matplotlib-users Mailing List for matplotlib

>>>>> "Parekh," == Parekh, Hardik (IndSys, GE Interlogix) <har...@ge...> writes:
 Parekh> Hi, I'm pretty impressed with the abilities of
 Parekh> matplotlib. But, I want to do dynamic plotting in
 Parekh> python. So does this tool support dynamic plotting? I
 Parekh> didn't see any snapshots or demo for dynamic plotting.
 Parekh> Please let me know if I can do dynamic plotting or not,
 Parekh> and if I can, how? Is there any tutorial on it?
Hi,
Thanks for the email. In the future if you post the question to the
mailing list I'll be more likely to see it quickly and others can help
you as well
http://lists.sourceforge.net/mailman/listinfo/matplotlib-users.
Now, on to your question. Yes, matplotlib does support dynamic
plotting, if I understand your question correctly. You can update the
graphs in an animated fashion using either the gtk timer or idle
functionality.
See examples/system_monitor.py in the src distribution. Let me know
if you need something other than this functionality, and I'll see
about getting it incorporated.
John Hunter

Gary Ruben wrote:
>Hi John,
>Just pre-empting Nathan's reply, a search revealed this:
><http://www.cairographics.org/>
>The reason I'm replying is to note that I'd vote for SVG support ahead of the currently X-specific Cairo. In fact I'd support WPG and WMF/EMF ahead of Cairo based on portability arguments. Yes, I'm primarily a Win32 user. Nathan may know something about how much simpler it would be to create a Cairo backend than an SVG backend which may argue against this.
>
Cairo is intended to have an svg backend because people want to be able 
to screenshot ->svg and print->svg and so on. Also, cairo is an 
immediate mode system, suitable for interactive things, svg isn't. 
Having said that, either would be good, and svg would have greater value 
at the moment (cairo will be fast, but it ain't at the moment).
njh

John Hunter wrote:
>>>>>>"Nathan" == Nathan Hurst <nj...@ha...> writes:
>>>>>> 
>>>>>>
>
> Nathan> has anyone investigated making a cario backend for
> Nathan> matplotlib? I'd do it myself (due to interests sake), but
> Nathan> I have a supervisor breathing down my neck...
>
>I'm not familiar with it, and a quick google didn't turn up anything
>that looked relevant. Do you have a link?
>
www.cairographics.org
the python-cairo interface works too.
njh

Hi John,
Just pre-empting Nathan's reply, a search revealed this:
<http://www.cairographics.org/>
The reason I'm replying is to note that I'd vote for SVG support ahead of the currently X-specific Cairo. In fact I'd support WPG and WMF/EMF ahead of Cairo based on portability arguments. Yes, I'm primarily a Win32 user. Nathan may know something about how much simpler it would be to create a Cairo backend than an SVG backend which may argue against this.
Gary
----- Original Message -----
From: John Hunter <jdh...@ac...>
Date: 2003年10月29日 15:58:38 -0600
To: Nathan Hurst <nj...@ha...>
Subject: Re: [Matplotlib-users] Cairo backend
> >>>>> "Nathan" == Nathan Hurst <nj...@ha...> writes:
> 
> Nathan> has anyone investigated making a cario backend for
> Nathan> matplotlib? I'd do it myself (due to interests sake), but
> Nathan> I have a supervisor breathing down my neck...
> 
> I'm not familiar with it, and a quick google didn't turn up anything
> that looked relevant. Do you have a link?
> 
> JDH
> 
> 
> 
> 
> 
> -------------------------------------------------------
> This SF.net email is sponsored by: SF.net Giveback Program.
> Does SourceForge.net help you be more productive? Does it
> help you create better code? SHARE THE LOVE, and help us help
> YOU! Click Here: http://sourceforge.net/donate/
> _______________________________________________
> Matplotlib-users mailing list
> Mat...@li...
> https://lists.sourceforge.net/lists/listinfo/matplotlib-users
-- 
__________________________________________________________
Sign-up for your own personalized E-mail at Mail.com
http://www.mail.com/?sr=signup
CareerBuilder.com has over 400,000 jobs. Be smarter about your job search
http://corp.mail.com/careers

