Fixing coersion bug at SparseDataFrame construction

Author: Carlos Souza

doc/source/whatsnew/v0.20.0.txt

@@ -1045,7 +1045,7 @@ Bug Fixes
 - Bug in ``pd.concat()`` in which concatting with an empty dataframe with ``join='inner'`` was being improperly handled (:issue:`15328`)
 - Bug in ``groupby.agg()`` incorrectly localizing timezone on ``datetime`` (:issue:`15426`, :issue:`10668`, :issue:`13046`)
 
-
+- Bug in ``SparseDataFrame`` construction with lists not coercing to dtype (:issue:`15682`)
 
 - Bug in ``.read_csv()`` with ``parse_dates`` when multiline headers are specified (:issue:`15376`)
 - Bug in ``groupby.transform()`` that would coerce the resultant dtypes back to the original (:issue:`10972`, :issue:`11444`)

pandas/sparse/frame.py

@@ -159,7 +159,7 @@ def _init_dict(self, data, index, columns, dtype=None):
                     v = [v.get(i, nan) for i in index]
 
                 v = sp_maker(v)
-            sdict[k] = v
+            sdict[k] = v.astype(np.dtype(dtype)) if dtype is not None else v
 
         # TODO: figure out how to handle this case, all nan's?
         # add in any other columns we want to have (completeness)

pandas/tests/sparse/test_frame.py

@@ -28,7 +28,6 @@
 
 
 class TestSparseDataFrame(tm.TestCase, SharedWithSparse):
-
     klass = SparseDataFrame
 
     def setUp(self):
@@ -758,7 +757,8 @@ def test_sparse_frame_fillna_limit(self):
     def test_rename(self):
         # just check this works
         renamed = self.frame.rename(index=str)  # noqa
-        renamed = self.frame.rename(columns=lambda x: '%s%d' % (x, len(x)))  # noqa
+        renamed = self.frame.rename(
+            columns=lambda x: '%s%d' % (x, len(x)))  # noqa
 
     def test_corr(self):
         res = self.frame.corr()
@@ -967,7 +967,6 @@ def _check(frame, orig):
     def test_shift(self):
 
         def _check(frame, orig):
-
             shifted = frame.shift(0)
             exp = orig.shift(0)
             tm.assert_frame_equal(shifted.to_dense(), exp)
@@ -1060,7 +1059,7 @@ def test_sparse_pow_issue(self):
         df = SparseDataFrame({'A': [nan, 0, 1]})
 
         # note that 2 ** df works fine, also df ** 1
-        result = 1**df
+        result = 1 ** df
 
         r1 = result.take([0], 1)['A']
         r2 = result['A']
@@ -1126,7 +1125,7 @@ def test_isnotnull(self):
         tm.assert_frame_equal(res.to_dense(), exp)
 
 
-@pytest.mark.parametrize('index', [None, list('ab')])    # noqa: F811
+@pytest.mark.parametrize('index', [None, list('ab')])  # noqa: F811
 @pytest.mark.parametrize('columns', [None, list('cd')])
 @pytest.mark.parametrize('fill_value', [None, 0, np.nan])
 @pytest.mark.parametrize('dtype', [bool, int, float, np.uint16])
@@ -1180,7 +1179,7 @@ def test_from_to_scipy(spmatrix, index, columns, fill_value, dtype):
     tm.assert_equal(sdf.to_coo().dtype, np.object_)
 
 
-@pytest.mark.parametrize('fill_value', [None, 0, np.nan])    # noqa: F811
+@pytest.mark.parametrize('fill_value', [None, 0, np.nan])  # noqa: F811
 def test_from_to_scipy_object(spmatrix, fill_value):
     # GH 4343
     dtype = object
@@ -1226,7 +1225,6 @@ def test_from_to_scipy_object(spmatrix, fill_value):
 
 
 class TestSparseDataFrameArithmetic(tm.TestCase):
-
     def test_numeric_op_scalar(self):
         df = pd.DataFrame({'A': [nan, nan, 0, 1, ],
                            'B': [0, 1, 2, nan],
@@ -1255,7 +1253,6 @@ def test_comparison_op_scalar(self):
 
 
 class TestSparseDataFrameAnalytics(tm.TestCase):
-
     def setUp(self):
         self.data = {'A': [nan, nan, nan, 0, 1, 2, 3, 4, 5, 6],
                      'B': [0, 1, 2, nan, nan, nan, 3, 4, 5, 6],
@@ -1299,3 +1296,12 @@ def test_numpy_func_call(self):
                  'std', 'min', 'max']
         for func in funcs:
             getattr(np, func)(self.frame)
+
+    def test_type_coercion_at_construction(self):
+        # GH 15682
+        df = pd.SparseDataFrame(
+            {'a': [1, 0, 0], 'b': [0, 1, 0], 'c': [0, 0, 1]}, dtype='uint8',
+            default_fill_value=0)
+        result = df.dtypes[0]
+        expected = np.dtype('uint8')
+        assert result == expected