matplotlib-users Mailing List for matplotlib

belinda thom wrote:
> Hi,
> 
> I could've sworn I recently read about how to draw a line on a plot 
> in such a way that you see it regardless of what the axis ranges are.
If you need horizontal or vertical lines specified in normalized 
coordinates (0 to 1 spans the axes) then you can use the Axes.axhline() 
and Axes.axvline() methods or their pylab function counterparts. There 
are also methods for specifying the length of the line in normalized 
coordinates and the position in data coordinates: axhspan() and 
axvspan(). See examples/axhspan.py for examples of both types using the 
pylab interface.
Eric

Alan G Isaac wrote:
> I assume there must be a reason for this::
> 
> >>> import pylab
> >>> pylab.__version__
> Traceback (most recent call last):
> File "<stdin>", line 1, in <module>
> AttributeError: 'module' object has no attribute '__version__'
> 
> That has always bothered me.
> But of course you can::
> 
> >>> import matplotlib
> >>> matplotlib.__version__
> '0.87.7'
After a little experimentation I have tentatively concluded that 
__version__ only gets imported if it is a package attribute--that is, if 
it is defined in __init__.py. Matplotlib is a package but pylab is just 
a module, so it does not seem to be possible to give it matplotlib's 
__version__. I could not find anything in the python documentation 
about this, though.
Eric

Hi,
I could've sworn I recently read about how to draw a line on a plot 
in such a way that you see it regardless of what the axis ranges are.
Now that I need it, I can't find it anywhere. I'm not even sure what 
to search for.
I suspect this is really trivial and hope some more advanced users 
can enlighten me.
Thanks,
--b

Hello,
Regarding http://matplotlib.sourceforge.net/installing.html, there is 
some unreliable info.
In particular, under the OS X topic,
"Robery Kern has built an all-in-one installer which includes scipy, 
Numeric, numarray, matplotlib, ipython, VTK, MayaVi, PIL, the 
enthought tool suite and much more; see >MacEnthon."
should be removed.
Whoever maintains this page should make the appropriate change.
On Dec 6, 2006, at 11:38 PM, Robert Kern wrote:
> belinda thom wrote:
>> I do not want to compile code myself unless absolutely necessary. I
>> was wondering what was up with the "MacEnton" suite; clicking on the
>> link described at the above web page informs me that the "MacEnthon"
>> page does not exist.
>
> Umm, ignore it. It targetted a now-old Python distribution, and I 
> don't have
> time to update it anymore. References recommending it should be 
> removed.
>
> -- 
> Robert Kern

In the process of creating a Cocoa app for scientific data
acquisition, I've built an NSImageView subclass that displays an MPL
figure in the Cocoa NSView hierarchy. It's based on the CocoaAgg
backend code in MPL. There are also some subclasses that allow the use
of Cocoa Bindings to supply the data, axis labels, and units for the
figure. Finally, I've created an IB palette for adding one of the
PlotView subclasses to a Cocoa app and setting its bindings in
Interface Builder.
For plotting data passed via NSData instances is converted to a numpy
array via numpy.frombuffer using the PlotView's .dtype parameter as
the array dtype (see the source for DataPlotView in PlotView.py).
I think the ability to have such a powerful plotting library available
from Cocoa with so few lines of code illustrates the huge power of the
combination of Cocoa, PyObjC, and the many good Python libraries such
as MPL.
CocoaMPLPlotView requires PyObjc, numpy, and matplotlib.
You can download a copy of the CocoaMPLPlotView and
CocoaMPLPlotViewPalette source from
http://fairhall-lab.physiol.washington.edu/~bwark/mpl/.
I don't have time to provide support for these, but would love to hear
if anyone finds them useful. Obviously if you have comments, bugs, or
patches, I'd love to hear those as well. Happy coding.
Thanks,
Barry

Sorry John,
I see this was fixed a while ago - I was still using 0.87.3 from the 
last Enthought edition. Now that there's a scipy installer, I should 
upgrade numpy/scipy/mpl to something more current.
Gary R.
Gary Ruben wrote:
> While I think of it, I think the default zorder of legends should be 
> bigger so that, by default it overlays all plot lines and symbols.
> 
> Gary R.

I assume there must be a reason for this::
 >>> import pylab
 >>> pylab.__version__
 Traceback (most recent call last):
 File "<stdin>", line 1, in <module>
 AttributeError: 'module' object has no attribute '__version__'
That has always bothered me.
But of course you can::
 >>> import matplotlib
 >>> matplotlib.__version__
 '0.87.7'
hth,
Alan Isaac

I didn't see it either. For questions like this, I recommend heading to
the #python channel on irc.freenode.net. If you find out how, please
post.
josh
On Fri, 8 Dec 2006, belinda thom wrote:
> Hi,
>
> Perhaps I'm missing something really basic, but I can't figure out
> how to query pylab as to what version it is (the usual import <foo>,
> <foo>.__version doesn't work).
>
> Advice appreciated.
>
> --b
>
> -------------------------------------------------------------------------
> Take Surveys. Earn Cash. Influence the Future of IT
> Join SourceForge.net's Techsay panel and you'll get the chance to share your
> opinions on IT & business topics through brief surveys - and earn cash
> http://www.techsay.com/default.php?page=join.php&p=sourceforge&CID=DEVDEV
> _______________________________________________
> Matplotlib-users mailing list
> Mat...@li...
> https://lists.sourceforge.net/lists/listinfo/matplotlib-users
>

Hi,
Perhaps I'm missing something really basic, but I can't figure out 
how to query pylab as to what version it is (the usual import <foo>, 
<foo>.__version doesn't work).
Advice appreciated.
--b

Hi again,
At http://matplotlib.sourceforge.net/installing.html under topic OS X:
 "All of the backends run on OS X. Chris Barker has built a binary 
package (fink users see below) for matplotlib which is hosted on 
pythonmac, and works with Agg, Wx and Tk; see the step-by-step 
instructions kindly provided by Michael Tobis."
note that the step-by-step link leads to:
 Under revision. Please come back soon.
------
It is because I've had some installation troubles that I have been 
looking at the OS X install info so closely.
For instance: originally I tried to use macports for all python site- 
packages-related installs (as well as using their python24). The ONLY 
backend that seems to work thru that route is WxAgg (http:// 
howdy.physics.nyu.edu/index.php/Numpy_For_Mac_Using_MacPython 
corroborates this). I was able to achieve results w/backend WxAgg and 
numerix Numeric. However, when I tried to change the toolbar to 
classic, WxAgg crapped out.
W/MacPorts, I've been able to get the TkAgg backend to work, but 
_NOT_ with numerix set to Numeric. Success was only achieved w/TkAgg 
when numerix was set to Numpy. When I try the WXAgg backend (along w/ 
the pythonmac 2.6.3.3 wxPython dmg for python 2.4), a simple plot 
([1,2,3]) craps out with:
exceptions.MemoryError Traceback (most 
recent call last)
<snip>
---> 63 self.bitmap = _convert_agg_to_wx_bitmap 
(self.get_renderer(), None)
 64 if repaint:
 65 self.gui_repaint()
<snip>
MemoryError: _wxagg.convert_agg_to_wx_bitmap(): could not create the 
wx.Bitmap