API/BUG: Categorical.is_dtype_equal doesn't compare to Series

[chris-b1]

xref #16639

In [25]: c1 = pd.Categorical(['a', 'b'])

In [26]: c2 = pd.Categorical(['a', 'b'])

In [27]: s = pd.Series(c1)

In [29]: c1.is_dtype_equal(c2)
Out[29]: True

In [30]: c1.is_dtype_equal(s)
Out[30]: False

I think [30] should be true?

[jreback]

yep [30] should be True.

[jschendel]

Looks like this is fixed in 0.22.0 (so really probably fixed in 0.21.x):

In [2]: pd.__version__
Out[2]: '0.22.0'

In [3]: c1 = pd.Categorical(['a', 'b'])

In [4]: c2 = pd.Categorical(['a', 'b'])

In [5]: s = pd.Series(c1)

In [6]: c1.is_dtype_equal(c2)
Out[6]: True

In [7]: c1.is_dtype_equal(s)
Out[7]: True

Could still use a test to prevent regression though.

[jreback]

if u can put up a test would be great