BUG/API: allow get_indexer to work with nans

[cpcloud]

• clear the FIXME in Categorical in Index([1,2,.np.nan]).get_indexer([np.nan]) returns wrong value? #7820 (comment)

closes #7820

[jreback]

looks ok conceptually...but failing tests.

[cpcloud]

yeah a few kinks to work out

[jreback]

@cpcloud can you circle back to this when you have a chacne?

[cpcloud]

sure

[jankatins]

If this is fixed, the workaround in categorical.py should be removed:

        # FIXME: the following can be removed after https://github.com/pydata/pandas/issues/7820
        # is fixed.
        # float levels do currently return -1 for np.nan, even if np.nan is included in the index
        # "repair" this here
        if com.isnull(rvalue).any() and com.isnull(self.levels).any():
            nan_pos = np.where(com.isnull(self.levels))
            lindexer[lindexer == -1] = nan_pos

[This is in #8007, so not yet merged]

[jreback]

@cpcloud ?

[jreback]

@cpcloud ?

[jreback]

@cpcloud status?

[jreback]

@cpcloud can you update?

[jreback]

closing