CLN: Assorted (#50433)

* CLN: Assorted

* troubleshoot

* troubleshoot

* troubleshoot

* restore

* restore

* troubleshoot

* revert copy=False

* troubleshoot

* remove warning filter

Author: jbrockmendel

pandas/_libs/intervaltree.pxi.in

@@ -121,9 +121,8 @@ cdef class IntervalTree(IntervalMixin):
         """
         if self._na_count > 0:
             return False
-        values = [self.right, self.left]
 
-        sort_order = np.lexsort(values)
+        sort_order = self.left_sorter
         return is_monotonic(sort_order, False)[0]
 
     def get_indexer(self, scalar_t[:] target) -> np.ndarray:

pandas/_libs/tslibs/offsets.pyi

@@ -12,6 +12,7 @@ from typing import (
 
 import numpy as np
 
+from pandas._libs.tslibs.nattype import NaTType
 from pandas._typing import npt
 
 from .timedeltas import Timedelta
@@ -51,6 +52,8 @@ class BaseOffset:
     def __radd__(self, other: _DatetimeT) -> _DatetimeT: ...
     @overload
     def __radd__(self, other: _TimedeltaT) -> _TimedeltaT: ...
+    @overload
+    def __radd__(self, other: NaTType) -> NaTType: ...
     def __sub__(self: _BaseOffsetT, other: BaseOffset) -> _BaseOffsetT: ...
     @overload
     def __rsub__(self, other: npt.NDArray[np.object_]) -> npt.NDArray[np.object_]: ...

pandas/_libs/tslibs/util.pxd

@@ -83,7 +83,7 @@ cdef inline bint is_integer_object(object obj) nogil:
 
 cdef inline bint is_float_object(object obj) nogil:
     """
-    Cython equivalent of `isinstance(val, (float, np.complex_))`
+    Cython equivalent of `isinstance(val, (float, np.float_))`
 
     Parameters
     ----------

pandas/core/accessor.py

@@ -6,6 +6,7 @@
 """
 from __future__ import annotations
 
+from typing import final
 import warnings
 
 from pandas.util._decorators import doc
@@ -16,6 +17,7 @@ class DirNamesMixin:
     _accessors: set[str] = set()
     _hidden_attrs: frozenset[str] = frozenset()
 
+    @final
     def _dir_deletions(self) -> set[str]:
         """
         Delete unwanted __dir__ for this object.

pandas/core/arrays/datetimelike.py

@@ -189,7 +189,7 @@ class DatetimeLikeArrayMixin(OpsMixin, NDArrayBackedExtensionArray):
     Shared Base/Mixin class for DatetimeArray, TimedeltaArray, PeriodArray
 
     Assumes that __new__/__init__ defines:
-        _data
+        _ndarray
         _freq
 
     and that the inheriting class has methods:
@@ -1418,9 +1418,8 @@ def __add__(self, other):
             # as is_integer returns True for these
             if not is_period_dtype(self.dtype):
                 raise integer_op_not_supported(self)
-            result = cast("PeriodArray", self)._addsub_int_array_or_scalar(
-                other * self.freq.n, operator.add
-            )
+            obj = cast("PeriodArray", self)
+            result = obj._addsub_int_array_or_scalar(other * obj.freq.n, operator.add)
 
         # array-like others
         elif is_timedelta64_dtype(other_dtype):
@@ -1435,9 +1434,8 @@ def __add__(self, other):
         elif is_integer_dtype(other_dtype):
             if not is_period_dtype(self.dtype):
                 raise integer_op_not_supported(self)
-            result = cast("PeriodArray", self)._addsub_int_array_or_scalar(
-                other * self.freq.n, operator.add
-            )
+            obj = cast("PeriodArray", self)
+            result = obj._addsub_int_array_or_scalar(other * obj.freq.n, operator.add)
         else:
             # Includes Categorical, other ExtensionArrays
             # For PeriodDtype, if self is a TimedeltaArray and other is a
@@ -1477,9 +1475,8 @@ def __sub__(self, other):
             # as is_integer returns True for these
             if not is_period_dtype(self.dtype):
                 raise integer_op_not_supported(self)
-            result = cast("PeriodArray", self)._addsub_int_array_or_scalar(
-                other * self.freq.n, operator.sub
-            )
+            obj = cast("PeriodArray", self)
+            result = obj._addsub_int_array_or_scalar(other * obj.freq.n, operator.sub)
 
         elif isinstance(other, Period):
             result = self._sub_periodlike(other)
@@ -1500,9 +1497,8 @@ def __sub__(self, other):
         elif is_integer_dtype(other_dtype):
             if not is_period_dtype(self.dtype):
                 raise integer_op_not_supported(self)
-            result = cast("PeriodArray", self)._addsub_int_array_or_scalar(
-                other * self.freq.n, operator.sub
-            )
+            obj = cast("PeriodArray", self)
+            result = obj._addsub_int_array_or_scalar(other * obj.freq.n, operator.sub)
         else:
             # Includes ExtensionArrays, float_dtype
             return NotImplemented

pandas/core/arrays/interval.py

@@ -812,6 +812,8 @@ def argsort(
         ascending = nv.validate_argsort_with_ascending(ascending, (), kwargs)
 
         if ascending and kind == "quicksort" and na_position == "last":
+            # TODO: in an IntervalIndex we can re-use the cached
+            #  IntervalTree.left_sorter
             return np.lexsort((self.right, self.left))
 
