matplotlib-users Mailing List for matplotlib

>>>>> "James" == James Landry <jwl...@gm...> writes:

    James> Maybe this is a dumb question, but I've installed tk-devel
    James> and tcl-devel.  Still doesn't work.  Do I need to reinstall
    James> matplotlib or do it from source?

Yes, after installing the dev headers you will need to rebuild
matplotlib from src.  I suggest you capture the output of the build
process is case you need to post it

> sudo rm -rf build
> sudo python setup.py install >& build.out

JDH

    James> James

    James> On 11/25/05, John Hunter <jdh...@ac...>
    James> wrote:
    >>  >>>>> "James" == James Landry <jwl...@gm...> writes:
    >> 
    James> I get the following message, when I try to use it, which I
    James> don't understand, because I have tkinter installed:
    >>  It means that matpltolib was not able to find the tk src code
    >> needed to compile the tkagg extension.  Usually this is because
    >> you haven't installed the tk-dev and tcl-dev packages.
    >> 
    >> JDH
    >> 



    James> -- James W. Landry jwl...@gm...

Maybe this is a dumb question, but I've installed tk-devel and tcl-devel.
Still doesn't work.  Do I need to reinstall matplotlib or do it from source=
?

James

On 11/25/05, John Hunter <jdh...@ac...> wrote:
>
> >>>>> "James" =3D=3D James Landry <jwl...@gm...> writes:
>
>     James> I get the following message, when I try to use it, which I
>     James> don't understand, because I have tkinter installed:
>
> It means that matpltolib was not able to find the tk src code needed
> to compile the tkagg extension.  Usually this is because you haven't
> installed the tk-dev and tcl-dev packages.
>
> JDH
>



--
James W. Landry
jwl...@gm...

>>>>> "James" == James Landry <jwl...@gm...> writes:

    James> I get the following message, when I try to use it, which I
    James> don't understand, because I have tkinter installed:

It means that matpltolib was not able to find the tk src code needed
to compile the tkagg extension.  Usually this is because you haven't
installed the tk-dev and tcl-dev packages.

JDH

I tried using tkagg by setting

backend  : TkAgg

in my matplotlibrc, which includes only the following lines:

backend         : TkAgg

# latex options for matplotlib

font.latex.package    : type1cm
text.usetex        : False
text.tex.engine        : latex

I get the following message, when I try to use it, which I don't understand=
,
because I have tkinter installed:

[jwlandry@merrimac flow]$ ~/grain/plots/hopper/pof-w-insert.py data_hopper-
40k-r1.5.funnel-pof.combo data_hopper-40k-r2.funnel-pof.combo data_hopper-
40k-r4.funnel-pof.combo data_hopper-40k-r6.funnel-pof.combo
Traceback (most recent call last):
  File "/home/jwlandry/grain/plots/hopper/pof-w-insert.py", line 12, in ?
    from pylab import *
  File "/usr/lib/python2.4/site-packages/pylab.py", line 1, in ?
    from matplotlib.pylab import *
  File "/usr/lib/python2.4/site-packages/matplotlib/pylab.py", line 217, in
?
    new_figure_manager, draw_if_interactive, show =3D pylab_setup()
  File "/usr/lib/python2.4/site-packages/matplotlib/backends/__init__.py",
line 24, in pylab_setup
    globals(),locals(),[backend_name])
  File
"/usr/lib/python2.4/site-packages/matplotlib/backends/backend_tkagg.py",
line 8, in ?
    import tkagg                 # Paint image to Tk photo blitter extensio=
n
  File "/usr/lib/python2.4/site-packages/matplotlib/backends/tkagg.py", lin=
e
1, in ?
    import _tkagg
ImportError: No module named _tkagg

James

On 11/25/05, John Hunter <jdh...@ac...> wrote:
>
> >>>>> "James" =3D=3D James Landry <jwl...@gm...> writes:
>
>     James> When it doesn't work, a dialog box opens with this message
>
>     James> Save figure failure: None: Interrupted system call
>
> What backend are you using?  I have seen something like this on GTK*.
> I'm pretty sure it is a threading problem.  Basically, we need a
> backend dependent thread-safe way to make system calls.  Not a trivial
> task.  I've posted a thread about this to the pygtk mailing list a few
> months ago, but have yet to find a reliable recipe.  gtk gurus, please
> advise.
>
> In the meantime, you may want to try a different GUI backend, eg,
> tkagg.  If this helps, let us know, because it will help determine
> that this is indeed your problem.
>
>
> JDH
>



--
James W. Landry
jwl...@gm...