>>>>> "Juraj" == Juraj Krempasky <jur...@ps...> writes:
 Juraj> the afm files were missing probably because I made an rpm
 Juraj> package with python setup.py which I installed (I didn't do
 Juraj> the setup.py install). Nevertheless the PS output by
 Juraj> setting the AFMPATH works fine. I really appreciate the
 Juraj> font management with matplotlib. However, for making plots
 Juraj> to be included in Latex I'd like to have the possibility to
 Juraj> scale also the axes fonts. For example Gnuplot knows a
 Juraj> command "set size" which then scales whole plot by leaving
 Juraj> all fonts the same - and this makes then small plots look
 Juraj> good. In matplotlib I can scale text, but I don't know how
 Juraj> to scale axes fonts along with axis labels. Is it possible?
Yes, this is possible. You need to get a list of the text handles and
then set the text properties. I'm away from my desk so I can't test
the code below, but here is the basic idea
# handle graphicsversion
t = gca().get_xticklabels()
set(t, 'fontsize', 12)
# or you can use object oriented interface
ax = subplot(211)
plot(something)
t = ax.get_xticklabels()
for label in t:
 t.set_fontsize(12)
 t.set_fontweight('bold')
I have been planning on modifying the axes constructor to take a font
dictionary (see help(text) for more info on font dictionaries) which
changes the default fonts for the axes, but haven't gotten to it.
But the approach above should work for you. See the tutorial on the
web site for more info on setting text properties, and let me know if
you run into any troubles.
Good luck,
John Hunter

>>>>> "Nathan" == Nathan Hurst <nj...@ha...> writes:
 Nathan> has anyone investigated making a cario backend for
 Nathan> matplotlib? I'd do it myself (due to interests sake), but
 Nathan> I have a supervisor breathing down my neck...
I'm not familiar with it, and a quick google didn't turn up anything
that looked relevant. Do you have a link?
JDH

has anyone investigated making a cario backend for matplotlib?
I'd do it myself (due to interests sake), but I have a supervisor 
breathing down my neck...
njh

I just set up a new mailing list for discussion related to the
development of matplotlib. Please join if interested.
John Hunter

>>>>> "Juraj" == Juraj Krempasky <jur...@ps...> writes:
 [ Standard response: please post questions to the mailing list
 when possible since others can benefit from the discussion, and
 provide help
 https://lists.sourceforge.net/lists/listinfo/matplotlib-users 
 ]
 Juraj> I suppose the above matplotlib.use('GTK') should be
 Juraj> matplotlib.use('PS'), isn't it?
Yes, that's correct. Thanks for catching it.
 Juraj> But then I got the following message:
 Juraj> Error: Could not find any AFM files; please set AFMPATH to
 Juraj> point to some readable adobe font metric files
Matplotlib comes with a set of AFM files. Did you do a normal 
 > python setup.py install
with matplotlib? If so, distutils should put the files in a place
where matplotlib can find them even w/o setting AFMPATH (eg,
/usr/local/share/matplotlib or /usr/share/matplotlib). If they are
not there, then my guess is something went funny with your install.
These files reside in the src distribution in the dir
matplotlib/fonts/afm, so you can get them there and set AFMPATH
accordingly.
The easiest way to find the AFM fonts on your system is to do
 > locate afm | less 
and see which dirs pop up. On my system, dirs include
/usr/share/fonts/default/Type1, /usr/share/fonts/afms/adobe, etc...
They you can set your AFMPATH to point to these dirs.
 Juraj> Any idea where there files are on Linux RH7.3. The size is
 Juraj> then defined again like savefig('xx.ps',dpi=60)?
You don't need to set the DPI for a postscript file since postscript
is scalable. The DPI is determined by your output device, eg, a
1200x1200 DPI printer. This parameter is ignored by matplotlib on the
PS backend.
Good luck,
John Hunter