         # TODO: other cases we can use lexsort for?  much more performant.

pandas/core/groupby/groupby.py

@@ -32,7 +32,6 @@ class providing the base-class of operations.
     cast,
     final,
 )
-import warnings
 
 import numpy as np
 
@@ -2200,13 +2199,8 @@ def sem(self, ddof: int = 1, numeric_only: bool = False):
             counts = self.count()
             result_ilocs = result.columns.get_indexer_for(cols)
             count_ilocs = counts.columns.get_indexer_for(cols)
-            with warnings.catch_warnings():
-                # TODO(2.0): once iloc[:, foo] = bar depecation is enforced,
-                #  this catching will be unnecessary
-                warnings.filterwarnings(
-                    "ignore", ".*will attempt to set the values inplace.*"
-                )
-                result.iloc[:, result_ilocs] /= np.sqrt(counts.iloc[:, count_ilocs])
+
+            result.iloc[:, result_ilocs] /= np.sqrt(counts.iloc[:, count_ilocs])
         return result
 
     @final

pandas/core/internals/array_manager.py

@@ -863,8 +863,6 @@ def column_setitem(
 
         This is a method on the ArrayManager level, to avoid creating an
         intermediate Series at the DataFrame level (`s = df[loc]; s[idx] = value`)
-
-
         """
         if not is_integer(loc):
             raise TypeError("The column index should be an integer")

pandas/core/internals/blocks.py

@@ -325,6 +325,7 @@ def apply(self, func, **kwargs) -> list[Block]:
 
         return self._split_op_result(result)
 
+    @final
     def reduce(self, func) -> list[Block]:
         # We will apply the function and reshape the result into a single-row
         #  Block with the same mgr_locs; squeezing will be done at a higher level
@@ -1957,19 +1958,6 @@ class ObjectBlock(NumpyBlock):
     __slots__ = ()
     is_object = True
 
-    def reduce(self, func) -> list[Block]:
-        """
-        For object-dtype, we operate column-wise.
-        """
-        assert self.ndim == 2
-
-        res = func(self.values)
-
-        assert isinstance(res, np.ndarray)
-        assert res.ndim == 1
-        res = res.reshape(-1, 1)
-        return [self.make_block_same_class(res)]
-
     @maybe_split
     def convert(
         self,
@@ -1980,7 +1968,9 @@ def convert(
         attempt to cast any object types to better types return a copy of
         the block (if copy = True) by definition we ARE an ObjectBlock!!!!!
         """
-        if self.dtype != object:
+        if self.dtype != _dtype_obj:
+            # GH#50067 this should be impossible in ObjectBlock, but until
+            #  that is fixed, we short-circuit here.
             return [self]
 
         values = self.values

pandas/core/series.py

@@ -856,7 +856,6 @@ def __array__(self, dtype: npt.DTypeLike | None = None) -> np.ndarray:
 
     # coercion
     __float__ = _coerce_method(float)
-    __long__ = _coerce_method(int)
     __int__ = _coerce_method(int)
 
     # ----------------------------------------------------------------------

pandas/core/tools/datetimes.py

@@ -398,10 +398,7 @@ def _convert_listlike_datetimes(
 
     elif is_datetime64_ns_dtype(arg_dtype):
         if not isinstance(arg, (DatetimeArray, DatetimeIndex)):
-            try:
-                return DatetimeIndex(arg, tz=tz, name=name)
-            except ValueError:
-                pass
+            return DatetimeIndex(arg, tz=tz, name=name)
         elif utc:
             # DatetimeArray, DatetimeIndex
             return arg.tz_localize("utc")

pandas/tests/arithmetic/test_timedelta64.py

@@ -1751,7 +1751,7 @@ def test_td64arr_floordiv_td64arr_with_nat(
         expected = np.array([1.0, 1.0, np.nan], dtype=np.float64)
         expected = tm.box_expected(expected, xbox)
         if box is DataFrame and using_array_manager:
-            # INFO(ArrayManager) floorfiv returns integer, and ArrayManager
+            # INFO(ArrayManager) floordiv returns integer, and ArrayManager
             # performs ops column-wise and thus preserves int64 dtype for
             # columns without missing values
             expected[[0, 1]] = expected[[0, 1]].astype("int64")

pandas/tests/indexes/test_base.py

@@ -891,7 +891,6 @@ def test_isin_nan_common_float64(self, nulls_fixture):
         "index",
         [
             Index(["qux", "baz", "foo", "bar"]),
-            # float64 Index overrides isin, so must be checked separately
             NumericIndex([1.0, 2.0, 3.0, 4.0], dtype=np.float64),
         ],
     )

pandas/tests/reshape/concat/test_concat.py

@@ -745,7 +745,7 @@ def test_concat_retain_attrs(data):
 @td.skip_array_manager_invalid_test
 @pytest.mark.parametrize("df_dtype", ["float64", "int64", "datetime64[ns]"])
 @pytest.mark.parametrize("empty_dtype", [None, "float64", "object"])
-def test_concat_ignore_emtpy_object_float(empty_dtype, df_dtype):
+def test_concat_ignore_empty_object_float(empty_dtype, df_dtype):
     # https://github.com/pandas-dev/pandas/issues/45637
     df = DataFrame({"foo": [1, 2], "bar": [1, 2]}, dtype=df_dtype)
     empty = DataFrame(columns=["foo", "bar"], dtype=empty_dtype)