From 6c835114ee3d923b727aca15d37046fc8da6e654 Mon Sep 17 00:00:00 2001
From: Irv Lustig <irv@princeton.com>
Date: Sat, 1 May 2021 19:59:39 -0400
Subject: [PATCH 01/12] TYP: ExtensionArray.__getitem__

---
 pandas/core/arrays/_mixins.py      | 15 +++++++++++++--
 pandas/core/arrays/base.py         | 10 +++++++++-
 pandas/core/arrays/datetimelike.py | 20 +++++++++++++-------
 pandas/core/indexes/base.py        |  6 +-----
 pandas/core/missing.py             |  6 +-----
 5 files changed, 37 insertions(+), 20 deletions(-)

diff --git a/pandas/core/arrays/_mixins.py b/pandas/core/arrays/_mixins.py
index e97687de34273..976e0e73bbbfd 100644
--- a/pandas/core/arrays/_mixins.py
+++ b/pandas/core/arrays/_mixins.py
@@ -6,6 +6,7 @@
     Sequence,
     TypeVar,
     cast,
+    overload,
 )
 
 import numpy as np
@@ -13,7 +14,7 @@
 from pandas._libs import lib
 from pandas._typing import (
     F,
-    PositionalIndexer2D,
+    PositionalIndexer,
     Shape,
     type_t,
 )
@@ -276,9 +277,19 @@ def __setitem__(self, key, value):
     def _validate_setitem_value(self, value):
         return value
 
+    @overload
+    def __getitem__(self, item: int | np.integer) -> Any:
+        ...
+
+    @overload
+    def __getitem__(
+        self: NDArrayBackedExtensionArrayT, item: slice | np.ndarray | Sequence[int]
+    ) -> NDArrayBackedExtensionArrayT:
+        ...
+
     def __getitem__(
         self: NDArrayBackedExtensionArrayT,
-        key: PositionalIndexer2D,
+        key: PositionalIndexer,
     ) -> NDArrayBackedExtensionArrayT | Any:
         if lib.is_integer(key):
             # fast-path
diff --git a/pandas/core/arrays/base.py b/pandas/core/arrays/base.py
index 5a2643dd531ed..94ce7cfccd02a 100644
--- a/pandas/core/arrays/base.py
+++ b/pandas/core/arrays/base.py
@@ -16,6 +16,7 @@
     Sequence,
     TypeVar,
     cast,
+    overload,
 )
 
 import numpy as np
@@ -288,6 +289,13 @@ def _from_factorized(cls, values, original):
     # ------------------------------------------------------------------------
     # Must be a Sequence
     # ------------------------------------------------------------------------
+    @overload
+    def __getitem__(self, item: int | np.integer) -> Any:
+        ...
+
+    @overload
+    def __getitem__(self, item: slice | np.ndarray | Sequence[int]) -> ExtensionArray:
+        ...
 
     def __getitem__(self, item: PositionalIndexer) -> ExtensionArray | Any:
         """
@@ -729,7 +737,7 @@ def fillna(self, value=None, method=None, limit=None):
             new_values = self.copy()
         return new_values
 
-    def dropna(self):
+    def dropna(self: ExtensionArrayT) -> ExtensionArrayT:
         """
         Return ExtensionArray without NA values.
 
diff --git a/pandas/core/arrays/datetimelike.py b/pandas/core/arrays/datetimelike.py
index 93df88aba2cba..814e96d5d2c49 100644
--- a/pandas/core/arrays/datetimelike.py
+++ b/pandas/core/arrays/datetimelike.py
@@ -48,7 +48,7 @@
     Dtype,
     DtypeObj,
     NpDtype,
-    PositionalIndexer2D,
+    PositionalIndexer,
 )
 from pandas.compat.numpy import function as nv
 from pandas.errors import (
@@ -309,8 +309,18 @@ def __array__(self, dtype: NpDtype | None = None) -> np.ndarray:
             return np.array(list(self), dtype=object)
         return self._ndarray
 
