TYP/CLN: assorted cleanups (pandas-dev#38959)

Author: jbrockmendel

pandas/_libs/tslibs/offsets.pyx

@@ -3565,7 +3565,7 @@ cpdef to_offset(freq):
             f"to_offset does not support tuples {freq}, pass as a string instead"
         )
 
-    elif isinstance(freq, timedelta):
+    elif PyDelta_Check(freq):
         return delta_to_tick(freq)
 
     elif isinstance(freq, str):

pandas/conftest.py

@@ -1002,14 +1002,6 @@ def tz_aware_fixture(request):
 tz_aware_fixture2 = tz_aware_fixture
 
 
-@pytest.fixture(scope="module")
-def datetime_tz_utc():
-    """
-    Yields the UTC timezone object from the datetime module.
-    """
-    return timezone.utc
-
-
 @pytest.fixture(params=["utc", "dateutil/UTC", utc, tzutc(), timezone.utc])
 def utc_fixture(request):
     """
@@ -1189,7 +1181,7 @@ def any_nullable_int_dtype(request):
 
 
 @pytest.fixture(params=tm.ALL_EA_INT_DTYPES + tm.FLOAT_EA_DTYPES)
-def any_numeric_dtype(request):
+def any_nullable_numeric_dtype(request):
     """
     Parameterized fixture for any nullable integer dtype and
     any float ea dtypes.

pandas/core/internals/blocks.py

@@ -658,7 +658,7 @@ def _astype(self, dtype: DtypeObj, copy: bool) -> ArrayLike:
             values = values.astype(dtype, copy=copy)
 
         else:
-            values = astype_nansafe(values, dtype, copy=True)
+            values = astype_nansafe(values, dtype, copy=copy)
 
         return values
 

pandas/core/internals/concat.py

@@ -419,7 +419,7 @@ def _get_empty_dtype_and_na(join_units: Sequence[JoinUnit]) -> Tuple[DtypeObj, A
         return np.dtype("M8[ns]"), np.datetime64("NaT", "ns")
     elif "timedelta" in upcast_classes:
         return np.dtype("m8[ns]"), np.timedelta64("NaT", "ns")
-    else:  # pragma
+    else:
         try:
             common_dtype = np.find_common_type(upcast_classes, [])
         except TypeError:

pandas/core/series.py

@@ -4485,7 +4485,7 @@ def replace(
             method=method,
         )
 
-    def _replace_single(self, to_replace, method, inplace, limit):
+    def _replace_single(self, to_replace, method: str, inplace: bool, limit):
         """
         Replaces values in a Series using the fill method specified when no
         replacement value is given in the replace method

pandas/io/pytables.py

@@ -80,6 +80,8 @@
 if TYPE_CHECKING:
     from tables import Col, File, Node
 
+    from pandas.core.internals import Block
+
 
 # versioning attribute
 _version = "0.15.2"
@@ -3860,20 +3862,17 @@ def _create_axes(
         for a in new_non_index_axes:
             obj = _reindex_axis(obj, a[0], a[1])
 
-        def get_blk_items(mgr, blocks):
-            return [mgr.items.take(blk.mgr_locs) for blk in blocks]
-
         transposed = new_index.axis == 1
 
         # figure out data_columns and get out blocks
         data_columns = self.validate_data_columns(
             data_columns, min_itemsize, new_non_index_axes
         )
 
-        block_obj = self.get_object(obj, transposed)._consolidate()
+        frame = self.get_object(obj, transposed)._consolidate()
 
         blocks, blk_items = self._get_blocks_and_items(
-            block_obj, table_exists, new_non_index_axes, self.values_axes, data_columns
+            frame, table_exists, new_non_index_axes, self.values_axes, data_columns
         )
 
         # add my values
@@ -3978,35 +3977,39 @@ def get_blk_items(mgr, blocks):
 
     @staticmethod
     def _get_blocks_and_items(
-        block_obj, table_exists, new_non_index_axes, values_axes, data_columns
+        frame: DataFrame,
+        table_exists: bool,
+        new_non_index_axes,
+        values_axes,
+        data_columns,
     ):
         # Helper to clarify non-state-altering parts of _create_axes
 
-        def get_blk_items(mgr, blocks):
-            return [mgr.items.take(blk.mgr_locs) for blk in blocks]
+        def get_blk_items(mgr):
+            return [mgr.items.take(blk.mgr_locs) for blk in mgr.blocks]
 