>>>>> "Samir" == Samir Patel <sp...@la...> writes:
 Samir> PS solution works for me. One thing is that afm fonts are
 Samir> in gnome-print package (Took me some time to find where
 Samir> this fonts came from). Now I am trying to see whether I can
 Samir> make it work with Quixote (web server). I will try GD
 Samir> solution later on. Thanks for your suggestions. - Samir
matplotlib also ships with some AFM fonts which are put in a system
dir by the setup.py script. They reside in the fonts/afm dir of the
matplotlib src distribution. 
I'm curious, postscript is usually not such a great format for serving
web images. Do you plan on using a PS->PNG converter on the server
side? By the way, if you are on linux, the GD setup is fairly
painless if you follow the steps in the "Quick install guide for GD
output" on http://matplotlib.sourceforge.net/backends.html. I need to
generate some server side graphics myself -- that was one of my main
incentives for porting to GD.
Good luck -- if you have a publicly accessible web page serving up
matplotlib graphics in the future, let me know as I'd be interested in
checking it out.
JDH

First of all, one of the easiest graph package yet fill with tons of
capability and future potential.
Now my issue:
1. I am trying to integrate matplot with quixote. It fails because of gtk (I 
think). How can I use matplot without gui requirement.
2. I am trying to save an image without showing it. Here is small program
**************************************************************
from matplotlib.matlab import *
plot([1,2,3])
savefig('test.png')
#show()
***************************************************************
This does not create test.png file, but if I uncomment last line, it create
test.png file. How can I create a test.png file without showing it?
_________________________________________________________________
Enjoy MSN 8 patented spam control and more with MSN 8 Dial-up Internet 
Service. Try it FREE for one month! http://join.msn.com/?page=dept/dialup

Commented out line 43 in _matlab_helpers fix my issue. Is it because I don't 
have any window to destroy?
Here is small example (which creates problem without commenting line 43)
from matplotlib import use
use('GD')
from matplotlib.matlab import *
bar([0,1], [0,2])
savefig('test.png', dpi=100)
close('all')
bar([0,1], [1,1])
savefig('test1.png', dpi=100)
close('all')
********************New destroy function in _matlab_helpers (Need a fix if you 
have window to destroy 
*******************
 def destroy(num):
 if not GcfBase.has_fignum(num): return
 figwin = GcfBase.figs[num]
 oldQue = GcfBase._activeQue[:]
 GcfBase._activeQue = []
 for f in oldQue:
 if f != figwin: GcfBase._activeQue.append(f)
 del GcfBase.figs[num]
 #figwin.window.destroy()
 destroy = staticmethod(destroy)

http://biggles.sourceforge.net/libplot/
Just FYI.
njh

>>>>> "Samir" == Samir Patel <sp...@la...> writes:
Please email the list directly or CC it so others can benefit from the
discussion. For reason unclear to me, my posts to the list are held
up in some nebulous zone for many hours....
 Samir> I created image with GD library. I had 2 small issues: I
 Samir> add GDFONTPATH and while running from web server TTFQuery
 Samir> wanted to create a font
 Samir> "./usr/lib/python2.3/site-packages/ttfquery/font.cache" and
 Samir> it did not had permission. I simply created this file with
 Samir> write permission.
It looks like you are running in an environment where HOME is not set.
The TTFQuery file _scriptregistry.py first checks for the HOME
environment variable and tries to put the font cache there. If it
fails, it falls back on a python dir (which works well for linux and
win32 respectively). Also, make sure you are using version 0.2.6 of
TTFQuery.
So either 1) set HOME to point to where you want to font cache to go,
or 2) Edit _scriptregistry.py and manually set the registryFile
variable at the top of that script.
 Samir> Unfortunately I cannot put my stuff on public internet
 Samir> because I am working on my development machine which is
 Samir> inside my companies firewall and I have to create some good
 Samir> graphics forms for user to use it. I am planning to do both
 Samir> things withing a month or so. But I will be glad to
 Samir> transfer whatever knowledge and programs I have to start
 Samir> you on publishing graph on web.
I understand. If you have a publicly viewable screenshot that you can
share for the screenshot page, let me know. Otherwise, just pass on
what you learn about the mechanics of getting it up and running.
Thanks,
John Hunter

