > Is there a compatible matplotlib as well? I was o.k. with mpl from
>> > enthought until I switched numerix to numpy. That made mpl unhappy.
>> > I downloaded 0.87.5 but I broke something in the process because now
>> > even switching back to Numeric doesn't make mpl happy.
>> > """
>> >
>> > Thanks,
>> > Sven
>> >
>> > Charlie Moad schrieb:
>> > > That error was relating to the source release. Can you please post
>> > > your error for the binary?
>> > >
>> > > On 9/6/06, Sven Schreiber <sve...@gm...> wrote:
>> > >> Charlie Moad schrieb:
>> > >> > Minor rev bump for numpy 1.0b5 compatibility. This release should
>> > >> > remain compatible with future 1.0 releases of numpy.
>> > >> >
>> > >>
>> > >> I keep running into the ImportError problem described on the
>> devel list,
>> > >> with the win32 2.4 binary (exe). Is that still just a case of
>> waiting
>> > >> for the mirrors to update, or is there a deeper problem? Are there
>> > >> alternative download links? How to tell whether it's the "right"
>> binary,
>> > >> since version numbers are the same?
>> > >>
>> > >> Thanks for your help,
>> > >> Sven
>> > >>
>> > >
>> >
>> >
>>
> 

Esdras Caleb wrote:
>i see the tutorial but only do plot dont plot the points it do a
>interpolation in the opints and give a line or points folloing the line...
>someone can say to me hoy only draw points?
It sounds like you want to specify that the plot command should use a marker
other than line, which is the default.
For instance if you have been doing
plot(x,y) 
and getting a line, then try doing this instead
plot(x,y,'.') 
to get unconnected dots.
You can use a variety of markers; there's a list in the pylab 
documentation somewhere.
Hope this helps,
Vicki Laidler
 

Sorry all for the error. I just uploaded "working" (at least for me)
versions to sourceforge again. For those who are sicking of playing
with mirrors here is a direct download. Don't expect this link to be
good for a long time though.
http://euclid.uits.iupui.edu/mplfiles/
- Charlie
On 9/6/06, Charlie Moad <cw...@gm...> wrote:
> The source error must of propagated to those builds. I will post new
> ones shortly.
>
> On 9/6/06, Sven Schreiber <sve...@gm...> wrote:
> > Well the thread on the devel list that I referred to explicitly has the
> > win32 exe in its title ("Missing __init__.py in
> > matplotlib-0.87.5.win32-py2.4.exe ?") . The starting post there pretty
> > much says it all.
> >
> > There's also a recent post on the numpy list that sounds like it's maybe
> > the same problem (quote):
> > """
> > Is there a compatible matplotlib as well? I was o.k. with mpl from
> > enthought until I switched numerix to numpy. That made mpl unhappy.
> > I downloaded 0.87.5 but I broke something in the process because now
> > even switching back to Numeric doesn't make mpl happy.
> > """
> >
> > Thanks,
> > Sven
> >
> > Charlie Moad schrieb:
> > > That error was relating to the source release. Can you please post
> > > your error for the binary?
> > >
> > > On 9/6/06, Sven Schreiber <sve...@gm...> wrote:
> > >> Charlie Moad schrieb:
> > >> > Minor rev bump for numpy 1.0b5 compatibility. This release should
> > >> > remain compatible with future 1.0 releases of numpy.
> > >> >
> > >>
> > >> I keep running into the ImportError problem described on the devel list,
> > >> with the win32 2.4 binary (exe). Is that still just a case of waiting
> > >> for the mirrors to update, or is there a deeper problem? Are there
> > >> alternative download links? How to tell whether it's the "right" binary,
> > >> since version numbers are the same?
> > >>
> > >> Thanks for your help,
> > >> Sven
> > >>
> > >
> >
> >
>

>>>>> "Esdras" == Esdras Caleb <ach...@gm...> writes:
 Esdras> i see the tutorial but only do plot dont plot the points
 Esdras> it do a interpolation in the opints and give a line or
 Esdras> points folloing the line... someone can say to me hoy
 Esdras> only draw points? PS: sorry john i dont see...
As I said before offlist, the 2nd example in the tutorial with a
figure entitled "Using format strings" covers this.
You will notice that the plot there *does not* have a line connecting
the markers. See also the documentation for plot. Basically, you
want to set a line marker in the plot format string
plot(x, y, '-') # a solid line connecting the data points
plot(x, y, 'o') # circle markers, no solid line
plot(x, y, 's') # square markers, no solid line
plot(x, y, '-o') # circle markers, with connecting solid line