-        blocks = block_obj._mgr.blocks
-        blk_items = get_blk_items(block_obj._mgr, blocks)
+        blocks: List["Block"] = list(frame._mgr.blocks)
+        blk_items: List[Index] = get_blk_items(frame._mgr)
 
         if len(data_columns):
             axis, axis_labels = new_non_index_axes[0]
             new_labels = Index(axis_labels).difference(Index(data_columns))
-            mgr = block_obj.reindex(new_labels, axis=axis)._mgr
+            mgr = frame.reindex(new_labels, axis=axis)._mgr
 
             blocks = list(mgr.blocks)
-            blk_items = get_blk_items(mgr, blocks)
+            blk_items = get_blk_items(mgr)
             for c in data_columns:
-                mgr = block_obj.reindex([c], axis=axis)._mgr
+                mgr = frame.reindex([c], axis=axis)._mgr
                 blocks.extend(mgr.blocks)
-                blk_items.extend(get_blk_items(mgr, mgr.blocks))
+                blk_items.extend(get_blk_items(mgr))
 
         # reorder the blocks in the same order as the existing table if we can
         if table_exists:
             by_items = {
                 tuple(b_items.tolist()): (b, b_items)
                 for b, b_items in zip(blocks, blk_items)
             }
-            new_blocks = []
+            new_blocks: List["Block"] = []
             new_blk_items = []
             for ea in values_axes:
                 items = tuple(ea.values)
@@ -4875,7 +4878,7 @@ def _unconvert_index(
 
 
 def _maybe_convert_for_string_atom(
-    name: str, block, existing_col, min_itemsize, nan_rep, encoding, errors
+    name: str, block: "Block", existing_col, min_itemsize, nan_rep, encoding, errors
 ):
     if not block.is_object:
         return block.values
@@ -4895,11 +4898,12 @@ def _maybe_convert_for_string_atom(
     elif not (inferred_type == "string" or dtype_name == "object"):
         return block.values
 
-    block = block.fillna(nan_rep, downcast=False)
-    if isinstance(block, list):
-        # Note: because block is always object dtype, fillna goes
-        #  through a path such that the result is always a 1-element list
-        block = block[0]
+    blocks: List["Block"] = block.fillna(nan_rep, downcast=False)
+    # Note: because block is always object dtype, fillna goes
+    #  through a path such that the result is always a 1-element list
+    assert len(blocks) == 1
+    block = blocks[0]
+
     data = block.values
 
     # see if we have a valid string type

pandas/tests/indexes/datetimes/test_timezones.py

@@ -1159,7 +1159,6 @@ def test_dti_union_mixed(self):
     @pytest.mark.parametrize(
         "tz", [None, "UTC", "US/Central", dateutil.tz.tzoffset(None, -28800)]
     )
-    @pytest.mark.usefixtures("datetime_tz_utc")
     def test_iteration_preserves_nanoseconds(self, tz):
         # GH 19603
         index = DatetimeIndex(

pandas/tests/indexes/timedeltas/test_partial_slicing.py

@@ -1,5 +1,4 @@
 import numpy as np
-import pytest
 
 from pandas import Series, timedelta_range
 import pandas._testing as tm
@@ -22,13 +21,6 @@ def test_partial_slice(self):
         expected = s.iloc[:134]
         tm.assert_series_equal(result, expected)
 
-        result = s["6 days, 23:11:12"]
-        assert result == s.iloc[133]
-
-        msg = r"^Timedelta\('50 days 00:00:00'\)$"
-        with pytest.raises(KeyError, match=msg):
-            s["50 days"]
-
     def test_partial_slice_high_reso(self):
 
         # higher reso

pandas/tests/series/indexing/test_getitem.py

@@ -18,6 +18,7 @@
     Timestamp,
     date_range,
     period_range,
+    timedelta_range,
 )
 import pandas._testing as tm
 from pandas.core.indexing import IndexingError
@@ -121,6 +122,23 @@ def test_getitem_scalar_categorical_index(self):
         result = ser[cats[0]]
         assert result == expected
 
+    def test_getitem_str_with_timedeltaindex(self):
+        rng = timedelta_range("1 day 10:11:12", freq="h", periods=500)
+        ser = Series(np.arange(len(rng)), index=rng)
+
+        key = "6 days, 23:11:12"
+        indexer = rng.get_loc(key)
+        assert indexer == 133
+
+        result = ser[key]
+        assert result == ser.iloc[133]
+
+        msg = r"^Timedelta\('50 days 00:00:00'\)$"
+        with pytest.raises(KeyError, match=msg):
+            rng.get_loc("50 days")
+        with pytest.raises(KeyError, match=msg):
+            ser["50 days"]
+
 
 class TestSeriesGetitemSlices:
     def test_getitem_partial_str_slice_with_datetimeindex(self):

pandas/tests/series/indexing/test_setitem.py

@@ -167,19 +167,19 @@ def test_setitem_boolean_td64_values_cast_na(self, value):
         expected = Series([NaT, 1, 2], dtype="timedelta64[ns]")
         tm.assert_series_equal(series, expected)
 
