CLN: follow-ups, whatsnew notes, use tm.get_obj (#44998)

Author: jbrockmendel

doc/source/whatsnew/v1.4.0.rst

@@ -633,7 +633,7 @@ Datetimelike
 - Bug in adding a ``np.timedelta64`` object to a :class:`BusinessDay` or :class:`CustomBusinessDay` object incorrectly raising (:issue:`44532`)
 - Bug in :meth:`Index.insert` for inserting ``np.datetime64``, ``np.timedelta64`` or ``tuple`` into :class:`Index` with ``dtype='object'`` with negative loc adding ``None`` and replacing existing value (:issue:`44509`)
 - Bug in :meth:`Series.mode` with ``DatetimeTZDtype`` incorrectly returning timezone-naive and ``PeriodDtype`` incorrectly raising (:issue:`41927`)
-- Bug in :class:`DateOffset`` addition with :class:`Timestamp` where ``offset.nanoseconds`` would not be included in the result. (:issue:`43968`)
+- Bug in :class:`DateOffset`` addition with :class:`Timestamp` where ``offset.nanoseconds`` would not be included in the result. (:issue:`43968`,:issue:`36589`)
 -
 
 Timedelta
@@ -845,15 +845,16 @@ Sparse
 ExtensionArray
 ^^^^^^^^^^^^^^
 - Bug in :func:`array` failing to preserve :class:`PandasArray` (:issue:`43887`)
-- NumPy ufuncs ``np.abs``, ``np.positive``, ``np.negative`` now correctly preserve dtype when called on ExtensionArrays that implement ``__abs__, __pos__, __neg__``, respectively. In particular this is fixed for :class:`TimedeltaArray` (:issue:`43899`)
+- NumPy ufuncs ``np.abs``, ``np.positive``, ``np.negative`` now correctly preserve dtype when called on ExtensionArrays that implement ``__abs__, __pos__, __neg__``, respectively. In particular this is fixed for :class:`TimedeltaArray` (:issue:`43899`,:issue:`23316`)
 - NumPy ufuncs ``np.minimum.reduce`` and ``np.maximum.reduce`` now work correctly instead of raising ``NotImplementedError`` on :class:`Series` with ``IntegerDtype`` or ``FloatDtype`` (:issue:`43923`)
 - Avoid raising ``PerformanceWarning`` about fragmented DataFrame when using many columns with an extension dtype (:issue:`44098`)
 - Bug in :class:`IntegerArray` and :class:`FloatingArray` construction incorrectly coercing mismatched NA values (e.g. ``np.timedelta64("NaT")``) to numeric NA (:issue:`44514`)
 - Bug in :meth:`BooleanArray.__eq__` and :meth:`BooleanArray.__ne__` raising ``TypeError`` on comparison with an incompatible type (like a string). This caused :meth:`DataFrame.replace` to sometimes raise a ``TypeError`` if a nullable boolean column was included (:issue:`44499`)
 - Bug in :func:`array` incorrectly raising when passed a ``ndarray`` with ``float16`` dtype (:issue:`44715`)
 - Bug in calling ``np.sqrt`` on :class:`BooleanArray` returning a malformed :class:`FloatingArray` (:issue:`44715`)
 - Bug in :meth:`Series.where` with ``ExtensionDtype`` when ``other`` is a NA scalar incompatible with the series dtype (e.g. ``NaT`` with a numeric dtype) incorrectly casting to a compatible NA value (:issue:`44697`)
-- Fixed bug in :meth:`Series.replace` with ``FloatDtype``, ``string[python]``, or ``string[pyarrow]`` dtype not being preserved when possible (:issue:`33484`)
+- Fixed bug in :meth:`Series.replace` with ``FloatDtype``, ``string[python]``, or ``string[pyarrow]`` dtype not being preserved when possible (:issue:`33484`,:issue:`40732`,:issue:`31644`,:issue:`41215`,:issue:`25438`)
+-
 
