BUG: .loc against CategoricalIndex may result in normal Index

[sinhrks]

.loc against CategoricalIndex with values included in its categories but not appears as codes results in normal Index

import numpy as np
import pandas as pd

index = pd.CategoricalIndex(list('aabbca'), categories=list('cabe'))
df = pd.DataFrame({'A' : np.arange(6,dtype='int64')}, index=index)

# OK
df.loc[['a', 'b']].index
# CategoricalIndex([u'a', u'a', u'a', u'b', u'b'], categories=[u'c', u'a', u'b', u'e'], ordered=False, dtype='category')

# NG, must be CategoricalIndex
df.loc[['a', 'b', 'e']].index
# Index([u'a', u'a', u'a', u'b', u'b', u'e'], dtype='object')

Even though this is tested here, it doesn't check dtype.

• https://github.com/pydata/pandas/blob/master/pandas/tests/test_indexing.py#L4781

[jreback]

yeh that last does seem buggy.