TST/REF: collect tests by method from tests.internals (#37420)

Author: jbrockmendel

pandas/tests/frame/methods/test_equals.py

@@ -0,0 +1,23 @@
+from pandas import DataFrame
+import pandas._testing as tm
+
+
+class TestEquals:
+    def test_dataframe_not_equal(self):
+        # see GH#28839
+        df1 = DataFrame({"a": [1, 2], "b": ["s", "d"]})
+        df2 = DataFrame({"a": ["s", "d"], "b": [1, 2]})
+        assert df1.equals(df2) is False
+
+    def test_equals_different_blocks(self):
+        # GH#9330
+        df0 = DataFrame({"A": ["x", "y"], "B": [1, 2], "C": ["w", "z"]})
+        df1 = df0.reset_index()[["A", "B", "C"]]
+        # this assert verifies that the above operations have
+        # induced a block rearrangement
+        assert df0._mgr.blocks[0].dtype != df1._mgr.blocks[0].dtype
+
+        # do the real tests
+        tm.assert_frame_equal(df0, df1)
+        assert df0.equals(df1)
+        assert df1.equals(df0)

pandas/tests/frame/methods/test_infer_objects.py

@@ -0,0 +1,42 @@
+from datetime import datetime
+
+from pandas import DataFrame
+import pandas._testing as tm
+
+
+class TestInferObjects:
+    def test_infer_objects(self):
+        # GH#11221
+        df = DataFrame(
+            {
+                "a": ["a", 1, 2, 3],
+                "b": ["b", 2.0, 3.0, 4.1],
+                "c": [
+                    "c",
+                    datetime(2016, 1, 1),
+                    datetime(2016, 1, 2),
+                    datetime(2016, 1, 3),
+                ],
+                "d": [1, 2, 3, "d"],
+            },
+            columns=["a", "b", "c", "d"],
+        )
+        df = df.iloc[1:].infer_objects()
+
+        assert df["a"].dtype == "int64"
+        assert df["b"].dtype == "float64"
+        assert df["c"].dtype == "M8[ns]"
+        assert df["d"].dtype == "object"
+
+        expected = DataFrame(
+            {
+                "a": [1, 2, 3],
+                "b": [2.0, 3.0, 4.1],
+                "c": [datetime(2016, 1, 1), datetime(2016, 1, 2), datetime(2016, 1, 3)],
+                "d": [2, 3, "d"],
+            },
+            columns=["a", "b", "c", "d"],
+        )
+        # reconstruct frame to verify inference is same
+        result = df.reset_index(drop=True)
+        tm.assert_frame_equal(result, expected)

pandas/tests/frame/methods/test_to_dict_of_blocks.py

@@ -0,0 +1,52 @@
+import numpy as np
+
+from pandas import DataFrame
+from pandas.core.arrays import PandasArray
+
+
+class TestToDictOfBlocks:
+    def test_copy_blocks(self, float_frame):
+        # GH#9607
+        df = DataFrame(float_frame, copy=True)
+        column = df.columns[0]
+
+        # use the default copy=True, change a column
+        blocks = df._to_dict_of_blocks(copy=True)
+        for dtype, _df in blocks.items():
+            if column in _df:
+                _df.loc[:, column] = _df[column] + 1
+
+        # make sure we did not change the original DataFrame
+        assert not _df[column].equals(df[column])
+
+    def test_no_copy_blocks(self, float_frame):
+        # GH#9607
+        df = DataFrame(float_frame, copy=True)
+        column = df.columns[0]
+
+        # use the copy=False, change a column
+        blocks = df._to_dict_of_blocks(copy=False)
+        for dtype, _df in blocks.items():
+            if column in _df:
+                _df.loc[:, column] = _df[column] + 1
+
+        # make sure we did change the original DataFrame
+        assert _df[column].equals(df[column])
+
+
+def test_to_dict_of_blocks_item_cache():
+    # Calling to_dict_of_blocks should not poison item_cache
+    df = DataFrame({"a": [1, 2, 3, 4], "b": ["a", "b", "c", "d"]})
+    df["c"] = PandasArray(np.array([1, 2, None, 3], dtype=object))
+    mgr = df._mgr
+    assert len(mgr.blocks) == 3  # i.e. not consolidated
+
+    ser = df["b"]  # populations item_cache["b"]
+
+    df._to_dict_of_blocks()
+
+    # Check that the to_dict_of_blocks didnt break link between ser and df
+    ser.values[0] = "foo"
+    assert df.loc[0, "b"] == "foo"
+
+    assert df["b"] is ser

