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

Author: kernc

pandas/core/categorical.py

@@ -602,6 +602,21 @@ def _get_categories(self):
     categories = property(fget=_get_categories, fset=_set_categories,
                           doc=_categories_doc)
 
+    def _codes_for_groupby(self, sort=False):
+        """ Return a Categorical adjusted for groupby """
+        cat = self
+        # sort=False should order groups in as-encountered order (GH-8868)
+        if not sort:
+            cat = self.unique()
+            all_categories = self.categories
+            # But all categories should be present, including those missing
+            # from the data (GH-13179), which .unique() dropped
+            cat.add_categories(all_categories[
+                                   ~all_categories.isin(cat.categories)],
+                               inplace=True)
+            cat = self.reorder_categories(cat.categories)
+        return cat
+
     _ordered = None
 
     def set_ordered(self, value, inplace=False):

pandas/core/groupby.py

@@ -2300,28 +2300,7 @@ def __init__(self, index, grouper=None, obj=None, name=None, level=None,
             # a passed Categorical
             elif is_categorical_dtype(self.grouper):
 
-                # must have an ordered categorical
-                if self.sort:
-                    if not self.grouper.ordered:
-
-                        # technically we cannot group on an unordered
-                        # Categorical
-                        # but this a user convenience to do so; the ordering
-                        # is preserved and if it's a reduction it doesn't make
-                        # any difference
-                        pass
-
-                # fix bug #GH8868 sort=False being ignored in categorical
-                # 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)
+                self.grouper = self.grouper._codes_for_groupby(self.sort)
 
                 # we make a CategoricalIndex out of the cat grouper
                 # preserving the categories / ordered attributes

pandas/indexes/category.py

@@ -550,6 +550,10 @@ def _append_same_dtype(self, to_concat, name):
         result.name = name
         return result
 
+    def _codes_for_groupby(self, sort):
+        """ Return a Categorical adjusted for groupby """
+        return self.values._codes_for_groupby(sort)
+
     @classmethod
     def _add_comparison_methods(cls):
         """ add in comparison methods """