>>>>> "Samir" == Samir Patel <sp...@la...> writes:
 Samir> I am trying to save an image without showing it. Here is
 Samir> small program
There are a couple of ways to do it. It would help to know what
platform you are on. On linux/unix, there are the postscript and GD
backends which can generate images 'offline' w/o showing them. At
present, there is no way to generate a GTK image (GTK is the default
backend) w/o showing it. You should read the web pages
 http://matplotlib.sourceforge.net/installing.html
 http://matplotlib.sourceforge.net/backends.html
for more information on these issues.
Postscript is a high quality output that is more useful on linux/unix
platforms where there are usually viewers, converters and printers by
default. GD can be used to make PNGs and other image formats without
showing the image, but as far as I know there is no good windows
installer for gd module yet. I've spent some time working on it, and
have been corresponding with the author of gdmodule, and can probably
come up with one in the near future.
So that's the long answer. The short answer is -- if you are on
linux, install gd module following the instructions on the matplotlib
web site and use the GD or PS backend. If you are on windows, wait a
little bit and I'll see about getting the gd installer going, which is
something I've been meaning to do anyway.
Cheers,
John Hunter

PS solution works for me. One thing is that afm fonts are in gnome-print 
package (Took me some time to find where this fonts came from). Now I am 
trying to see whether I can make it work with Quixote (web server). I will 
try GD solution later on. Thanks for your suggestions.
- Samir

Samir,
Either call it with "python test.py -dGD" or use this code, with 2 new
lines at the beginning.
from matplotlib import use
use('GD')
from matplotlib.matlab import *
plot([1,2,3])
savefig('test.png')
----------------
Instructions for installing stuff for GD on the website. But if you just
need an image, do -dPS or use('PS') for postscript output.
	-C
 
--
Ask not what your computer can do for you; ask.... [ Uh-oh. ]

First of all, one of the easiest graph package yet fill with tons of 
capability and future potential. 
Now my issue:
I am trying to save an image without showing it. Here is small program
**************************************************************
from matplotlib.matlab import *
plot([1,2,3])
savefig('test.png')
#show()
***************************************************************
This does not create test.png file, but if I uncomment last line, it create 
test.png file. How can I create a test.png file without showing it? 

>>>>> "Charles" == Charles R Twardy <ct...@ma...> writes:
 Charles> That's installed now, but python figtext2.py -dGD makes
 Charles> it complain about GDFONTPATH even though
Mike Fletcher was kind enough to apply the ttfquery patch and release
TTFQuery-0.2.6 today, so GD users should grab that version from
http://sourceforge.net/projects/ttfquery to enable nonstandard true
type fonts.
JDH

>>>>> "Barry" == Barry Drake <bl...@ad...> writes:
 Barry> Thanks John. That fixed the hanging process. Now a
 Barry> message is coming back "none active" from the function:
 Barry> I haven't traced it back further yet.
Just kill the line -- that was a diagnostic debug message that should
have been removed before the release. 
Andrew Straw noted that a change I made in the axes module in 0.30
introduced a performance hit on his system. If you are experiencing
similar problems, I also fixed this. plots should load approx
twice as fast with the changes in CVS. 
JDH

Thanks John. That fixed the hanging process.
Now a message is coming back "none active" from the
function:
 def _get_active(self):
 if len(GcfBase._activeQue)==0:
 print 'None active'
 return None
 else: return self._activeQue[-1]
in the _matlab_helpers.py file. 
I haven't traced it back further yet.
Cheers.
Barry
--- John Hunter <jdh...@ac...> wrote:
> >>>>> "Barry" == Barry Drake <bl...@ad...>
> writes:
> 
> Barry> I'm observing some difference in behavior
> between the two
> Barry> versions. Running some example code in
> 0.29, after closing
> Barry> the plot windows, the python.exe process
> quits normally.
> Barry> The same example using 0.30 hangs the
> process and I have to
> Barry> kill it by hand. This occurs both at the
> command line and
> Barry> from within my editor (normally I execute
> from the editor).
> 
> Barry> Any ideas? Do I need to change some set
> up from when I was
> Barry> using 0.29?
> 
> Yes. I made several changes to this part of
> matplotlib to iron out
> some interactive bugs, and a new one crept in; I
> didn't notice it in
> my tests. The quick fix is to add the following
> code to
> matplotlib/backends/backend_gtk.py at line 773
> 
> if GcfGTK.get_num_figwins()==0 and
> ShowOn().is_mainloop_on():
> gtk.mainquit()
> 
> This is in the GcfGTK.destroy function, so the
> entire function should
> read
> 
> def destroy(num):
> if not GcfBase.has_fignum(num): return 
> figwin = GcfGTK.figs[num]
> figwin.figure.drawable= None
> GcfBase.destroy(num)
> if GcfGTK.get_num_figwins()==0 and
> ShowOn().is_mainloop_on():
> gtk.mainquit()
> destroy = staticmethod(destroy)
> 
> This should restore the old functionality -- let me
> know if you see
> anymore strangeness. I'll check the change into
> CVS.
> 
> [BTW, I posted the 0.30 announce to this mailing
> list this afternoon,
> but it didn't seem to make it. Hmmm.... Have I
> been moderated out of
> existence :-) ]
> 
> Thanks for the bug report, 
> John Hunter

