Question

So the situation is like this. Suppose we have:

class Test(datetime):
    def a(self):
        return '{0:02d}{1:02d}{2:02d}{3:02d}'.format(self.year % 100, self.month, self.day, self.hour)
x = [Test(2010, 05, 02, 06), Test(2010, 05, 02, 12)]
y = [0, 1]

Now trying to use:

import matplotlib.pyplot as plt
plt.plot(x, y)

gives a blank graph and the following error:

/usr/local/lib/python2.7/dist-packages/matplotlib/pyplot.py in plot(*args, **kwargs)
   2456         ax.hold(hold)
   2457     try:
-> 2458         ret = ax.plot(*args, **kwargs)
   2459         draw_if_interactive()
   2460     finally:
/usr/local/lib/python2.7/dist-packages/matplotlib/axes.pyc in plot(self, *args, **kwargs)
   3847 
   3848         for line in self._get_lines(*args, **kwargs):
-> 3849             self.add_line(line)
   3850             lines.append(line)
   3851 
/usr/local/lib/python2.7/dist-packages/matplotlib/axes.pyc in add_line(self, line)
   1441             line.set_clip_path(self.patch)
   1442 
-> 1443         self._update_line_limits(line)
   1444         if not line.get_label():
   1445             line.set_label('_line%d'%len(self.lines))
/usr/local/lib/python2.7/dist-packages/matplotlib/axes.pyc in _update_line_limits(self, line)
   1449 
   1450     def _update_line_limits(self, line):
-> 1451         p = line.get_path()
   1452         if p.vertices.size > 0:
   1453             self.dataLim.update_from_path(p, self.ignore_existing_data_limits,
/usr/local/lib/python2.7/dist-packages/matplotlib/lines.pyc in get_path(self)
    642         """
    643         if self._invalidy or self._invalidx:
--> 644             self.recache()
    645         return self._path
    646 
/usr/local/lib/python2.7/dist-packages/matplotlib/lines.pyc in recache(self, always)
    390                 x = ma.asarray(xconv, np.float_)
    391             else:
--> 392                 x = np.asarray(xconv, np.float_)
    393             x = x.ravel()
    394         else:
/usr/local/lib/python2.7/dist-packages/numpy/core/numeric.pyc in asarray(a, dtype, order)
    233 
    234     """
--> 235     return array(a, dtype, copy=False, order=order)
    236 
    237 def asanyarray(a, dtype=None, order=None):
TypeError: float() argument must be a string or a number

Do any of you know what’s going on? And yes, I know there is a date2num that I could use on x, but I’d like to understand the reason for the error here if it’s possible.

Answer 1

Matplotlib uses the __class__ attribute of an object to determine how to handle it. Your Test class isn’t the same as the datetime class:

>>> from datetime import datetime
>>> class Test(datetime):
...   def a(self):
...     return '{0:02d}{1:02d}{2:02d}{3:02d}'.format(self.year % 100, self.month, self.day, self.hour)
...
>>>
>>> t = Test(2010, 05, 02, 06)
>>> d = datetime(2010, 05, 02, 06)
>>>
>>> getattr(t, '__class__')
<class '__main__.Test'>
>>> getattr(d, '__class__')
<type 'datetime.datetime'>

So matplotlib doesn’t recognise your Test class as a datetime.

Here’s an ugly hack that works around the problem by registering the Test class so that its handled the same way a datetime is. Check out the matplotlib.units documentation for more information. This code is not guaranteed to work with future versions of matplotlib:

>>> import matplotlib.pyplot as plt
>>> import matplotlib.units, matplotlib.dates
>>> import datetime
>>>
>>> class Test(datetime.datetime):
...   def a(self):
...     return '{0:02d}{1:02d}{2:02d}{3:02d}'.format(self.year % 100, self.month, self.day, self.hour)
...
>>> matplotlib.units.registry[Test] = matplotlib.dates.DateConverter()
>>>
>>> x = [Test(2010, 05, 02, 06), Test(2010, 05, 02, 12)]
>>> y = [0, 1]
>>> plt.plot(x, y)
[<matplotlib.lines.Line2D object at 0x02F6E730>]

(I tested this using python 2.7.2 with matplotlib 1.1.0 on Windows)