 Styler
 ^^^^^^
@@ -874,7 +875,7 @@ Other
 - Bug in :meth:`RangeIndex.union` with another ``RangeIndex`` with matching (even) ``step`` and starts differing by strictly less than ``step / 2`` (:issue:`44019`)
 - Bug in :meth:`RangeIndex.difference` with ``sort=None`` and ``step<0`` failing to sort (:issue:`44085`)
 - Bug in :meth:`Series.to_frame` and :meth:`Index.to_frame` ignoring the ``name`` argument when ``name=None`` is explicitly passed (:issue:`44212`)
-- Bug in :meth:`Series.replace` and :meth:`DataFrame.replace` with ``value=None`` and ExtensionDtypes (:issue:`44270`)
+- Bug in :meth:`Series.replace` and :meth:`DataFrame.replace` with ``value=None`` and ExtensionDtypes (:issue:`44270`,:issue:`37899`)
 - Bug in :meth:`FloatingArray.equals` failing to consider two arrays equal if they contain ``np.nan`` values (:issue:`44382`)
 - Bug in :meth:`DataFrame.shift` with ``axis=1`` and ``ExtensionDtype`` columns incorrectly raising when an incompatible ``fill_value`` is passed (:issue:`44564`)
 - Bug in :meth:`DataFrame.diff` when passing a NumPy integer object instead of an ``int`` object (:issue:`44572`)

pandas/core/dtypes/common.py

@@ -128,10 +128,12 @@ def ensure_python_int(value: int | np.integer) -> int:
     ------
     TypeError: if the value isn't an int or can't be converted to one.
     """
-    if not is_scalar(value):
-        raise TypeError(
-            f"Value needs to be a scalar value, was type {type(value).__name__}"
-        )
+    if not (is_integer(value) or is_float(value)):
+        if not is_scalar(value):
+            raise TypeError(
+                f"Value needs to be a scalar value, was type {type(value).__name__}"
+            )
+        raise TypeError(f"Wrong type {type(value)} for value {value}")
     try:
         new_value = int(value)
         assert new_value == value

pandas/core/indexes/range.py

@@ -1041,11 +1041,7 @@ def _arith_method(self, other, op):
                 rstop = op(left.stop, right)
 
             res_name = ops.get_op_result_name(self, other)
-            with warnings.catch_warnings():
-                warnings.simplefilter("ignore", DeprecationWarning)
-                # The constructor validation can lead to a DeprecationWarning
-                #  from numpy, e.g. with RangeIndex + np.datetime64("now")
-                result = type(self)(rstart, rstop, rstep, name=res_name)
+            result = type(self)(rstart, rstop, rstep, name=res_name)
 
             # for compat with numpy / Int64Index
             # even if we can represent as a RangeIndex, return

pandas/tests/apply/test_frame_transform.py

@@ -141,8 +141,7 @@ def test_transform_bad_dtype(op, frame_or_series, request):
         )
 
     obj = DataFrame({"A": 3 * [object]})  # DataFrame that will fail on most transforms
-    if frame_or_series is not DataFrame:
-        obj = obj["A"]
+    obj = tm.get_obj(obj, frame_or_series)
 
     # tshift is deprecated
     warn = None if op != "tshift" else FutureWarning

pandas/tests/apply/test_invalid_arg.py

@@ -352,8 +352,7 @@ def test_transform_reducer_raises(all_reductions, frame_or_series, op_wrapper):
     op = op_wrapper(all_reductions)
 
     obj = DataFrame({"A": [1, 2, 3]})
-    if frame_or_series is not DataFrame:
-        obj = obj["A"]
+    obj = tm.get_obj(obj, frame_or_series)
 
     msg = "Function did not transform"
     with pytest.raises(ValueError, match=msg):

pandas/tests/frame/indexing/test_mask.py

@@ -96,9 +96,8 @@ def test_mask_pos_args_deprecation(self, frame_or_series):
         # https://github.com/pandas-dev/pandas/issues/41485
         obj = DataFrame({"a": range(5)})
         expected = DataFrame({"a": [-1, 1, -1, 3, -1]})
-        if frame_or_series is Series:
-            obj = obj["a"]
-            expected = expected["a"]
+        obj = tm.get_obj(obj, frame_or_series)
+        expected = tm.get_obj(expected, frame_or_series)
 
         cond = obj % 2 == 0
         msg = (

pandas/tests/frame/indexing/test_where.py

@@ -708,8 +708,7 @@ def test_where_interval_noop(self):
 
 def test_where_try_cast_deprecated(frame_or_series):
     obj = DataFrame(np.random.randn(4, 3))
-    if frame_or_series is not DataFrame:
-        obj = obj[0]
+    obj = tm.get_obj(obj, frame_or_series)
 
     mask = obj > 0
 

pandas/tests/frame/methods/test_append.py

@@ -15,8 +15,7 @@
 class TestDataFrameAppend:
     def test_append_multiindex(self, multiindex_dataframe_random_data, frame_or_series):
         obj = multiindex_dataframe_random_data
-        if frame_or_series is Series:
-            obj = obj["A"]
+        obj = tm.get_obj(obj, frame_or_series)
 
         a = obj[:5]
         b = obj[5:]

