BUG: Fix .groupby(categorical, sort=False) failing

Author: kernc

doc/source/whatsnew/v0.20.0.txt

@@ -539,6 +539,7 @@ Bug Fixes
 
 - Bug in ``resample``, where a non-string ```loffset`` argument would not be applied when resampling a timeseries (:issue:`13218`)
 
+- Bug in ``.groupby`` where ```.groupby(categorical, sort=False)`` would raise ``ValueError`` due to non-matching categories (:issue:`13179`)
 
 
 

pandas/core/groupby.py

@@ -2315,6 +2315,11 @@ def __init__(self, index, grouper=None, obj=None, name=None, level=None,
                 # groupby
                 else:
                     cat = self.grouper.unique()
+                    all_categories = self.grouper.categories
+                    cat.add_categories(
+                        all_categories[
+                            ~all_categories.isin(cat.categories)],
+                        inplace=True)  # GH-13179
                     self.grouper = self.grouper.reorder_categories(
                         cat.categories)
 

pandas/tests/groupby/test_categorical.py

@@ -284,6 +284,30 @@ def test_groupby_multi_categorical_as_index(self):
 
             tm.assert_frame_equal(result, expected, check_index_type=True)
 
+    def test_groupby_preserve_categories(self):
+        # GH-13179
+        categories = list('abc')
+
+        # ordered=True
+        df = DataFrame({'A': pd.Categorical(list('ba'),
+                                            categories=categories,
+                                            ordered=True)})
+        index = pd.CategoricalIndex(categories, categories, ordered=True)
+        tm.assert_index_equal(df.groupby('A', sort=True).first().index, index)
+        tm.assert_index_equal(df.groupby('A', sort=False).first().index, index)
+
+        # ordered=False
+        df = DataFrame({'A': pd.Categorical(list('ba'),
+                                            categories=categories,
+                                            ordered=False)})
+        sort_index = pd.CategoricalIndex(categories, categories, ordered=False)
+        nosort_index = pd.CategoricalIndex(list('bac'), list('bac'),
+                                           ordered=False)
+        tm.assert_index_equal(df.groupby('A', sort=True).first().index,
+                              sort_index)
+        tm.assert_index_equal(df.groupby('A', sort=False).first().index,
+                              nosort_index)
+
     def test_groupby_preserve_categorical_dtype(self):
         # GH13743, GH13854
         df = DataFrame({'A': [1, 2, 1, 1, 2],