BUG: Load data from a CategoricalIndex for dtype comparison, closes #… (#16738)

* BUG: Load data from a CategoricalIndex for dtype comparison, closes #16627

* Enable is_dtype_equal on CategoricalIndex, fixed some doc typos, added ordered CategoricalIndex test

* Flake8 windows suggestion

* Fixed some documentation/formatting issues, clarified the purpose of the test case.

(cherry picked from commit 5b88d2f)

Author: thequackdaddy

.github/PULL_REQUEST_TEMPLATE.md

@@ -1,4 +1,4 @@
  - [ ] closes #xxxx
  - [ ] tests added / passed
- - [ ] passes ``git diff upstream/master --name-only -- '*.py' | flake8 --diff``
+ - [ ] passes ``git diff upstream/master --name-only -- '*.py' | flake8 --diff`` (On Windows, ``git diff upstream/master -u -- "*.py" | flake8 --diff`` might work as an alternative.)
  - [ ] whatsnew entry

doc/source/contributing.rst

@@ -525,6 +525,12 @@ run this slightly modified command::
 
    git diff master --name-only -- '*.py' | grep 'pandas/' | xargs flake8
 
+Note that on Windows, ``grep``, ``xargs``, and other tools are likely
+unavailable. However, this has been shown to work on smaller commits in the
+standard Windows command line::
+
+   git diff master -u -- "*.py" | flake8 --diff
+
 Backwards Compatibility
 ~~~~~~~~~~~~~~~~~~~~~~~
 

doc/source/whatsnew/v0.20.3.txt

@@ -78,6 +78,7 @@ Sparse
 Reshaping
 ^^^^^^^^^
 
+- Bug in joining on a ``MultiIndex`` with a ``category`` dtype for a level (:issue:`16627`).
 
 
 Numeric

pandas/core/indexes/category.py

@@ -560,6 +560,9 @@ def take(self, indices, axis=0, allow_fill=True,
                                            na_value=-1)
         return self._create_from_codes(taken)
 
+    def is_dtype_equal(self, other):
+        return self._data.is_dtype_equal(other)
+
     take_nd = take
 
     def map(self, mapper):

pandas/tests/test_join.py

@@ -1,11 +1,11 @@
 # -*- coding: utf-8 -*-
 
 import numpy as np
-from pandas import Index
+from pandas import Index, DataFrame, Categorical, merge
 
 from pandas._libs import join as _join
 import pandas.util.testing as tm
-from pandas.util.testing import assert_almost_equal
+from pandas.util.testing import assert_almost_equal, assert_frame_equal
 
 
 class TestIndexer(object):
@@ -192,3 +192,43 @@ def test_inner_join_indexer2():
 
     exp_ridx = np.array([0, 1, 2, 3], dtype=np.int64)
     assert_almost_equal(ridx, exp_ridx)
+
+
+def test_merge_join_categorical_multiindex():
+    # From issue 16627
+    a = {'Cat1': Categorical(['a', 'b', 'a', 'c', 'a', 'b'],
+                             ['a', 'b', 'c']),
+         'Int1': [0, 1, 0, 1, 0, 0]}
+    a = DataFrame(a)
+
+    b = {'Cat': Categorical(['a', 'b', 'c', 'a', 'b', 'c'],
+                            ['a', 'b', 'c']),
+         'Int': [0, 0, 0, 1, 1, 1],
+         'Factor': [1.1, 1.2, 1.3, 1.4, 1.5, 1.6]}
+    b = DataFrame(b).set_index(['Cat', 'Int'])['Factor']
+
+    expected = merge(a, b.reset_index(), left_on=['Cat1', 'Int1'],
+                     right_on=['Cat', 'Int'], how='left')
+    result = a.join(b, on=['Cat1', 'Int1'])
+    expected = expected.drop(['Cat', 'Int'], axis=1)
+    assert_frame_equal(expected, result)
+
+    # Same test, but with ordered categorical
+    a = {'Cat1': Categorical(['a', 'b', 'a', 'c', 'a', 'b'],
+                             ['b', 'a', 'c'],
+                             ordered=True),
+         'Int1': [0, 1, 0, 1, 0, 0]}
+    a = DataFrame(a)
+
+    b = {'Cat': Categorical(['a', 'b', 'c', 'a', 'b', 'c'],
+                            ['b', 'a', 'c'],
+                            ordered=True),
+         'Int': [0, 0, 0, 1, 1, 1],
+         'Factor': [1.1, 1.2, 1.3, 1.4, 1.5, 1.6]}
+    b = DataFrame(b).set_index(['Cat', 'Int'])['Factor']
+
+    expected = merge(a, b.reset_index(), left_on=['Cat1', 'Int1'],
+                     right_on=['Cat', 'Int'], how='left')
+    result = a.join(b, on=['Cat1', 'Int1'])
+    expected = expected.drop(['Cat', 'Int'], axis=1)
+    assert_frame_equal(expected, result)