matplotlib-devel Mailing List for matplotlib

Hi Tony,
have you solved this issue?
What was the problem?
I just installed the ScipySuperpack
http://macinscience.org/?page_id=6
on a clean OSX 10.6 machine
running
http://matplotlib.sourceforge.net/plot_directive/mpl_examples/pylab_examples/date_demo_rrule.py
for example from the command line works awesome, but not from TextMate.
This sounds like the same problem you had..
I'd appreciate any pointers!
Cheers,
Claus
Tony Yu-3 wrote:
> 
> I updated to the newest release of OS X---10.6.2---last night and I've
> been getting a weird windowing error for certain backends. 
> 
> I'm able to reproduce the error with the following: run a simple plot
> script (which calls plt.show) from the terminal. Close the figure. The
> problem is:
> 
> * tkagg: window closes, but it doesn't return control to the terminal
> (i.e. terminal hangs).
> * macosx: axis is cleared from the figure window, but the window (w/o
> axis) remains and terminal hangs.
> * qt4agg: works fine.
> 
> Alternatively, I can run the script from ipython. In this case, tkagg and
> qt4agg work as normal. The macosx backend hangs, the window does not
> close, and control is not returned to ipython.
> 
> In all cases, the plot shows up fine (it just doesn't always close
> properly). Running the script with python -vv doesn't help: All the debug
> code it prints stops before plt.show is called. I wish I could provide
> more debug output, but I'm not sure how.
> 
> Any ideas?
> 
> -Tony
> 
> PS. To add another piece to the mystery, I've been having problems running
> plot scripts from my text editor (Textmate): python hangs **before**
> plotting. I've traced the hang to a call to
> matplotlib.backends.pylab_setup.new_figure_manager, but I really don't
> understand where exactly hangs. Again: this only happens when running from
> Textmate, and started happening about a week ago (following a Textmate
> update). My guess is that this is a Textmate-related issue, but I thought
> I'd mention it in case these two issues combined suggest a more
> fundamental problem with my python install.
> ------------------------------------------------------------------------------
> Join us December 9, 2009 for the Red Hat Virtual Experience,
> a free event focused on virtualization and cloud computing. 
> Attend in-depth sessions from your desk. Your couch. Anywhere.
> http://p.sf.net/sfu/redhat-sfdev2dev
> _______________________________________________
> Matplotlib-devel mailing list
> Mat...@li...
> https://lists.sourceforge.net/lists/listinfo/matplotlib-devel
> 
> 
-- 
View this message in context: http://old.nabble.com/Weird-windowing-issues-since-last-update-of-OS-X-10.6-tp26658846p27061342.html
Sent from the matplotlib - devel mailing list archive at Nabble.com.

Hi,
how about adding a special minor locator class like this simple one?
class AutoDateMinorLocator(AutoDateLocator):
    """An extension to the
:class:`~matplotlib.dates.AutoDateLocator` for minor locators"""
    def viewlim_to_dt(self):
        major_ticks = self.axis.get_majorticklocs()[:2]
        print num2date(major_ticks[0], self.tz),
num2date(major_ticks[-1], self.tz)
        return num2date(major_ticks[0], self.tz),
num2date(major_ticks[-1], self.tz)
    def datalim_to_dt(self):
        return self.get_view_interval()
Maybe a better implementation should exclude minor locations where
major locations are defined.
--
Stephane Raynaud