Generic solution to categorical problem

rs2 · Jul 17, 2017 · 218da66 · 218da66
1 parent 48e7163
commit 218da66
Show file tree

Hide file tree

Showing 2 changed files with 10 additions and 11 deletions.
diff --git a/pandas/core/reshape/merge.py b/pandas/core/reshape/merge.py
@@ -896,12 +896,6 @@ def _maybe_coerce_merge_keys(self):
                 if lk.is_dtype_equal(rk):
                     continue
 
-                # if we are dates with differing categories
-                # then allow them to proceed because
-                # coercing to object below results in integers.
-                if is_datetimelike(lk.categories) and is_datetimelike(rk.categories):
-                    continue
-
             elif is_categorical_dtype(lk) or is_categorical_dtype(rk):
                 pass
 
@@ -923,11 +917,13 @@ def _maybe_coerce_merge_keys(self):
             # Houston, we have a problem!
             # let's coerce to object
             if name in self.left.columns:
+                typ = lk.categories.dtype if is_categorical_dtype(lk) else object
                 self.left = self.left.assign(
-                    **{name: self.left[name].astype(object)})
+                    **{name: self.left[name].astype(typ)})
             if name in self.right.columns:
+                typ = rk.categories.dtype if is_categorical_dtype(rk) else object
                 self.right = self.right.assign(
-                    **{name: self.right[name].astype(object)})
+                    **{name: self.right[name].astype(typ)})
 
     def _validate_specification(self):
         # Hm, any way to make this logic less complicated??

diff --git a/pandas/tests/reshape/test_merge.py b/pandas/tests/reshape/test_merge.py
@@ -1515,7 +1515,7 @@ def test_self_join_multiple_categories(self):
 
         assert_frame_equal(result, df)
 
-    def test_dtype_on_categorical_dates(self):
+    def test_categorical_dates(self):
         # GH 16900
         # dates should not be coerced to ints
 
@@ -1534,10 +1534,13 @@ def test_dtype_on_categorical_dates(self):
         df2['date'] = df2['date'].astype('category')
 
         result = pd.merge(df, df2, how='outer', on=['date'])
-        assert result['date'].dtype == 'category'
+        assert result.shape == (3, 3)
+        assert result['date'].iloc[0] == pd.Timestamp('2001-01-01')
+        assert result['date'].iloc[-1] == pd.Timestamp('2001-01-03')
 
         result_inner = pd.merge(df, df2, how='inner', on=['date'])
-        assert result_inner['date'].dtype == 'category'
+        assert result_inner.shape == (1, 3)
+        assert result_inner['date'].iloc[-1] == pd.Timestamp('2001-01-01')
 
 
 @pytest.fixture