Thanks to all (particularly Andrew and Charles) who contributed code
and bug reports leading to this release. Since my last post this
morning, I've cleared up several bugs in log scaling and interactive
mode. Definitely upgrade!
[Below is the general announcement]
matplotlib is a 2D plotting package for python with a matlab
compatible syntax and output tested under linux and windows platforms.
matplotlib-0.30 is available for download at
http://matplotlib.sourceforge.net, and has many new features since the
last major release.
Multiple outputs
 matplotlib now supports postscript and GD output, as well as the
 traditional GTK backend. The postscript backend goes a long way
 towards the goal of acheiving publication quality output. The GD
 backend allows you to use matplotlib even in environments with no X
 server, such as for a web application server serving dynamic charts.
Log scaling
 
 With the help of Andrew Straw, matplotlib now has log axis
 capabilities, with new commands semilogx, semilogy and loglog
 See http://matplotlib.sourceforge.net/screenshots.html#log_shot
Legends
 With the help of Charles R. Twardy, matplotlib now has a matlab
 compatible legend command. See
 http://matplotlib.sourceforge.net/screenshots.html#legend_demo
Numerous bug fixes and minor additions
 DPI parameter allows multiple output resolutions with correct
 scaling
 
 Several bug fixes in GTK interactive mode using
 examples/interactive2.py
 Multiple ways to specify colors, including matlab compatible format
 strings, RGB tuples, and html-style hex color strings
 Rewrite of line class for much greater compatibility with matlab
 handle graphics commands and flexibility in choosing line styles and
 markers
See http://matplotlib.sourceforge.net/screenshots.html for screenshots
and http://matplotlib.sourceforge.net/whats_new.html for more detailed
information on what's new.
John Hunter

[This didn't make it through when I sent it this afternoon. Not sure
why; sorry if you get a duplicate. ]
matplotlib-0.30 is available for download at
http://matplotlib.sourceforge.net, and has many new features since the
last major release.
Multiple outputs
 matplotlib now supports postscript and GD output, as well as the
 traditional GTK backend. The postscript backend goes a long way
 towards the goal of acheiving publication quality output. The GD
 backend allows you to use matplotlib even in environments with no X
 server, such as for a web application server serving dynamic charts.
Log scaling
 
 With the help of Andrew Straw, matplotlib now has log axis
 capabilities, with new commands semilogx, semilogy and loglog
 See http://matplotlib.sourceforge.net/screenshots.html#log_shot
Legends
 With the help of Charles R. Twardy, matplotlib now has a matlab
 compatible legend command. See
 http://matplotlib.sourceforge.net/screenshots.html#legend_demo
Numerous bug fixes and minor additions
 DPI parameter allows multiple output resolutions with correct
 scaling
 
 Several bug fixes in GTK interactive mode using
 examples/interactive2.py
 Multiple ways to specify colors, including matlab compatible format
 strings, RGB tuples, and html-style hex color strings
 Rewrite of line class for much greater compatibility with matlab
 handle graphics commands and flexibility in choosing line styles and
 markers
See http://matplotlib.sourceforge.net/screenshots.html for screenshots
and http://matplotlib.sourceforge.net/whats_new.html for more detailed
information on what's new.
John Hunter
Hi Nigel,
There has been a discussion on the scipy mailing list about a desire
to include your matfile in the scipy distribution. There is a
problem, however, in that scipy is distributed under a BSD compatible
license and the maintainers are hesitant to put GPL'd code in for fear
that it would hinder adoption by the business community. Would you be
willing to release a version of your matfile under a BSD/Python
compatible license for inclusion in scipy? There is currently support
for matfile version4 , but version 5 would be wonderful.
BTW, I have also done some work in matlab/python compatibility in a
plotting library I developed http://matplotlib.sourceforge.net.
Thought you might be interested.
John Hunter

