TimeSeries.plot ignores color kwarg

[andreas-h]

I know this looks like a duplicate of #1636, but even with pandas 0.8.1, I don't get the expected plotting behaviour:

In [1]: from numpy import arange

In [2]: import matplotlib.pyplot as plt

In [3]: import pandas

In [4]: pandas.__version__
Out[4]: '0.8.1'

In [5]: plt.figure()
Out[5]: <matplotlib.figure.Figure at 0x33afd50>

In [6]: plt.plot(arange(12), arange(12), color='green')     # this line is green
Out[6]: [<matplotlib.lines.Line2D at 0x337c150>]

In [7]: pandas.Series(arange(12) + 1).plot(color='green')   # this line is blue
Out[7]: <matplotlib.axes.AxesSubplot at 0x33c80d0>

[wesm]

thanks for the report. indeed this was not fixed by #1636 and has been remedied here

[jorisvandenbossche]

Is it possoble this bug is not yet fixed? I think it is for a normal series, but not for a timeseries as mentioned in the title. An example in the same way as above:

In [9]: pandas.Series(np.arange(12) + 1).plot(color='green')  # this line is indeed green now
Out[9]: <matplotlib.axes.AxesSubplot at 0x55d1e70>

In [12]: pandas.Series(np.arange(12) + 1, index=pandas.date_range('1/1/2000', periods=12)).plot(color='green')  # this is still blue
Out[12]: <matplotlib.axes.AxesSubplot at 0x58a8850>

And a bigger example (the case where I experienced it): http://nbviewer.ipython.org/3857689/

[brendam]

It is still a problem. I've submitted a patch, which is waiting for review. A work around is to add style='b' to the plot statement. eg:
pandas.Series(np.arange(12) + 1, index=pandas.date_range('1/1/2000', periods=12)).plot(style='b', color='green')

[jorisvandenbossche]

OK, thanks!

[wesm]

reopened til patch is merged

[wesm]

merged the PR. I think we're good to go