Hist fillstyle error

[revkarol]

Hi,

I've a histogram of type Hist which I try to set the fill style.

noiseon.fillstyle= 'solid'

which gives me the following error.  Anybody else have this problem?

  File "./plot_hvonoff_noise.py", line 44, in <module>

    r2m.hist(stack, stacked=False) # label='Assembly', histtype='step')

  File "/usr/local/lib/python2.7/dist-packages/rootpy-dev-py2.7.egg/rootpy/plotting/root2matplotlib.py", line 242, in hist

    returns.append(_hist(h, axes=axes, logy=logy, **kwargs))

  File "/usr/local/lib/python2.7/dist-packages/rootpy-dev-py2.7.egg/rootpy/plotting/root2matplotlib.py", line 275, in _hist

    zorder=zorder)

  File "/usr/local/lib/python2.7/dist-packages/rootpy-dev-py2.7.egg/rootpy/plotting/root2matplotlib.py", line 685, in fill_between

    return axes.fill_between(x, top, bottom, **kwargs)

  File "/usr/lib/pymodules/python2.7/matplotlib/axes.py", line 6555, in fill_between

    collection = mcoll.PolyCollection(polys, **kwargs)

  File "/usr/lib/pymodules/python2.7/matplotlib/collections.py", line 637, in __init__

    Collection.__init__(self,**kwargs)

  File "/usr/lib/pymodules/python2.7/matplotlib/collections.py", line 112, in __init__

    self.update(kwargs)

  File "/usr/lib/pymodules/python2.7/matplotlib/artist.py", line 659, in update

    raise AttributeError('Unknown property %s'%k)

Karol.

[Noel Dawe]

Hi Karol,

What matplotlib version are you using and what is the actual exception message (the value of "Unknown property %s'%k")?

Noel

[revkarol]

Hi Noel,

The unknown property was 'hatch'.  I had version 1.1.1 of matplotlib.  I upgraded to 1.3.0 and all seems to be well.  Should have thought of that myself.

Thanks,

Karol.

[Noel Dawe]

Hi Karol,

No problem! Glad to see it's fixed in 1.3.0. I had a similar problem recently where matplotlib had not been updated on one of my computers in a while and the plots were not displaying correctly. matplotlib is changing rapidly so it's hard to maintain support for older releases while taking advantage of the new features where some of the new features are actually required to get the desired results in rootpy's root2matplotlib. I think we will continue to recommend the latest matplotlib release if possible.

Cheers,

Noel