i see the tutorial but only do plot dont plot the points it do a
interpolation in the opints and give a line or points folloing the line...
someone can say to me hoy only draw points?
PS: sorry john i dont see...
2006年9月1日, Esdras Caleb <ach...@gm...>:
> i see but when i put my point every time they are convered by aproximation
> in one line no in one expiremintal grafic..
>
> 2006年9月1日, John Hunter < jdh...@ac...>:
>
> > >>>>> "Esdras" == Esdras Caleb < ach...@gm...> writes:
> >
> > Esdras> no i gess i not clear i wnt draw_poits not plot a
> > Esdras> line. How i do it? when i plot the array the result is a
> > Esdras> line not a goup of poits like must be...
> >
> > I see -- take a look at the tutorial
> >
> > http://matplotlib.sourceforge.net/tutorial.html
> >
> > JDH
> >
>
>

The source error must of propagated to those builds. I will post new
ones shortly.
On 9/6/06, Sven Schreiber <sve...@gm...> wrote:
> Well the thread on the devel list that I referred to explicitly has the
> win32 exe in its title ("Missing __init__.py in
> matplotlib-0.87.5.win32-py2.4.exe ?") . The starting post there pretty
> much says it all.
>
> There's also a recent post on the numpy list that sounds like it's maybe
> the same problem (quote):
> """
> Is there a compatible matplotlib as well? I was o.k. with mpl from
> enthought until I switched numerix to numpy. That made mpl unhappy.
> I downloaded 0.87.5 but I broke something in the process because now
> even switching back to Numeric doesn't make mpl happy.
> """
>
> Thanks,
> Sven
>
> Charlie Moad schrieb:
> > That error was relating to the source release. Can you please post
> > your error for the binary?
> >
> > On 9/6/06, Sven Schreiber <sve...@gm...> wrote:
> >> Charlie Moad schrieb:
> >> > Minor rev bump for numpy 1.0b5 compatibility. This release should
> >> > remain compatible with future 1.0 releases of numpy.
> >> >
> >>
> >> I keep running into the ImportError problem described on the devel list,
> >> with the win32 2.4 binary (exe). Is that still just a case of waiting
> >> for the mirrors to update, or is there a deeper problem? Are there
> >> alternative download links? How to tell whether it's the "right" binary,
> >> since version numbers are the same?
> >>
> >> Thanks for your help,
> >> Sven
> >>
> >
>
>

Hi! (Background: I'm on a Mac G4 running OS 10.4.7, "Python 2.4.3 (#1, 
Apr 7 2006, 10:54:33) [GCC 4.0.1 (Apple Computer, Inc. build 5250)] on 
darwin", matplotlib version 0.87.4. I'm trying to solve some x-axis 
tick labeling problems I'm having using plot_date; I'm toying with the 
example date_demo_convert.py to try to learn what changes have what 
effect. I have the folder MPL_examples on my Desktop, and am running my 
modifications from a terminal therein.) In the terminal, I launch 
python, import matplotlib, type help(matplotlib.<matplotlib module 
member>) (it doesn't seem to matter which member I request help on) and 
instead of getting the doc displayed in the terminal, a new GUI instance 
of python is launched and displays the following window:
Nothing happens in the terminal, so I try to quit this new instance of 
python from the menu bar, and the terminal prints:
"Traceback (most recent call last):
 File "<stdin>", line 1, in ?
 File 
"/Library/Frameworks/Python.framework/Versions/2.4/lib/python2.4/site.py", 
line 327, in __call__
 import pydoc
 File 
"/Library/Frameworks/Python.framework/Versions/2.4//lib/python2.4/pydoc.py", 
line 57, in ?
 from collections import deque
ImportError: cannot import name deque"
the dock says the new instance of python is still running, and I have to 
<ctrl>-c in the terminal to terminate that new instance (which also 
terminates
the original non-GUI instance of python; if instead of trying to quit 
from the menu bar, I go straight to the terminal and type <ctrl>-c, I 
can terminate the new python instance without getting the exception 
traceback, but the terminal instance of python still quits).
Now, as if this all weren't strange enough, here's the kicker: all this 
_only_ seems to happen if I try to get help on matplotlib members when 
I'm running python from within the MPL_examples directory - if I do 
python, import matplotlib, help(matplotlib.<module member>) in a 
terminal based in another directory, I get the expected behavior, i.e., 
a text printout of the member's doc! (One thing I haven't tried, 'cause 
it doesn't _seem_ like it should make a difference, is placing the 
MPL_examples directory someplace different and then repeating the 
experiment.)
Is this a "bug"? Should I file a bug report?
DG