pandas/tests/frame/methods/test_asfreq.py

@@ -88,17 +88,15 @@ def test_asfreq_keep_index_name(self, frame_or_series):
         index_name = "bar"
         index = date_range("20130101", periods=20, name=index_name)
         obj = DataFrame(list(range(20)), columns=["foo"], index=index)
-        if frame_or_series is Series:
-            obj = obj["foo"]
+        obj = tm.get_obj(obj, frame_or_series)
 
         assert index_name == obj.index.name
         assert index_name == obj.asfreq("10D").index.name
 
     def test_asfreq_ts(self, frame_or_series):
         index = period_range(freq="A", start="1/1/2001", end="12/31/2010")
         obj = DataFrame(np.random.randn(len(index), 3), index=index)
-        if frame_or_series is Series:
-            obj = obj[0]
+        obj = tm.get_obj(obj, frame_or_series)
 
         result = obj.asfreq("D", how="end")
         exp_index = index.asfreq("D", how="end")
@@ -115,8 +113,7 @@ def test_asfreq_resample_set_correct_freq(self, frame_or_series):
         # we test if .asfreq() and .resample() set the correct value for .freq
         dti = to_datetime(["2012-01-01", "2012-01-02", "2012-01-03"])
         obj = DataFrame({"col": [1, 2, 3]}, index=dti)
-        if frame_or_series is Series:
-            obj = obj["col"]
+        obj = tm.get_obj(obj, frame_or_series)
 
         # testing the settings before calling .asfreq() and .resample()
         assert obj.index.freq is None

pandas/tests/frame/methods/test_at_time.py

@@ -31,8 +31,7 @@ def test_localized_at_time(self, tzstr, frame_or_series):
     def test_at_time(self, frame_or_series):
         rng = date_range("1/1/2000", "1/5/2000", freq="5min")
         ts = DataFrame(np.random.randn(len(rng), 2), index=rng)
-        if frame_or_series is not DataFrame:
-            ts = ts[0]
+        ts = tm.get_obj(ts, frame_or_series)
         rs = ts.at_time(rng[1])
         assert (rs.index.hour == rng[1].hour).all()
         assert (rs.index.minute == rng[1].minute).all()
@@ -46,8 +45,7 @@ def test_at_time_midnight(self, frame_or_series):
         # midnight, everything
         rng = date_range("1/1/2000", "1/31/2000")
         ts = DataFrame(np.random.randn(len(rng), 3), index=rng)
-        if frame_or_series is not DataFrame:
-            ts = ts[0]
+        ts = tm.get_obj(ts, frame_or_series)
 
         result = ts.at_time(time(0, 0))
         tm.assert_equal(result, ts)
@@ -56,8 +54,7 @@ def test_at_time_nonexistent(self, frame_or_series):
         # time doesn't exist
         rng = date_range("1/1/2012", freq="23Min", periods=384)
         ts = DataFrame(np.random.randn(len(rng)), rng)
-        if frame_or_series is not DataFrame:
-            ts = ts[0]
+        ts = tm.get_obj(ts, frame_or_series)
         rs = ts.at_time("16:00")
         assert len(rs) == 0
 
@@ -87,8 +84,7 @@ def test_at_time_tz(self):
     def test_at_time_raises(self, frame_or_series):
         # GH#20725
         obj = DataFrame([[1, 2, 3], [4, 5, 6]])
-        if frame_or_series is not DataFrame:
-            obj = obj[0]
+        obj = tm.get_obj(obj, frame_or_series)
         msg = "Index must be DatetimeIndex"
         with pytest.raises(TypeError, match=msg):  # index is not a DatetimeIndex
             obj.at_time("00:00")

pandas/tests/frame/methods/test_between_time.py

@@ -23,8 +23,7 @@ def test_between_time_formats(self, frame_or_series):
         # GH#11818
         rng = date_range("1/1/2000", "1/5/2000", freq="5min")
         ts = DataFrame(np.random.randn(len(rng), 2), index=rng)
-        if frame_or_series is Series:
-            ts = ts[0]
+        ts = tm.get_obj(ts, frame_or_series)
 
         strings = [
             ("2:00", "2:30"),
@@ -62,8 +61,7 @@ def test_between_time_types(self, frame_or_series):
         # GH11818
         rng = date_range("1/1/2000", "1/5/2000", freq="5min")
         obj = DataFrame({"A": 0}, index=rng)
-        if frame_or_series is Series:
-            obj = obj["A"]
+        obj = tm.get_obj(obj, frame_or_series)
 