pandas/tests/frame/methods/test_values.py

@@ -102,3 +102,62 @@ def test_interleave_with_tzaware(self, timezone_frame):
             dtype=object,
         ).T
         tm.assert_numpy_array_equal(result, expected)
+
+    def test_values_interleave_non_unique_cols(self):
+        df = DataFrame(
+            [[Timestamp("20130101"), 3.5], [Timestamp("20130102"), 4.5]],
+            columns=["x", "x"],
+            index=[1, 2],
+        )
+
+        df_unique = df.copy()
+        df_unique.columns = ["x", "y"]
+        assert df_unique.values.shape == df.values.shape
+        tm.assert_numpy_array_equal(df_unique.values[0], df.values[0])
+        tm.assert_numpy_array_equal(df_unique.values[1], df.values[1])
+
+    def test_values_numeric_cols(self, float_frame):
+        float_frame["foo"] = "bar"
+
+        values = float_frame[["A", "B", "C", "D"]].values
+        assert values.dtype == np.float64
+
+    def test_values_lcd(self, mixed_float_frame, mixed_int_frame):
+
+        # mixed lcd
+        values = mixed_float_frame[["A", "B", "C", "D"]].values
+        assert values.dtype == np.float64
+
+        values = mixed_float_frame[["A", "B", "C"]].values
+        assert values.dtype == np.float32
+
+        values = mixed_float_frame[["C"]].values
+        assert values.dtype == np.float16
+
+        # GH#10364
+        # B uint64 forces float because there are other signed int types
+        values = mixed_int_frame[["A", "B", "C", "D"]].values
+        assert values.dtype == np.float64
+
+        values = mixed_int_frame[["A", "D"]].values
+        assert values.dtype == np.int64
+
+        # B uint64 forces float because there are other signed int types
+        values = mixed_int_frame[["A", "B", "C"]].values
+        assert values.dtype == np.float64
+
+        # as B and C are both unsigned, no forcing to float is needed
+        values = mixed_int_frame[["B", "C"]].values
+        assert values.dtype == np.uint64
+
+        values = mixed_int_frame[["A", "C"]].values
+        assert values.dtype == np.int32
+
+        values = mixed_int_frame[["C", "D"]].values
+        assert values.dtype == np.int64
+
+        values = mixed_int_frame[["A"]].values
+        assert values.dtype == np.int32
+
+        values = mixed_int_frame[["C"]].values
+        assert values.dtype == np.uint8

pandas/tests/frame/test_arithmetic.py

@@ -14,6 +14,32 @@
 from pandas.core.computation.expressions import _MIN_ELEMENTS, NUMEXPR_INSTALLED
 from pandas.tests.frame.common import _check_mixed_float, _check_mixed_int
 
+
+class DummyElement:
+    def __init__(self, value, dtype):
+        self.value = value
+        self.dtype = np.dtype(dtype)
+
+    def __array__(self):
+        return np.array(self.value, dtype=self.dtype)
+
+    def __str__(self) -> str:
+        return f"DummyElement({self.value}, {self.dtype})"
+
+    def __repr__(self) -> str:
+        return str(self)
+
+    def astype(self, dtype, copy=False):
+        self.dtype = dtype
+        return self
+
+    def view(self, dtype):
+        return type(self)(self.value.view(dtype), dtype)
+
+    def any(self, axis=None):
+        return bool(self.value)
+
+
 # -------------------------------------------------------------------
 # Comparisons
 
@@ -782,6 +808,77 @@ def test_frame_with_frame_reindex(self):
         )
         tm.assert_frame_equal(result, expected)
 
+    @pytest.mark.parametrize(
+        "value, dtype",
+        [
+            (1, "i8"),
+            (1.0, "f8"),
+            (2 ** 63, "f8"),
+            (1j, "complex128"),
+            (2 ** 63, "complex128"),
+            (True, "bool"),
+            (np.timedelta64(20, "ns"), "<m8[ns]"),
+            (np.datetime64(20, "ns"), "<M8[ns]"),
+        ],
+    )
+    @pytest.mark.parametrize(
+        "op",
+        [
+            operator.add,
+            operator.sub,
+            operator.mul,
+            operator.truediv,
+            operator.mod,
+            operator.pow,
+        ],
+        ids=lambda x: x.__name__,
+    )
+    def test_binop_other(self, op, value, dtype):
+        skip = {
+            (operator.add, "bool"),
+            (operator.sub, "bool"),
+            (operator.mul, "bool"),
+            (operator.truediv, "bool"),
+            (operator.mod, "i8"),
+            (operator.mod, "complex128"),
+            (operator.pow, "bool"),
+        }
+        if (op, dtype) in skip:
+            pytest.skip(f"Invalid combination {op},{dtype}")
+
+        e = DummyElement(value, dtype)
+        s = DataFrame({"A": [e.value, e.value]}, dtype=e.dtype)
+
+        invalid = {
+            (operator.pow, "<M8[ns]"),
+            (operator.mod, "<M8[ns]"),
+            (operator.truediv, "<M8[ns]"),
+            (operator.mul, "<M8[ns]"),
+            (operator.add, "<M8[ns]"),
+            (operator.pow, "<m8[ns]"),
+            (operator.mul, "<m8[ns]"),
+        }
+
+        if (op, dtype) in invalid:
+            msg = (
+                None
+                if (dtype == "<M8[ns]" and op == operator.add)
+                or (dtype == "<m8[ns]" and op == operator.mul)
+                else (
+                    f"cannot perform __{op.__name__}__ with this "
+                    "index type: (DatetimeArray|TimedeltaArray)"
+                )
+            )
+
+            with pytest.raises(TypeError, match=msg):
+                op(s, e.value)
+        else:
+            # FIXME: Since dispatching to Series, this test no longer
+            # asserts anything meaningful
+            result = op(s, e.value).dtypes
+            expected = op(s, value).dtypes
+            tm.assert_series_equal(result, expected)
+
 
 def test_frame_with_zero_len_series_corner_cases():
     # GH#28600