REGR: allow merging on object boolean columns (#21310)

Author: jorisvandenbossche

doc/source/whatsnew/v0.23.1.txt

@@ -65,15 +65,14 @@ In addition, ordering comparisons will raise a ``TypeError`` in the future.
   a tz-aware time instead of tz-naive (:issue:`21267`) and :attr:`DatetimeIndex.date`
   returned incorrect date when the input date has a non-UTC timezone (:issue:`21230`).
 - Fixed regression in :meth:`pandas.io.json.json_normalize` when called with ``None`` values
-  in nested levels in JSON (:issue:`21158`).
+  in nested levels in JSON, and to not drop keys with value as `None` (:issue:`21158`, :issue:`21356`).
 - Bug in :meth:`~DataFrame.to_csv` causes encoding error when compression and encoding are specified (:issue:`21241`, :issue:`21118`)
 - Bug preventing pandas from being importable with -OO optimization (:issue:`21071`)
 - Bug in :meth:`Categorical.fillna` incorrectly raising a ``TypeError`` when `value` the individual categories are iterable and `value` is an iterable (:issue:`21097`, :issue:`19788`)
 - Fixed regression in constructors coercing NA values like ``None`` to strings when passing ``dtype=str`` (:issue:`21083`)
 - Regression in :func:`pivot_table` where an ordered ``Categorical`` with missing
   values for the pivot's ``index`` would give a mis-aligned result (:issue:`21133`)
-- Fixed Regression in :func:`nested_to_record` which now flattens list of dictionaries and doesnot drop keys with value as `None` (:issue:`21356`)
-
+- Fixed regression in merging on boolean index/columns (:issue:`21119`).
 
 .. _whatsnew_0231.performance:
 

pandas/core/reshape/merge.py

@@ -28,6 +28,7 @@
     is_int_or_datetime_dtype,
     is_dtype_equal,
     is_bool,
+    is_bool_dtype,
     is_list_like,
     is_datetimelike,
     _ensure_int64,
@@ -974,9 +975,14 @@ def _maybe_coerce_merge_keys(self):
 
             # Check if we are trying to merge on obviously
             # incompatible dtypes GH 9780, GH 15800
-            elif is_numeric_dtype(lk) and not is_numeric_dtype(rk):
+
+            # boolean values are considered as numeric, but are still allowed
+            # to be merged on object boolean values
+            elif ((is_numeric_dtype(lk) and not is_bool_dtype(lk))
+                    and not is_numeric_dtype(rk)):
                 raise ValueError(msg)
-            elif not is_numeric_dtype(lk) and is_numeric_dtype(rk):
+            elif (not is_numeric_dtype(lk)
+                    and (is_numeric_dtype(rk) and not is_bool_dtype(rk))):
                 raise ValueError(msg)
             elif is_datetimelike(lk) and not is_datetimelike(rk):
                 raise ValueError(msg)

pandas/tests/reshape/merge/test_merge.py

@@ -1526,6 +1526,27 @@ def test_merge_on_ints_floats_warning(self):
             result = B.merge(A, left_on='Y', right_on='X')
             assert_frame_equal(result, expected[['Y', 'X']])
 
+    def test_merge_incompat_infer_boolean_object(self):
+        # GH21119: bool + object bool merge OK
+        df1 = DataFrame({'key': Series([True, False], dtype=object)})
+        df2 = DataFrame({'key': [True, False]})
+
+        expected = DataFrame({'key': [True, False]}, dtype=object)
+        result = pd.merge(df1, df2, on='key')
+        assert_frame_equal(result, expected)
+        result = pd.merge(df2, df1, on='key')
+        assert_frame_equal(result, expected)
+
+        # with missing value
+        df1 = DataFrame({'key': Series([True, False, np.nan], dtype=object)})
+        df2 = DataFrame({'key': [True, False]})
+
+        expected = DataFrame({'key': [True, False]}, dtype=object)
+        result = pd.merge(df1, df2, on='key')
+        assert_frame_equal(result, expected)
+        result = pd.merge(df2, df1, on='key')
+        assert_frame_equal(result, expected)
+
     @pytest.mark.parametrize('df1_vals, df2_vals', [
         ([0, 1, 2], ["0", "1", "2"]),
         ([0.0, 1.0, 2.0], ["0", "1", "2"]),
@@ -1538,6 +1559,8 @@ def test_merge_on_ints_floats_warning(self):
             pd.date_range('20130101', periods=3, tz='US/Eastern')),
         ([0, 1, 2], Series(['a', 'b', 'a']).astype('category')),
         ([0.0, 1.0, 2.0], Series(['a', 'b', 'a']).astype('category')),
+        # TODO ([0, 1], pd.Series([False, True], dtype=bool)),
+        ([0, 1], pd.Series([False, True], dtype=object))
     ])
     def test_merge_incompat_dtypes(self, df1_vals, df2_vals):
         # GH 9780, GH 15800