matplotlib-devel Mailing List for matplotlib

If I do 
 ax.yaxis.set_major_formatter(ScalarFormatter(useMathText=True))
only the "x1e6" multiplier is rendered in mathtext, which makes the
tick labels and the multiplier look different. Is this done on
purpose, or an inadvertent omission?
The following one-line change makes the tick labels use mathtext.
--- ticker.py.orig 2005年06月14日 00:40:26.000000000 +0300
+++ ticker.py 2005年11月16日 12:11:29.055601000 +0200
@@ -337,6 +337,7 @@
 for loc in locs]
 sigfigs.sort()
 self.format = '%1.' + str(sigfigs[-1]) + 'f'
+ if self._useMathText: self.format = '$' + self.format + '$'
 
 def pprint_val(self, x):
 xp = (x-self.offset)/10**self.orderOfMagnitude
-- 
Jouni K Seppänen

Hi,
I have recently started to make plots with matplotlib. It's a great tool!
When trying to change the width of ticklines, I noticed that
axis.get_ticklines() only returns the line2D objects for the major
ticks. But one might want to change the linewidth of the minor ticks.
Hence, I propose something like
 def get_minor_ticklines(self):
 'Return the minor ticklines lines as a list of Line2D instance'
 lines =3D []
 for tick in self.minorTicks:
 lines.append(tick.tick1line)
 lines.append(tick.tick2line)
 return silent_list('Line2D minor ticklines', lines)
in class Axis. What do people think?
Also note, that, in fact, ticks are 'markers' and one has to change the
markeredgewidth (mew).
 - Christian
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> On Tue, 2005年11月15日 at 20:30 -0800,
> mat...@li... wrote:
> > Hi,
> > 
> > With matplotlib-0.84, pythonw embedding_in_qt.py
> > returns several
> > of these messages:
> > ___
> > Traceback (most recent call last):
> > File "embedding_in_qt.py", line 54, in sizeHint
> > w, h = self.fig.get_width_height() 
> > AttributeError: Figure instance has no attribute
> > 'get_width_height'
> > ___
> > With matplotlib-0.82 the example works fine.
> > Changelog shows 
> > 2005年06月18日 Move Figure.get_width_height() to
> > FigureCanvasBase and return 
> > int instead of float. - SC 
> > Is this a bug, or am I missing something?
> > 
> > Thanks,
> > 
> > Virgil
The self.fig.get_width_height() should be
self.get_width_height(). Changed in CVS.
Steve
Send instant messages to your online friends http://au.messenger.yahoo.com