         msg = r"Cannot convert arg \[datetime\.datetime\(2010, 1, 2, 1, 0\)\] to a time"
         with pytest.raises(ValueError, match=msg):
@@ -72,8 +70,7 @@ def test_between_time_types(self, frame_or_series):
     def test_between_time(self, inclusive_endpoints_fixture, frame_or_series):
         rng = date_range("1/1/2000", "1/5/2000", freq="5min")
         ts = DataFrame(np.random.randn(len(rng), 2), index=rng)
-        if frame_or_series is not DataFrame:
-            ts = ts[0]
+        ts = tm.get_obj(ts, frame_or_series)
 
         stime = time(0, 0)
         etime = time(1, 0)
@@ -107,8 +104,7 @@ def test_between_time(self, inclusive_endpoints_fixture, frame_or_series):
         # across midnight
         rng = date_range("1/1/2000", "1/5/2000", freq="5min")
         ts = DataFrame(np.random.randn(len(rng), 2), index=rng)
-        if frame_or_series is not DataFrame:
-            ts = ts[0]
+        ts = tm.get_obj(ts, frame_or_series)
         stime = time(22, 0)
         etime = time(9, 0)
 
@@ -135,8 +131,7 @@ def test_between_time(self, inclusive_endpoints_fixture, frame_or_series):
     def test_between_time_raises(self, frame_or_series):
         # GH#20725
         obj = DataFrame([[1, 2, 3], [4, 5, 6]])
-        if frame_or_series is not DataFrame:
-            obj = obj[0]
+        obj = tm.get_obj(obj, frame_or_series)
 
         msg = "Index must be DatetimeIndex"
         with pytest.raises(TypeError, match=msg):  # index is not a DatetimeIndex
@@ -215,8 +210,7 @@ def test_between_time_warn(self, include_start, include_end, frame_or_series):
         # GH40245
         rng = date_range("1/1/2000", "1/5/2000", freq="5min")
         ts = DataFrame(np.random.randn(len(rng), 2), index=rng)
-        if frame_or_series is not DataFrame:
-            ts = ts[0]
+        ts = tm.get_obj(ts, frame_or_series)
 
         stime = time(0, 0)
         etime = time(1, 0)

pandas/tests/frame/methods/test_first_and_last.py

@@ -13,14 +13,12 @@
 class TestFirst:
     def test_first_subset(self, frame_or_series):
         ts = tm.makeTimeDataFrame(freq="12h")
-        if frame_or_series is not DataFrame:
-            ts = ts["A"]
+        ts = tm.get_obj(ts, frame_or_series)
         result = ts.first("10d")
         assert len(result) == 20
 
         ts = tm.makeTimeDataFrame(freq="D")
-        if frame_or_series is not DataFrame:
-            ts = ts["A"]
+        ts = tm.get_obj(ts, frame_or_series)
         result = ts.first("10d")
         assert len(result) == 10
 
@@ -38,8 +36,7 @@ def test_first_subset(self, frame_or_series):
     def test_first_last_raises(self, frame_or_series):
         # GH#20725
         obj = DataFrame([[1, 2, 3], [4, 5, 6]])
-        if frame_or_series is not DataFrame:
-            obj = obj[0]
+        obj = tm.get_obj(obj, frame_or_series)
 
         msg = "'first' only supports a DatetimeIndex index"
         with pytest.raises(TypeError, match=msg):  # index is not a DatetimeIndex
@@ -51,14 +48,12 @@ def test_first_last_raises(self, frame_or_series):
 
     def test_last_subset(self, frame_or_series):
         ts = tm.makeTimeDataFrame(freq="12h")
-        if frame_or_series is not DataFrame:
-            ts = ts["A"]
+        ts = tm.get_obj(ts, frame_or_series)
         result = ts.last("10d")
         assert len(result) == 20
 
         ts = tm.makeTimeDataFrame(nper=30, freq="D")
-        if frame_or_series is not DataFrame:
-            ts = ts["A"]
+        ts = tm.get_obj(ts, frame_or_series)
         result = ts.last("10d")
         assert len(result) == 10
 

pandas/tests/frame/methods/test_pipe.py

@@ -21,17 +21,15 @@ def test_pipe(self, frame_or_series):
 
     def test_pipe_tuple(self, frame_or_series):
         obj = DataFrame({"A": [1, 2, 3]})
-        if frame_or_series is Series:
-            obj = obj["A"]
+        obj = tm.get_obj(obj, frame_or_series)
 
         f = lambda x, y: y
         result = obj.pipe((f, "y"), 0)
         tm.assert_equal(result, obj)
 
     def test_pipe_tuple_error(self, frame_or_series):
         obj = DataFrame({"A": [1, 2, 3]})
-        if frame_or_series is Series:
-            obj = obj["A"]
+        obj = tm.get_obj(obj, frame_or_series)
 
         f = lambda x, y: y
 

pandas/tests/frame/methods/test_reorder_levels.py

@@ -16,7 +16,7 @@ def test_reorder_levels(self, frame_or_series):
             names=["L0", "L1", "L2"],
         )
         df = DataFrame({"A": np.arange(6), "B": np.arange(6)}, index=index)
-        obj = df if frame_or_series is DataFrame else df["A"]
+        obj = tm.get_obj(df, frame_or_series)
 
