matplotlib-users Mailing List for matplotlib

>>>>> "Chris" == Chris Barker <Chr...@no...> writes:

    Chris> Wonderful! this is great news. How have you done it? Have
    Chris> you written a DVI driver? or do you use TeX and associated
    Chris> tools to create a bitmap and put that on the image? Or
    Chris> something else?

For *Agg, we use tex + dvipng and load the output as a transparent,
anti-aliased raster, caching the dvipng output in ~/.tex.cache for
efficiency.  For PS, we use tex + dvips + psfrag + latex + a patched
ps2epsi.  The latter is cumbersome, but it works.  In an ideal world,
we would simply use tex + dvips but embedding the postscript fragments
generated by dvips in arbitrary locations on the figure, but this has
proved challenging.

Having our own dvi parser would be a nice solution.  As you noted, the
format isn't too complicated, except for the font handling and the
fact that you need your own bytecode processing engine to extract the
information.  We're waiting until Robert Kern has another boring
experiment to site through for this one.

JDH

John Hunter wrote:
> In the next release of matplotlib, due out soon, we have support for
> mathematical expressions using TeX in the *Agg and PS backends.

Wonderful! this is great news. How have you done it? Have you written a 
DVI driver? or do you use TeX and associated tools to create a bitmap 
and put that on the image? Or something else?

-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 Wednesday 01 June 2005 17:00, John Hunter wrote:
> Currently, matplotlib mathtext handling does not allow "embedded" math
> expressions.  So the whole expression must be written in math mode.
> This is a pain and something we will fix when we can.
>

Okay.

> In the next release of matplotlib, due out soon, we have support for
> mathematical expressions using TeX in the *Agg and PS backends.  So if
> you have tex (and a few other tools) installed on your systems, you
> can use it to render all the text in your matplotlib figure.
>
> In the meantime, you can do something like this
>
>   title(r'$x\ \rm{and}\ y$')
>

Yes, that works fine here ! Many thanks for the tip !

Cheers,
Nicolas

>>>>> "Nicolas" == Nicolas Girard <nic...@ne...> writes:

    Nicolas> Hi all, I need to mix maths & text in labels such as
    Nicolas> title and xlabel.

    Nicolas> I first tried the most intuitive way:


    Nicolas>   title(r"$x$ and $y$")

    Nicolas> but it appeared not to work (don't focus of my silly
    Nicolas> sample title, the real one does make sense ;-)

    Nicolas> Then I tried

    Nicolas>   title(r"$x$ \text{and} $y$")

    Nicolas> but I got an error related to the fact that
    Nicolas> is_string_like() returned False when passing my title
    Nicolas> string to it.


Currently, matplotlib mathtext handling does not allow "embedded" math
expressions.  So the whole expression must be written in math mode.
This is a pain and something we will fix when we can.

In the next release of matplotlib, due out soon, we have support for
mathematical expressions using TeX in the *Agg and PS backends.  So if
you have tex (and a few other tools) installed on your systems, you
can use it to render all the text in your matplotlib figure.

In the meantime, you can do something like this

  title(r'$x\ \rm{and}\ y$')

JDH

Hi all,
I need to mix maths & text in labels such as title and xlabel.

I first tried the most intuitive way: 


  title(r"$x$ and $y$")

but it appeared not to work (don't focus of my silly sample title, the real 
one does make sense ;-)

Then I tried

  title(r"$x$ \text{and} $y$")

but I got an error related to the fact that is_string_like() returned False 
when passing my title string to it.

Could you please tell me how to achieve it ?

Thanks in advance,
cheers,
Nicolas

Has anybody considered writing a epix backend using pyepix? 
(http://claymore.engineer.gvsu.edu/~steriana/Python/pyepix/index.html).

A cursory inspection shows that it might solve many of the problems with 
mathematical typesetting that the psfrag backend was invented to solve.

-Jeff

-- 
Jeffrey S. Whitaker         Phone : (303)497-6313
NOAA/OAR/CDC  R/CDC1        FAX   : (303)497-6449
325 Broadway                Web   : http://www.cdc.noaa.gov/~jsw
Boulder, CO, USA 80305-3328 Office: Skaggs Research Cntr 1D-124

[...]

>> On 5/31/05, *John Gill* <jn...@eu... 
>> <mailto:jn...@eu...>> wrote:
>>
>>     I think this might be a matplotlib issue.
>>
>>     I've been having some refresh issues with matplotlib since
>>     switching to 2.6.  In particular I've noticed that if I put my
>>     mouse over the toolbar and let a tooltip pop up the canvas does
>>     not get repainted.
>>
>>     Up to now I'd been assuming this problem was isolated to me (I use
>>     ratpoison as a window manager, which can confuse some apps).
>>
>>     I've cc'ed the matplotlib list in case anyone can throw some light
>>     on this.
>>
>>     John
>>
[...]

> I can confirm this (mis)behaviour with matplotlib 0.80 and gtk+/pygtk 
> 2.6.  I had wanted to look into this matter, but didn't yet have time 
> to do it.  A simple guess would be that the expose-event is not 
> handled properly by the GTK backend.
>
> I noticed the problem when I put a menu-bar over a matplotlib GTK 
> Canvas.  When you leave an opened menu, the menu will not disappear.
>
> Niklas.
>

I have started looking at the code and did not quite understand what I 
found:

backends/backend_gtk.py has a class FigureCanvasGTK. Its method 
'expose_event' is responsible for redrawing.
In the situation I described, when I leave an opened menu that overlaps 
the graph, the expose_event is properly called, but the widget is not 
redrawn. If DBL_BUFFER is set (which it is), the widget is only redrawn, 
if self._draw_pixmap is set to True. Of course, this variable is not set 
to True when GTK closes a menu but only when matplotlib requests a 
redraw.  So, in conclusion, no redraw happens.

