Plotting looks broken

[dhuard]

I get the following traceback when trying the wiki example "Monthly Data along with an exponential moving average". This is with a fairly recent version of matplotlib.

[snip]

4 
    775         # render the axes
--> 776         for a in self.axes: a.draw(renderer)
    777 
    778         # render the figure text

/usr/local/lib64/python2.5/site-packages/matplotlib/axes.pyc in draw(self, renderer, inframe)
   1400 
   1401         for zorder, i, a in dsu:
-> 1402             a.draw(renderer)
   1403 
   1404         renderer.close_group('axes')

/usr/local/lib64/python2.5/site-packages/matplotlib/axis.pyc in draw(self, renderer, *args, **kwargs)
    672         renderer.open_group(__name__)
    673         interval = self.get_view_interval()
--> 674         for tick, loc, label in self.iter_ticks():
    675             if tick is None: continue
    676             if not interval_contains(interval, loc): continue

/usr/local/lib64/python2.5/site-packages/matplotlib/axis.pyc in iter_ticks(self)
    614         Iterate through all of the major and minor ticks.
    615         """
--> 616         majorLocs = self.major.locator()
    617         majorTicks = self.get_major_ticks(len(majorLocs))
    618         self.major.formatter.set_locs(majorLocs)

/usr/local/lib64/python2.5/site-packages/scikits/timeseries/lib/plotlib.pyc in __call__(self)
    444     def __call__(self):
    445         'Return the locations of the ticks.'
--> 446         self.verify_intervals()
    447         vmin, vmax = self.viewInterval.get_bounds()
    448         if vmax < vmin:

AttributeError: TimeSeries_DateLocator instance has no attribute 'verify_intervals'

[mattknox_ca]

reinvestigate this after numpy 1.0.5 is released. If it works with that, then we can chalk this one up to incompatibilities in older versions of the modules.