-    def test_setitem_boolean_nullable_int_types(self, any_numeric_dtype):
+    def test_setitem_boolean_nullable_int_types(self, any_nullable_numeric_dtype):
         # GH: 26468
-        ser = Series([5, 6, 7, 8], dtype=any_numeric_dtype)
-        ser[ser > 6] = Series(range(4), dtype=any_numeric_dtype)
-        expected = Series([5, 6, 2, 3], dtype=any_numeric_dtype)
+        ser = Series([5, 6, 7, 8], dtype=any_nullable_numeric_dtype)
+        ser[ser > 6] = Series(range(4), dtype=any_nullable_numeric_dtype)
+        expected = Series([5, 6, 2, 3], dtype=any_nullable_numeric_dtype)
         tm.assert_series_equal(ser, expected)
 
-        ser = Series([5, 6, 7, 8], dtype=any_numeric_dtype)
-        ser.loc[ser > 6] = Series(range(4), dtype=any_numeric_dtype)
+        ser = Series([5, 6, 7, 8], dtype=any_nullable_numeric_dtype)
+        ser.loc[ser > 6] = Series(range(4), dtype=any_nullable_numeric_dtype)
         tm.assert_series_equal(ser, expected)
 
-        ser = Series([5, 6, 7, 8], dtype=any_numeric_dtype)
-        loc_ser = Series(range(4), dtype=any_numeric_dtype)
+        ser = Series([5, 6, 7, 8], dtype=any_nullable_numeric_dtype)
+        loc_ser = Series(range(4), dtype=any_nullable_numeric_dtype)
         ser.loc[ser > 6] = loc_ser.loc[loc_ser > 1]
         tm.assert_series_equal(ser, expected)
 

pandas/tests/series/test_constructors.py

@@ -272,8 +272,8 @@ def test_constructor_index_dtype(self, dtype):
         [
             ([1, 2]),
             (["1", "2"]),
-            (list(pd.date_range("1/1/2011", periods=2, freq="H"))),
-            (list(pd.date_range("1/1/2011", periods=2, freq="H", tz="US/Eastern"))),
+            (list(date_range("1/1/2011", periods=2, freq="H"))),
+            (list(date_range("1/1/2011", periods=2, freq="H", tz="US/Eastern"))),
             ([Interval(left=0, right=5)]),
         ],
     )
@@ -628,10 +628,10 @@ def test_constructor_copy(self):
     @pytest.mark.parametrize(
         "index",
         [
-            pd.date_range("20170101", periods=3, tz="US/Eastern"),
-            pd.date_range("20170101", periods=3),
-            pd.timedelta_range("1 day", periods=3),
-            pd.period_range("2012Q1", periods=3, freq="Q"),
+            date_range("20170101", periods=3, tz="US/Eastern"),
+            date_range("20170101", periods=3),
+            timedelta_range("1 day", periods=3),
+            period_range("2012Q1", periods=3, freq="Q"),
             Index(list("abc")),
             pd.Int64Index([1, 2, 3]),
             RangeIndex(0, 3),
@@ -1038,16 +1038,16 @@ def test_construction_consistency(self):
 
         # make sure that we are not re-localizing upon construction
         # GH 14928
-        s = Series(pd.date_range("20130101", periods=3, tz="US/Eastern"))
+        ser = Series(date_range("20130101", periods=3, tz="US/Eastern"))
 
-        result = Series(s, dtype=s.dtype)
-        tm.assert_series_equal(result, s)
+        result = Series(ser, dtype=ser.dtype)
+        tm.assert_series_equal(result, ser)
 
-        result = Series(s.dt.tz_convert("UTC"), dtype=s.dtype)
-        tm.assert_series_equal(result, s)
+        result = Series(ser.dt.tz_convert("UTC"), dtype=ser.dtype)
+        tm.assert_series_equal(result, ser)
 
-        result = Series(s.values, dtype=s.dtype)
-        tm.assert_series_equal(result, s)
+        result = Series(ser.values, dtype=ser.dtype)
+        tm.assert_series_equal(result, ser)
 
     @pytest.mark.parametrize(
         "data_constructor", [list, np.array], ids=["list", "ndarray[object]"]
@@ -1374,7 +1374,7 @@ def test_convert_non_ns(self):
         # convert from a numpy array of non-ns timedelta64
         arr = np.array([1, 2, 3], dtype="timedelta64[s]")
         s = Series(arr)
-        expected = Series(pd.timedelta_range("00:00:01", periods=3, freq="s"))
+        expected = Series(timedelta_range("00:00:01", periods=3, freq="s"))
         tm.assert_series_equal(s, expected)
 
         # convert from a numpy array of non-ns datetime64