After consulting the PyGTK reference I tried to reactivate double 
buffering, which is turned of in the backend_gtk.py by the command

self.set_double_buffered(False)

I wonder why you set this? Do you want to provide your own double 
buffering? Why? If you set this value to True, everything will be fine.

Best regards,

Niklas Volbers.

Hello everyone,

I am trying to implement object picking in my matplotlib-based
plotting application.  I found two examples.  The 'picker_demo.py'
demo is very nice and looks a lot simpler than the 'object_picker.py'
which requires you to use your own Canvas object.  But since
object_picker needs to have the event.inaxes object, it seems
impossible to catch any events outside the axes' bounding box, like
'click on axes title', or 'click on ticks' or 'click on xlabel'.

Is there any possibility to do so ?

BTW, is it possible to detect double-clicks?

Best regards,

Niklas.

dimitri pater schrieb:

> Yes, this could could very well be a matplotlib issue. I will try to 
> do some tests, thanks for cc'ing it to the matplotlib list.
>
> Dimitri
>
> On 5/31/05, *John Gill* <jn...@eu... 
> <mailto:jn...@eu...>> wrote:
>
>     I think this might be a matplotlib issue.
>
>     I've been having some refresh issues with matplotlib since
>     switching to 2.6.  In particular I've noticed that if I put my
>     mouse over the toolbar and let a tooltip pop up the canvas does
>     not get repainted.
>
>     Up to now I'd been assuming this problem was isolated to me (I use
>     ratpoison as a window manager, which can confuse some apps).
>
>     I've cc'ed the matplotlib list in case anyone can throw some light
>     on this.
>
>     John
>
>     dimitri pater wrote:
>
>>     Hello,
>>     I upgraded to PyGTK 2.6 from 2.4 . Now, when I switch from page 2
>>     on a Gtk.Notebook back to page 1, page 1 still shows some
>>     elements from page 2 (it is not refreshed, just some parts). Both
>>     pages contain graphs created with Matplotlib. I never had
>>     problems like this with 2.4...
>>     Any clues somebody? If necessary, I will attach the source files.
>>
>>     Best regards,
>>     Dimitri
>>
>>------------------------------------------------------------------------
>>
>>_______________________________________________
>>pygtk mailing list   
>>py...@da... <mailto:py...@da...>
>>http://www.daa.com.au/mailman/listinfo/pygtk
>>Read the PyGTK FAQ: 
>>http://www.async.com.br/faq/pygtk/
>>  
>>
>
>
>
> -- 
> Please visit dimitri's website: www.serpia.com <http://www.serpia.com> 

I can confirm this (mis)behaviour with matplotlib 0.80 and gtk+/pygtk 
2.6.  I had wanted to look into this matter, but didn't yet have time to 
do it.  A simple guess would be that the expose-event is not handled 
properly by the GTK backend.

I noticed the problem when I put a menu-bar over a matplotlib GTK 
Canvas.  When you leave an opened menu, the menu will not disappear.

Niklas.

On May 31, 2005, at 2:37 PM, Chris Barker wrote:
> great! thanks for picking up the mantle on this. I'd been meaning to 
> get to it, but you know how it is.

Thank you for providing such excellent information on the build 
process.  Picking up where you left off was a snap.

> Does this version work with TK? I can't remember what I put in the 
> notes with the one I sent out.

No, I didn't have Tk installed at the time.  I'll remedy that situation 
and put up a new installer package in the same place.

Ken

I have a question about figure size in pylab, which presently includes the=
=20
toolbar. I think figure(figsize=3D(3,3)) should create a 3x3in plot window,=
 and=20
append the toolbar below it. The effect is more noticeable on the QtAgg and=
=20
WXAgg backends, where the coordinates are reported below the toolbar. Is th=
is=20
a bug or a feature?=20

Darren

On May 31, 2005, at 6:55 PM, Stephen Walton wrote:
> Ken McIvor wrote:
>>     http://agni.phys.iit.edu/~kmcivor/potpourri/wxmpl-demos.py
>
> An interesting looking script.  But 'import wxmpl' fails on my system 
> (MPL V0.80).  Am I missing something?

It's missing the `wxmpl' module, a library for rapidly embedding 
matplotlib in wxPython.  I have not yet had the gumption to proselytize 
it.  I hadn't intended for people to try to run the whole script, just 
to look at the plot_XZY() functions.

> As an aside, it would be nice if a demo/tutorial script like this 
> could be made backend-independent.  I normally use TkAgg in preference 
> to WX or WXAgg.

All of the aforementioned plot_XZY() functions accept a Figure as their 
only argument and draw their plots onto the figure, so they're already 
backend independent.  I have attached a script which uses 
`pylab.figure()' to handle the creation of the Figure and associated 
FigureCanvas.  Run it from the command line, and it will present you 
with a list of demos.  If your matplotlibrc isn't setup for interactive 
plotting via wxPython, Tkinter, or GTK you'll need to specify an 
appropriate backend to use:

	$ python oo-demos.py -dWXAgg

Ken

Stephen Walton wrote:
 > Ken McIvor wrote:
 >
 >
 >>    http://agni.phys.iit.edu/~kmcivor/potpourri/wxmpl-demos.py
 >
 >
 > An interesting looking script.  But 'import wxmpl' fails on my system
 > (MPL V0.80).  Am I missing something?

http://agni.phys.iit.edu/~kmcivor/wxmpl/

(a bit of poking around recursively up his directories :)

Best,

f