+    @overload
+    def __getitem__(self, item: int | np.integer) -> DTScalarOrNaT:
+        ...
+
+    @overload
+    def __getitem__(
+        self, item: slice | np.ndarray | Sequence[int]
+    ) -> DatetimeLikeArrayMixin:
+        ...
+
     def __getitem__(
-        self, key: PositionalIndexer2D
+        self, key: PositionalIndexer
     ) -> DatetimeLikeArrayMixin | DTScalarOrNaT:
         """
         This getitem defers to the underlying array, which by-definition can
@@ -1778,11 +1788,7 @@ def factorize(self, na_sentinel=-1, sort: bool = False):
             uniques = self.copy()  # TODO: copy or view?
             if sort and self.freq.n < 0:
                 codes = codes[::-1]
-                # TODO: overload __getitem__, a slice indexer returns same type as self
-                # error: Incompatible types in assignment (expression has type
-                # "Union[DatetimeLikeArrayMixin, Union[Any, Any]]", variable
-                # has type "TimelikeOps")
-                uniques = uniques[::-1]  # type: ignore[assignment]
+                uniques = uniques[::-1]
             return codes, uniques
         # FIXME: shouldn't get here; we are ignoring sort
         return super().factorize(na_sentinel=na_sentinel)
diff --git a/pandas/core/indexes/base.py b/pandas/core/indexes/base.py
index 1bae9947bd875..55168bd8d90b8 100644
--- a/pandas/core/indexes/base.py
+++ b/pandas/core/indexes/base.py
@@ -4595,11 +4595,7 @@ def __getitem__(self, key):
 
         result = getitem(key)
         if not is_scalar(result):
-            # error: Argument 1 to "ndim" has incompatible type "Union[ExtensionArray,
-            # Any]"; expected "Union[Union[int, float, complex, str, bytes, generic],
-            # Sequence[Union[int, float, complex, str, bytes, generic]],
-            # Sequence[Sequence[Any]], _SupportsArray]"
-            if np.ndim(result) > 1:  # type: ignore[arg-type]
+            if np.ndim(result) > 1:
                 deprecate_ndim_indexing(result)
                 return result
             # NB: Using _constructor._simple_new would break if MultiIndex
diff --git a/pandas/core/missing.py b/pandas/core/missing.py
index 8849eb0670faa..d6c0dcce63603 100644
--- a/pandas/core/missing.py
+++ b/pandas/core/missing.py
@@ -421,11 +421,7 @@ def interpolate_1d(
     if method in NP_METHODS:
         # np.interp requires sorted X values, #21037
 
-        # error: Argument 1 to "argsort" has incompatible type "Union[ExtensionArray,
-        # Any]"; expected "Union[Union[int, float, complex, str, bytes, generic],
-        # Sequence[Union[int, float, complex, str, bytes, generic]],
-        # Sequence[Sequence[Any]], _SupportsArray]"
-        indexer = np.argsort(inds[valid])  # type: ignore[arg-type]
+        indexer = np.argsort(inds[valid])
         result[invalid] = np.interp(
             inds[invalid], inds[valid][indexer], yvalues[valid][indexer]
         )

From 0c8d648821e7c60f8306e8ae7de97aebab122d55 Mon Sep 17 00:00:00 2001
From: Irv Lustig <irv@princeton.com>
Date: Sun, 2 May 2021 14:18:42 -0400
Subject: [PATCH 02/12] make base class use PositionalIndexer2D

---
 pandas/core/arrays/_mixins.py      | 12 ++++++------
 pandas/core/arrays/base.py         |  4 ++--
 pandas/core/arrays/datetimelike.py | 10 +++++++---
 3 files changed, 15 insertions(+), 11 deletions(-)

diff --git a/pandas/core/arrays/_mixins.py b/pandas/core/arrays/_mixins.py
index 976e0e73bbbfd..c6d8e4106dd1b 100644
--- a/pandas/core/arrays/_mixins.py
+++ b/pandas/core/arrays/_mixins.py
@@ -14,7 +14,7 @@
 from pandas._libs import lib
 from pandas._typing import (
     F,
-    PositionalIndexer,
+    PositionalIndexer2D,
     Shape,
     type_t,
 )
@@ -283,14 +283,14 @@ def __getitem__(self, item: int | np.integer) -> Any:
 
     @overload
     def __getitem__(
-        self: NDArrayBackedExtensionArrayT, item: slice | np.ndarray | Sequence[int]
-    ) -> NDArrayBackedExtensionArrayT:
+        self: NDArrayBackedExtensionArray, item: slice | np.ndarray | Sequence[int]
+    ) -> NDArrayBackedExtensionArray:
         ...
 
     def __getitem__(
-        self: NDArrayBackedExtensionArrayT,
-        key: PositionalIndexer,
-    ) -> NDArrayBackedExtensionArrayT | Any:
+        self: NDArrayBackedExtensionArray,
+        key: PositionalIndexer2D,
+    ) -> NDArrayBackedExtensionArray | Any:
         if lib.is_integer(key):
             # fast-path
             result = self._ndarray[key]
diff --git a/pandas/core/arrays/base.py b/pandas/core/arrays/base.py
index 94ce7cfccd02a..a83f9cd104326 100644
--- a/pandas/core/arrays/base.py
+++ b/pandas/core/arrays/base.py
@@ -25,7 +25,7 @@
 from pandas._typing import (
     ArrayLike,
     Dtype,
-    PositionalIndexer,
+    PositionalIndexer2D,
     Shape,
 )
 from pandas.compat import set_function_name
@@ -297,7 +297,7 @@ def __getitem__(self, item: int | np.integer) -> Any:
     def __getitem__(self, item: slice | np.ndarray | Sequence[int]) -> ExtensionArray:
         ...
 