>>>>> "Barry" == Barry Drake <bl...@ad...> writes:
 Barry> I'm observing some difference in behavior between the two
 Barry> versions. Running some example code in 0.29, after closing
 Barry> the plot windows, the python.exe process quits normally.
 Barry> The same example using 0.30 hangs the process and I have to
 Barry> kill it by hand. This occurs both at the command line and
 Barry> from within my editor (normally I execute from the editor).
 Barry> Any ideas? Do I need to change some set up from when I was
 Barry> using 0.29?
Yes. I made several changes to this part of matplotlib to iron out
some interactive bugs, and a new one crept in; I didn't notice it in
my tests. The quick fix is to add the following code to
matplotlib/backends/backend_gtk.py at line 773
 if GcfGTK.get_num_figwins()==0 and ShowOn().is_mainloop_on():
 gtk.mainquit()
This is in the GcfGTK.destroy function, so the entire function should
read
 def destroy(num):
 if not GcfBase.has_fignum(num): return 
 figwin = GcfGTK.figs[num]
 figwin.figure.drawable= None
 GcfBase.destroy(num)
 if GcfGTK.get_num_figwins()==0 and ShowOn().is_mainloop_on():
 gtk.mainquit()
 destroy = staticmethod(destroy)
This should restore the old functionality -- let me know if you see
anymore strangeness. I'll check the change into CVS.
[BTW, I posted the 0.30 announce to this mailing list this afternoon,
but it didn't seem to make it. Hmmm.... Have I been moderated out of
existence :-) ]
Thanks for the bug report, 
John Hunter

I installed the latest release 0.30 replacing 0.29,
after uninstalling 0.29.
I'm observing some difference in behavior between the
two versions. Running some example code in 0.29,
after closing the plot windows, the python.exe process
quits normally. The same example using 0.30 hangs the
process and I have to kill it by hand. This occurs
both at the command line and from within my editor
(normally I execute from the editor).
I'm using Windows XP pro and UltraEdit.
Any ideas? Do I need to change some set up from when
I was using 0.29?
Thanks.
Barry
--- John Hunter <jdh...@ac...> wrote:
> 
> matplotlib is a 2D plotting package for python with
> a matlab
> compatible syntax and output tested under linux and
> windows platforms.
> 
> matplotlib-0.30 is available for download at
> http://matplotlib.sourceforge.net, and has many new
> features since the
> last major release.
> 
> Multiple outputs
> 
> matplotlib now supports postscript and GD output,
> as well as the
> traditional GTK backend. The postscript backend
> goes a long way
> towards the goal of acheiving publication quality
> output. The GD
> backend allows you to use matplotlib even in
> environments with no X
> server, such as for a web application server
> serving dynamic charts.
> 
> Log scaling
> 
> With the help of Andrew Straw, matplotlib now has
> log axis
> capabilities, with new commands semilogx,
> semilogy and loglog
> See
>
http://matplotlib.sourceforge.net/screenshots.html#log_shot
> 
> Legends
> 
> With the help of Charles R. Twardy, matplotlib now
> has a matlab
> compatible legend command. See
> 
>
http://matplotlib.sourceforge.net/screenshots.html#legend_demo
> 
> Numerous bug fixes and minor additions
> 
> DPI parameter allows multiple output resolutions
> with correct
> scaling
> 
> Several bug fixes in GTK interactive mode using
> examples/interactive2.py
> 
> Multiple ways to specify colors, including matlab
> compatible format
> strings, RGB tuples, and html-style hex color
> strings
> 
> Rewrite of line class for much greater
> compatibility with matlab
> handle graphics commands and flexibility in
> choosing line styles and
> markers
> 
> See
> http://matplotlib.sourceforge.net/screenshots.html
> for screenshots
> and http://matplotlib.sourceforge.net/whats_new.html
> for more detailed
> information on what's new.
> 
> John Hunter
> _______________________________________________
> SciPy-user mailing list
> Sci...@sc...
> http://www.scipy.net/mailman/listinfo/scipy-user