Well the thread on the devel list that I referred to explicitly has the
win32 exe in its title ("Missing __init__.py in
matplotlib-0.87.5.win32-py2.4.exe ?") . The starting post there pretty
much says it all.
There's also a recent post on the numpy list that sounds like it's maybe
the same problem (quote):
"""
Is there a compatible matplotlib as well? I was o.k. with mpl from
enthought until I switched numerix to numpy. That made mpl unhappy.
I downloaded 0.87.5 but I broke something in the process because now
even switching back to Numeric doesn't make mpl happy.
"""
Thanks,
Sven
Charlie Moad schrieb:
> That error was relating to the source release. Can you please post
> your error for the binary?
> 
> On 9/6/06, Sven Schreiber <sve...@gm...> wrote:
>> Charlie Moad schrieb:
>> > Minor rev bump for numpy 1.0b5 compatibility. This release should
>> > remain compatible with future 1.0 releases of numpy.
>> >
>>
>> I keep running into the ImportError problem described on the devel list,
>> with the win32 2.4 binary (exe). Is that still just a case of waiting
>> for the mirrors to update, or is there a deeper problem? Are there
>> alternative download links? How to tell whether it's the "right" binary,
>> since version numbers are the same?
>>
>> Thanks for your help,
>> Sven
>>
> 

That error was relating to the source release. Can you please post
your error for the binary?
On 9/6/06, Sven Schreiber <sve...@gm...> wrote:
> Charlie Moad schrieb:
> > Minor rev bump for numpy 1.0b5 compatibility. This release should
> > remain compatible with future 1.0 releases of numpy.
> >
>
> I keep running into the ImportError problem described on the devel list,
> with the win32 2.4 binary (exe). Is that still just a case of waiting
> for the mirrors to update, or is there a deeper problem? Are there
> alternative download links? How to tell whether it's the "right" binary,
> since version numbers are the same?
>
> Thanks for your help,
> Sven
>

Hello,
I'm creating a program that plots data in real-time. I want the
x-axis to be the time in HH:MM:SS. I'm still playing with the date
formatters, but have run into a problem where it seems the axis puts
in more points on the axis to make the plot fit nicely. For instance,
I want to have 120 data points (seconds) on the graph. Initially, the
x-axis starts at, say 0 and goes to 120 (I'd expect this to be 119).
The following second the x-axis should go from 1 to 120, then 2 to
121, etc. But instead, the x-axis goes from 0 to 140. Once 141
seconds have passed by, the axis shifts and goes from 20 to 160.
Is there any way to make the axis behave so that it keeps just 120
values on the axis and increment the values as described above?
Thanks!
-Roberto.

Thank you! I realized later how to do this when I realized that I 
hadn't looked at the code for the canvases.
R.
On Aug 29, 2006, at 12:08 AM, John Hunter wrote:
>>>>>> "Richard" == Richard Harvey Chapman <hchapman-matplotlib- 
>>>>>> us...@3g...> writes:
>
> Richard> Briefly, is there a way to get a callback in the event
> Richard> that the canvas has been redrawn (e.g. from a zoom or
> Richard> pan)?
>
> Richard> matplotlib 0.86.2 python 2.4.1 wxPython 2.6.2.1 Mac OS
> Richard> X.4.7
>
> Richard> I draw a circle at a certain point. I let the user drag
> Richard> that point to a new location. However, if the plot is
> Richard> redrawn for any reason, I have to redraw my circle in a
> Richard> new position. I can't seem to find any way to get
> Richard> notified of a redraw.
>
> Have you dried connecting to a draw_event?
>
> def ondraw(event):
> # do something
> pass
>
> fig.canvas.mpl_connect('draw_event', ondraw)
>
> FYI, the widgets module matplotlib.widgets is a good place to look for
> examples of mpl event handling, in addition to the class documentation
> in the Event hierarchy in
> http://matplotlib.sourceforge.net/matplotlib.backend_bases.html
>
> JDH
R.

Alan G Isaac wrote:
> Does anyone really care about 25% enough to make this 
> worthwhile? Just wondering.
I tend to think not. You put 80,000 points in a PS, it's going to be 
big. That's all there is to it, it's the nature of Postscript.
I do think clipping is a good idea though.
What is the maximum precision in Postscript? It seems unlikely that you 
could plot 80,000 points and not have number of them overlap, unless 
it's clipped, so removing essentially redundant points may be another 
way to to go.
-Chris
-- 
Christopher Barker, Ph.D.
Oceanographer
 		
NOAA/OR&R/HAZMAT (206) 526-6959 voice
7600 Sand Point Way NE (206) 526-6329 fax
Seattle, WA 98115 (206) 526-6317 main reception
Chr...@no...