-    def __getitem__(self, item: PositionalIndexer) -> ExtensionArray | Any:
+    def __getitem__(self, item: PositionalIndexer2D) -> ExtensionArray | Any:
         """
         Select a subset of self.
 
diff --git a/pandas/core/arrays/datetimelike.py b/pandas/core/arrays/datetimelike.py
index 814e96d5d2c49..da45895f3627b 100644
--- a/pandas/core/arrays/datetimelike.py
+++ b/pandas/core/arrays/datetimelike.py
@@ -48,7 +48,7 @@
     Dtype,
     DtypeObj,
     NpDtype,
-    PositionalIndexer,
+    PositionalIndexer2D,
 )
 from pandas.compat.numpy import function as nv
 from pandas.errors import (
@@ -320,13 +320,17 @@ def __getitem__(
         ...
 
     def __getitem__(
-        self, key: PositionalIndexer
+        self, key: PositionalIndexer2D
     ) -> DatetimeLikeArrayMixin | DTScalarOrNaT:
         """
         This getitem defers to the underlying array, which by-definition can
         only handle list-likes, slices, and integer scalars
         """
-        result = super().__getitem__(key)
+        #  error: Invalid index type "Union[Union[int, integer[Any], slice,
+        # Sequence[int], ndarray], Tuple[Union[int, integer[Any], slice, Sequence[int],
+        # ndarray], Union[int, integer[Any], slice, Sequence[int], ndarray]]]" for
+        # "DatetimeLikeArrayMixin"; expected type "Union[int, integer[Any]]"
+        result = super().__getitem__(key)  # type: ignore[index]
         if lib.is_scalar(result):
             return result
 

From 94ced76e082879454b60c7097de6507c27b5453c Mon Sep 17 00:00:00 2001
From: Irv Lustig <irv@princeton.com>
Date: Sun, 16 May 2021 15:08:19 -0400
Subject: [PATCH 03/12] fix up getitem typing for DateTimeOps

---
 pandas/core/arrays/datetimelike.py | 8 ++++----
 pandas/core/arrays/datetimes.py    | 7 ++-----
 2 files changed, 6 insertions(+), 9 deletions(-)

diff --git a/pandas/core/arrays/datetimelike.py b/pandas/core/arrays/datetimelike.py
index de86455d248b9..e7f85a7cf0902 100644
--- a/pandas/core/arrays/datetimelike.py
+++ b/pandas/core/arrays/datetimelike.py
@@ -314,13 +314,13 @@ def __getitem__(self, item: int | np.integer) -> DTScalarOrNaT:
 
     @overload
     def __getitem__(
-        self, item: slice | np.ndarray | Sequence[int]
-    ) -> DatetimeLikeArrayMixin:
+        self: DatetimeLikeArrayT, item: slice | np.ndarray | Sequence[int]
+    ) -> DatetimeLikeArrayT:
         ...
 
     def __getitem__(
-        self, key: PositionalIndexer2D
-    ) -> DatetimeLikeArrayMixin | DTScalarOrNaT:
+        self: DatetimeLikeArrayT, key: PositionalIndexer2D
+    ) -> DatetimeLikeArrayT | DTScalarOrNaT:
         """
         This getitem defers to the underlying array, which by-definition can
         only handle list-likes, slices, and integer scalars
diff --git a/pandas/core/arrays/datetimes.py b/pandas/core/arrays/datetimes.py
index f07a04b8087e0..e9e2f0e24a817 100644
--- a/pandas/core/arrays/datetimes.py
+++ b/pandas/core/arrays/datetimes.py
@@ -8,7 +8,6 @@
 )
 from typing import (
     TYPE_CHECKING,
-    cast,
     overload,
 )
 import warnings
@@ -476,11 +475,9 @@ def _generate_range(
             index = cls._simple_new(arr, freq=None, dtype=dtype)
 
         if not left_closed and len(index) and index[0] == start:
-            # TODO: overload DatetimeLikeArrayMixin.__getitem__
-            index = cast(DatetimeArray, index[1:])
+            index = index[1:]
         if not right_closed and len(index) and index[-1] == end:
-            # TODO: overload DatetimeLikeArrayMixin.__getitem__
-            index = cast(DatetimeArray, index[:-1])
+            index = index[:-1]
 
         dtype = tz_to_dtype(tz)
         return cls._simple_new(index._ndarray, freq=freq, dtype=dtype)