>>>>> "Charlie" =3D=3D Charlie Moad <cw...@gm...> writes:

    Charlie> Can anybody think of a good way to disable a widget
    Charlie> (specifically SpanSelector) during plot interaction from
    Charlie> the toolbar?  It seems wrong when you are panning and the
    Charlie> SpanSelector is triggering over and over.  Is there a
    Charlie> trivial way to detect the state of the toolbar?

I think this will require some minor tweaks to backend_bases
NavigationToolbar2 and SpanSelector.  You'll want to modify
NavigationToolbar2 to make the mode "publicly accessible", eg a
get_mode method, which returns None, 'pan/zoom', or 'zoom-to-rect'
using the underlying _idPress, _idRelease, et al.  Then you'll want to
modify the span selector to be toolbar aware, eg, something along the
lines of

toolbar =3D getattr(self.ax.figure.canvas, 'toolbar')
if toolbar is not None:
    mode =3D toolbar.get_mode()
    if mode is None: suppress drawing....

The getattr check is required in case someone is using the span
selector in a canvas w/o the toolbar.
JDH

    Charlie> Thanks, Charlie


    Charlie> -------------------------------------------------------
    Charlie> This SF.Net email is sponsored by the JBoss Inc.  Get
    Charlie> Certified Today Register for a JBoss Training Course.
    Charlie> Free Certification Exam for All Training Attendees
    Charlie> Through End of 2005. For more info visit:
    Charlie> http://ads.osdn.com/?ad_idv28&alloc_id=16845&op=3Dclick
    Charlie> _______________________________________________
    Charlie> Matplotlib-users mailing list
    Charlie> Mat...@li...
    Charlie> https://lists.sourceforge.net/lists/listinfo/matplotlib-user=
s

>>>>> "Gerald" == Gerald John M Manipon <Ger...@jp...> writes:

    Gerald> Hello, I'm trying to plot some grid data using basemap and
    Gerald> imshow however haven't figured out how to specify the
    Gerald> color for nan values.  For example, in my data, valid
    Gerald> values range from 0.0-10.0.  I've assigned -9999.0 and
    Gerald> 1e30 to the missing values and when I imshow (setting
    Gerald> vmin=0.0, vmax=10.0), the missing values show up in the
    Gerald> low and high end of the spectrum, respectively.  Is there
    Gerald> a way to do this?

See the archives for a post by Danny Shevitz and followups in the
thread "obscure colormapping issue related to quantization"

http://article.gmane.org/gmane.comp.python.matplotlib.general/2856/match=obscure+colormapping+issue+related+to+quantization

Basically, you want to write a custom normalization and colormapping
instance, and recipe code is posted in the thread linked above.

JDH

>>>>> "Alan" == Alan Jackson <aja...@op...> writes:
    Alan> to show the failure, I thought I would ask to see of there
    Alan> are known incompatibilities between Linux and XP that might
    Alan> be causing my problem.

No, none that I am aware of.  I suggest you come up with that demo
script.


JDH

>>>>> "James" == James Landry <jwl...@gm...> writes:

    James> When it doesn't work, a dialog box opens with this message

    James> Save figure failure: None: Interrupted system call

What backend are you using?  I have seen something like this on GTK*.
I'm pretty sure it is a threading problem.  Basically, we need a
backend dependent thread-safe way to make system calls.  Not a trivial
task.  I've posted a thread about this to the pygtk mailing list a few
months ago, but have yet to find a reliable recipe.  gtk gurus, please
advise.

In the meantime, you may want to try a different GUI backend, eg,
tkagg.  If this helps, let us know, because it will help determine
that this is indeed your problem.


JDH

Thanks for the suggestion Jose.  I installed the new version and the same
problem appeared again.  I'm getting some ideas though.  When I tried it
again by uncommenting the rc line, it worked the first time but not the
second.  Doesn't it cache some of the information?  Could that be the
problem?

James Landry

On 11/25/05, Jos=E9 Matos <jao...@gm...> wrote:
>
> On 25/11/05, James Landry <jwl...@gm...> wrote:
> > I have intermittent problems with the savefig('foo.eps') command when I
> am
> > using my native LaTeX interpreter to produce eps files.  The really
> > frustrating thing is that it is intermittent.  Sometimes I can save a
> file
> > that looks fine and sometimes I can't.
> >
> >  My box is running Fedora Core 4 and kernel 2.6.14.
> >
> >  I have the following rpms:
> >  python-2.4.1-2
> >  ipython-0.6.15-1.fc4
> >  python-tools-2.4.1-2
> >  python-matplotlib-0.84-1.fc4
> >  python-numeric-23.7-2
>
>   Probably it is irrelevant to this problem and it does not help here
> but notice that 0.85 is available from extras.
>



--
James W. Landry
jwl...@gm...