         # no change, position
         result = obj.reorder_levels([0, 1, 2])
@@ -34,7 +34,7 @@ def test_reorder_levels(self, frame_or_series):
             names=["L1", "L2", "L0"],
         )
         expected = DataFrame({"A": np.arange(6), "B": np.arange(6)}, index=e_idx)
-        expected = expected if frame_or_series is DataFrame else expected["A"]
+        expected = tm.get_obj(expected, frame_or_series)
         tm.assert_equal(result, expected)
 
         result = obj.reorder_levels([0, 0, 0])
@@ -44,7 +44,7 @@ def test_reorder_levels(self, frame_or_series):
             names=["L0", "L0", "L0"],
         )
         expected = DataFrame({"A": np.arange(6), "B": np.arange(6)}, index=e_idx)
-        expected = expected if frame_or_series is DataFrame else expected["A"]
+        expected = tm.get_obj(expected, frame_or_series)
         tm.assert_equal(result, expected)
 
         result = obj.reorder_levels(["L0", "L0", "L0"])

pandas/tests/frame/methods/test_replace.py

@@ -483,8 +483,7 @@ def test_replace_with_empty_list(self, frame_or_series):
         # GH 21977
         ser = Series([["a", "b"], [], np.nan, [1]])
         obj = DataFrame({"col": ser})
-        if frame_or_series is Series:
-            obj = ser
+        obj = tm.get_obj(obj, frame_or_series)
         expected = obj
         result = obj.replace([], np.nan)
         tm.assert_equal(result, expected)
@@ -1332,8 +1331,7 @@ def test_replace_with_duplicate_columns(self, replacement):
     def test_replace_ea_ignore_float(self, frame_or_series, value):
         # GH#34871
         obj = DataFrame({"Per": [value] * 3})
-        if frame_or_series is not DataFrame:
-            obj = obj["Per"]
+        obj = tm.get_obj(obj, frame_or_series)
 
         expected = obj.copy()
         result = obj.replace(1.0, 0.0)

pandas/tests/frame/methods/test_sample.py

@@ -166,8 +166,7 @@ def test_sample_none_weights(self, obj):
     def test_sample_random_state(self, func_str, arg, frame_or_series):
         # GH#32503
         obj = DataFrame({"col1": range(10, 20), "col2": range(20, 30)})
-        if frame_or_series is Series:
-            obj = obj["col1"]
+        obj = tm.get_obj(obj, frame_or_series)
         result = obj.sample(n=3, random_state=eval(func_str)(arg))
         expected = obj.sample(n=3, random_state=com.random_state(eval(func_str)(arg)))
         tm.assert_equal(result, expected)
@@ -192,8 +191,7 @@ def test_sample_upsampling_without_replacement(self, frame_or_series):
         # GH#27451
 
         obj = DataFrame({"A": list("abc")})
-        if frame_or_series is Series:
-            obj = obj["A"]
+        obj = tm.get_obj(obj, frame_or_series)
 
         msg = (
             "Replace has to be set to `True` when "

pandas/tests/frame/methods/test_to_period.py

@@ -21,8 +21,7 @@ def test_to_period(self, frame_or_series):
             np.random.randn(len(dr), K), index=dr, columns=["A", "B", "C", "D", "E"]
         )
         obj["mix"] = "a"
-        if frame_or_series is Series:
-            obj = obj["A"]
+        obj = tm.get_obj(obj, frame_or_series)
 
         pts = obj.to_period()
         exp = obj.copy()
@@ -41,8 +40,7 @@ def test_to_period_without_freq(self, frame_or_series):
         )
 
         obj = DataFrame(np.random.randn(4, 4), index=idx, columns=idx)
-        if frame_or_series is Series:
-            obj = obj[idx[0]]
+        obj = tm.get_obj(obj, frame_or_series)
         expected = obj.copy()
         expected.index = exp_idx
         tm.assert_equal(obj.to_period(), expected)