Fixed DataFrame.__setitem__ for updating to sparse.

Closes pandas-dev#22367

Author: TomAugspurger

doc/source/whatsnew/v0.24.0.txt

@@ -731,6 +731,12 @@ Reshaping
 - Bug in :meth:`Series.replace` and meth:`DataFrame.replace` when dict is used as the `to_replace` value and one key in the dict is is another key's value, the results were inconsistent between using integer key and using string key (:issue:`20656`)
 -
 
+Sparse
+^^^^^^
+
+- Updating a boolean, datetime, or timedelta column to be Sparse now works (:issue:`22367`)
+
+
 Build Changes
 ^^^^^^^^^^^^^
 

pandas/core/internals/blocks.py

@@ -2298,7 +2298,8 @@ def _try_coerce_result(self, result):
         return result
 
     def should_store(self, value):
-        return issubclass(value.dtype.type, np.timedelta64)
+        return (issubclass(value.dtype.type, np.timedelta64) and
+                not is_extension_array_dtype(value))
 
     def to_native_types(self, slicer=None, na_rep=None, quoting=None,
                         **kwargs):
@@ -2337,7 +2338,8 @@ def _can_hold_element(self, element):
         return isinstance(element, (bool, np.bool_))
 
     def should_store(self, value):
-        return issubclass(value.dtype.type, np.bool_)
+        return (issubclass(value.dtype.type, np.bool_) and not
+                is_extension_array_dtype(value))
 
     def replace(self, to_replace, value, inplace=False, filter=None,
                 regex=False, convert=True, mgr=None):
@@ -2879,7 +2881,8 @@ def to_native_types(self, slicer=None, na_rep=None, date_format=None,
 
     def should_store(self, value):
         return (issubclass(value.dtype.type, np.datetime64) and
-                not is_datetimetz(value))
+                not is_datetimetz(value) and
+                not is_extension_array_dtype(value))
 
     def set(self, locs, values, check=False):
         """

pandas/tests/reshape/test_reshape.py

@@ -247,10 +247,16 @@ def test_dataframe_dummies_prefix_str(self, df, sparse):
                              dtype=np.uint8)
         expected = expected.astype({"C": np.int64})
         if sparse:
-            expected.iloc[1:] = expected.iloc[1:].astype(SparseDtype("uint8"))
-            # seemingly impossible to make expected .
-            # raise pytest.xfail(reason="can't make expected")
-            pass
+            # work around astyping & assigning with duplicate columns
+            # https://github.com/pandas-dev/pandas/issues/14427
+            expected = pd.concat([
+                pd.Series([1, 2, 3], name='C'),
+                pd.Series([1, 0, 1], name='bad_a', dtype='Sparse[uint8]'),
+                pd.Series([0, 1, 0], name='bad_b', dtype='Sparse[uint8]'),
+                pd.Series([1, 1, 0], name='bad_b', dtype='Sparse[uint8]'),
+                pd.Series([0, 0, 1], name='bad_c', dtype='Sparse[uint8]'),
+            ], axis=1)
+
         assert_frame_equal(result, expected)
 
     def test_dataframe_dummies_subset(self, df, sparse):
@@ -336,10 +342,10 @@ def test_dataframe_dummies_with_na(self, df, sparse, dtype):
         columns = ['A_a', 'A_b', 'A_nan', 'B_b', 'B_c', 'B_nan']
         expected[columns] = expected[columns].astype(e_dtype)
         if sparse:
-            expected[columns] = expected[columns].apply(
+            tmp = expected[columns].apply(
                 lambda x: pd.SparseSeries(x)
             )
-            raise pytest.xfail(reason="that apply is broken?")
+            expected[tmp.columns] = tmp
         assert_frame_equal(result, expected)
 
         result = get_dummies(df, dummy_na=False, sparse=sparse, dtype=dtype)

pandas/tests/sparse/frame/test_frame.py

@@ -582,6 +582,20 @@ def _check_frame(frame, orig):
 
         self._check_all(_check_frame)
 
+    @pytest.mark.parametrize('values', [
+        [True, False],
+        [0, 1],
+        [1, None],
+        ['a', 'b'],
+        [pd.Timestamp('2017'), pd.NaT],
+        [pd.Timedelta('10s'), pd.NaT],
+    ])
+    def test_setitem_more(self, values):
+        df = pd.DataFrame({"A": values})
+        df['A'] = pd.SparseArray(values)
+        expected = pd.DataFrame({'A': pd.SparseArray(values)})
+        tm.assert_frame_equal(df, expected)
+
     def test_setitem_corner(self):
         self.frame['a'] = self.frame['B']
         tm.assert_sp_series_equal(self.frame['a'], self.frame['B'],