On 9/6/06, Rob Hetland <he...@ta...> wrote:
>
> I have been having problems with postscript output from MPL on my
> various Macs that is just beyond the problem discussed in the tread
> here. But this seemed like the best place to bring it up. The basic
> problem is that Adobe products like Illustrator can't read MPL .eps
> output.
FWIW, I've had very good luck recently using inkscape
(http://www.inkscape.org/) to edit PS files, both old ones made by
who-knows-what and fresh mpl-off-svn ones. I've been able to fix old
labels that needed changes for figures whose data source is nowhere to
be found, change a line color and dashing/cap style, etc. I'm not a
graphics artist and have never used Illustrator, but for my needs,
inkscape has so far been an excellent (and free!) solution.
Note that this is on a Linux Ubuntu Dapper box, with the pstoedit
program installed, which inkscape uses to read *PS files into an
editable vector format.
Cheers,
f

On Wednesday 06 September 2006 11:04, Rob Hetland wrote:
> Well, the fonts are correct -- I checked in both the EPS source, and
> what Illustrator thinks the font is (after translation with pstopdf).
>
> The definition of the fonts looks right in the EPS source (checked by
> comparing against a sample from illustrator).
>
> If nobody else is having the same problem, typing pstopdf is really
> not a tremendous burden. I just thought I would raise this issue.
>
> BTW, SVG files seem to do fine with fonts. However, it can take a
> very long time to save large files in MPL and import large files into
> Illustrator, which seems to want to translate them into some native
> format, and does so poorly.
>
> If I ever figure out a better solution, I'll let you all know,
One of these days, we'll have to figure out how to embed individual glyphs 
rather than the entire font set.

Charlie Moad schrieb:
> Minor rev bump for numpy 1.0b5 compatibility. This release should
> remain compatible with future 1.0 releases of numpy.
> 
I keep running into the ImportError problem described on the devel list,
with the win32 2.4 binary (exe). Is that still just a case of waiting
for the mirrors to update, or is there a deeper problem? Are there
alternative download links? How to tell whether it's the "right" binary,
since version numbers are the same?
Thanks for your help,
Sven

Well, the fonts are correct -- I checked in both the EPS source, and 
what Illustrator thinks the font is (after translation with pstopdf).
The definition of the fonts looks right in the EPS source (checked by 
comparing against a sample from illustrator).
If nobody else is having the same problem, typing pstopdf is really 
not a tremendous burden. I just thought I would raise this issue.
BTW, SVG files seem to do fine with fonts. However, it can take a 
very long time to save large files in MPL and import large files into 
Illustrator, which seems to want to translate them into some native 
format, and does so poorly.
If I ever figure out a better solution, I'll let you all know,
-Rob
On Sep 6, 2006, at 8:46 AM, John Hunter wrote:
>>>>>> "Rob" == Rob Hetland <he...@ta...> writes:
>
> Rob> I have been having problems with postscript output from MPL
> Rob> on my various Macs that is just beyond the problem discussed
> Rob> in the tread here. But this seemed like the best place to
> Rob> bring it up. The basic problem is that Adobe products like
> Rob> Illustrator can't read MPL .eps output.
>
> Are you sure you have set the font paths properly *and* flushed your
> ~/.matplotlib/ttffont.cache before regenerating your files. Use
> --verbose-debug when creating your test EPS to make sure you are
> getting bitstream fonts. If you still have trouble, post an errant
> eps file somewhere and maybe someone can poke into it.
>
> Rob> I think that opening an EPS in Illustrator is the ultimate
> Rob> test of EPS format compliance, as those are the folks who
> Rob> made it.
>
> I think this is debatable, as it is a closed source implementation of
> an open document format specification. But I am willing to accept
> that it is more likely that Illustrator is right than that mpl is
> right.
>
> Rob> What can we do to make MPL output universally readable
> Rob> postscript?
>
> Find the bug and fix it :-)
>
> JDH
----
Rob Hetland, Associate Professor
Dept. of Oceanography, Texas A&M University
http://pong.tamu.edu/~rob
phone: 979-458-0096, fax: 979-845-6331

On Wed, 6 Sep 2006, John apparently wrote:=20
> could make it an rc param for those who want to trade=20
> accuracy for space.
Does anyone really care about 25% enough to make this=20
worthwhile? Just wondering.
Cheers,
Alan Isaac