Add test of empty dataframe in ExtensionDtype

Author: kotamatsuoka

pandas/tests/extension/arrow/test_bool.py

@@ -15,6 +15,13 @@ def dtype():
     return ArrowBoolDtype()
 
 
+@pytest.fixture
+def columns():
+    values = np.random.randint(0, 2, size=100, dtype=bool)
+    values[1] = ~values[0]
+    return ArrowBoolArray.from_scalars(values)
+
+
 @pytest.fixture
 def data():
     values = np.random.randint(0, 2, size=100, dtype=bool)
@@ -55,6 +62,10 @@ def test_from_dtype(self, data):
     def test_from_sequence_from_cls(self, data):
         super().test_from_sequence_from_cls(data)
 
+    @pytest.mark.xfail(reason="bad is-na for empty data")
+    def test_construct_empty_dataframe(self, columns, dtype):
+        super().test_construct_empty_dataframe(columns, dtype)
+
 
 class TestReduce(base.BaseNoReduceTests):
     def test_reduce_series_boolean(self):

pandas/tests/extension/base/constructors.py

@@ -83,3 +83,7 @@ def test_pandas_array_dtype(self, data):
         result = pd.array(data, dtype=np.dtype(object))
         expected = pd.arrays.PandasArray(np.asarray(data, dtype=object))
         self.assert_equal(result, expected)
+
+    def test_construct_empty_dataframe(self, columns, dtype):
+        # GH 33623
+        pd.DataFrame(columns=columns, dtype=dtype)

pandas/tests/extension/decimal/test_decimal.py

@@ -17,6 +17,11 @@ def dtype():
     return DecimalDtype()
 
 
+@pytest.fixture
+def columns():
+    return DecimalArray(make_data())
+
+
 @pytest.fixture
 def data():
     return DecimalArray(make_data())

pandas/tests/extension/json/test_json.py

@@ -15,6 +15,16 @@ def dtype():
     return JSONDtype()
 
 
+@pytest.fixture
+def columns():
+    data = make_data()
+
+    while len(data[0]) == len(data[1]):
+        data = make_data()
+
+    return JSONArray(data)
+
+
 @pytest.fixture
 def data():
     """Length-100 PeriodArray for semantics test."""

pandas/tests/extension/test_boolean.py

@@ -33,6 +33,11 @@ def dtype():
     return BooleanDtype()
 
 
+@pytest.fixture
+def columns(dtype):
+    return pd.array([True, False], dtype=dtype)
+
+
 @pytest.fixture
 def data(dtype):
     return pd.array(make_data(), dtype=dtype)

pandas/tests/extension/test_categorical.py

@@ -41,6 +41,11 @@ def dtype():
     return CategoricalDtype()
 
 
+@pytest.fixture
+def columns():
+    return Categorical(make_data())
+
+
 @pytest.fixture
 def data():
     """Length-100 array for this type.

pandas/tests/extension/test_common.py

@@ -79,9 +79,3 @@ def test_astype_no_copy():
 def test_is_extension_array_dtype(dtype):
     assert isinstance(dtype, dtypes.ExtensionDtype)
     assert is_extension_array_dtype(dtype)
-
-
-@pytest.mark.parametrize("columns, dtype", [(["a"], "string")])
-def test_construct_empty_dataframe_with_string_dtype(columns, dtype):
-    # GH 33623
-    pd.DataFrame(columns=columns, dtype=dtype)

pandas/tests/extension/test_datetime.py

@@ -13,6 +13,11 @@ def dtype(request):
     return DatetimeTZDtype(unit="ns", tz=request.param)
 
 
+@pytest.fixture
+def columns(dtype):
+    return DatetimeArray(pd.date_range("2000", periods=100, tz=dtype.tz), dtype=dtype)
+
+
 @pytest.fixture
 def data(dtype):
     data = DatetimeArray(pd.date_range("2000", periods=100, tz=dtype.tz), dtype=dtype)

pandas/tests/extension/test_integer.py

@@ -54,6 +54,11 @@ def dtype(request):
     return request.param()
 
 
+@pytest.fixture
+def columns():
+    return integer_array(make_data())
+
+
 @pytest.fixture
 def data(dtype):
     return integer_array(make_data(), dtype=dtype)
@@ -186,7 +191,9 @@ class TestInterface(base.BaseInterfaceTests):
 
 
 class TestConstructors(base.BaseConstructorsTests):
-    pass
+    @pytest.mark.xfail(reason="bad is-na for empty data")
+    def test_construct_empty_dataframe(self, columns, dtyoe):
+        super().test_construct_empty_dataframe(columns, dtype)
 
 
 class TestReshaping(base.BaseReshapingTests):

pandas/tests/extension/test_interval.py

@@ -35,6 +35,11 @@ def dtype():
     return IntervalDtype()
 
 
+@pytest.fixture
+def columns():
+    return IntervalArray(make_data())
+
+
 @pytest.fixture
 def data():
     """Length-100 PeriodArray for semantics test."""

pandas/tests/extension/test_numpy.py

@@ -37,6 +37,11 @@ def allow_in_pandas(monkeypatch):
         yield
 
 
+@pytest.fixture
+def columns():
+    return PandasArray(np.arange(1, 3))
+
+
 @pytest.fixture
 def data(allow_in_pandas, dtype):
     if dtype.numpy_dtype == "object":

pandas/tests/extension/test_period.py

@@ -15,6 +15,11 @@ def dtype():
     return PeriodDtype(freq="D")
 
 
+@pytest.fixture
+def columns(dtype):
+    return PeriodArray(np.arange(2020, 2021), freq=dtype.freq)
+
+
 @pytest.fixture
 def data(dtype):
     return PeriodArray(np.arange(1970, 2070), freq=dtype.freq)

pandas/tests/extension/test_sparse.py

@@ -29,6 +29,11 @@ def dtype():
     return SparseDtype()
 
 
+@pytest.fixture
+def columns():
+    return SparseArray(np.ones(10) * 2)
+
+
 @pytest.fixture(params=[0, np.nan])
 def data(request):
     """Length-100 PeriodArray for semantics test."""

pandas/tests/extension/test_string.py

@@ -13,6 +13,13 @@ def dtype():
     return StringDtype()
 
 
+@pytest.fixture
+def columns():
+    strings = np.random.choice(list(string.ascii_letters), size=1)
+
+    return StringArray._from_sequence(strings)
+
+
 @pytest.fixture
 def data():
     strings = np.random.choice(list(string.ascii_letters), size=100)