From 2773c8b39abca4156312b4c74b28cbe2fff6f9e2 Mon Sep 17 00:00:00 2001
From: Irv Lustig <irv@princeton.com>
Date: Sun, 23 May 2021 14:33:30 -0400
Subject: [PATCH 04/12] Make getitem on EA accept 1D, and change declaration
 for 2D arrays

---
 pandas/_typing.py                      |  7 +++---
 pandas/core/arrays/_mixins.py          | 10 +++++---
 pandas/core/arrays/base.py             | 12 ++++++---
 pandas/core/arrays/categorical.py      | 18 ++++++++++++-
 pandas/core/arrays/datetimelike.py     | 11 ++++----
 pandas/core/arrays/interval.py         | 16 +++++++++++-
 pandas/core/arrays/masked.py           | 15 ++++++++++-
 pandas/core/arrays/sparse/array.py     | 35 ++++++++++++++++++++++++--
 pandas/core/arrays/string_arrow.py     | 15 ++++++++++-
 pandas/core/indexes/base.py            | 12 ++++++---
 pandas/core/internals/array_manager.py | 21 +++++++---------
 pandas/core/internals/blocks.py        |  6 ++---
 pandas/core/internals/concat.py        |  6 ++---
 pandas/core/internals/ops.py           | 24 +++++++++---------
 14 files changed, 152 insertions(+), 56 deletions(-)

diff --git a/pandas/_typing.py b/pandas/_typing.py
index 7763b0ceb610a..2698ed696191b 100644
--- a/pandas/_typing.py
+++ b/pandas/_typing.py
@@ -211,7 +211,6 @@
 # TODO: add Ellipsis, see
 # https://github.com/python/typing/issues/684#issuecomment-548203158
 # https://bugs.python.org/issue41810
-PositionalIndexer = Union[int, np.integer, slice, Sequence[int], np.ndarray]
-PositionalIndexer2D = Union[
-    PositionalIndexer, Tuple[PositionalIndexer, PositionalIndexer]
-]
+PositionalIndexer = Union[int, np.integer, slice, List[int], np.ndarray]
+PositionalIndexerTuple = Tuple[PositionalIndexer, PositionalIndexer]
+PositionalIndexer2D = Union[PositionalIndexer, PositionalIndexerTuple]
diff --git a/pandas/core/arrays/_mixins.py b/pandas/core/arrays/_mixins.py
index db99fdacf3ccc..8ae32445be8b3 100644
--- a/pandas/core/arrays/_mixins.py
+++ b/pandas/core/arrays/_mixins.py
@@ -16,6 +16,7 @@
 from pandas._typing import (
     F,
     PositionalIndexer2D,
+    PositionalIndexerTuple,
     Shape,
     type_t,
 )
@@ -206,19 +207,20 @@ def _validate_setitem_value(self, value):
         return value
 
     @overload
-    def __getitem__(self, item: int | np.integer) -> Any:
+    def __getitem__(self, key: int | np.integer) -> Any:
         ...
 
     @overload
     def __getitem__(
-        self: NDArrayBackedExtensionArray, item: slice | np.ndarray | Sequence[int]
+        self: NDArrayBackedExtensionArray,
+        key: slice | np.ndarray | list[int] | PositionalIndexerTuple,
     ) -> NDArrayBackedExtensionArray:
         ...
 
     def __getitem__(
-        self: NDArrayBackedExtensionArray,
+        self: NDArrayBackedExtensionArrayT,
         key: PositionalIndexer2D,
-    ) -> NDArrayBackedExtensionArray | Any:
+    ) -> NDArrayBackedExtensionArrayT | Any:
         if lib.is_integer(key):
             # fast-path
             result = self._ndarray[key]
diff --git a/pandas/core/arrays/base.py b/pandas/core/arrays/base.py
index 666cc073cd3b5..c64ecff3a74f8 100644
--- a/pandas/core/arrays/base.py
+++ b/pandas/core/arrays/base.py
@@ -27,7 +27,7 @@
     ArrayLike,
     Dtype,
     FillnaOptions,
-    PositionalIndexer2D,
+    PositionalIndexer,
     Shape,
 )
 from pandas.compat import set_function_name
@@ -297,10 +297,14 @@ def __getitem__(self, item: int | np.integer) -> Any:
         ...
 
     @overload
-    def __getitem__(self, item: slice | np.ndarray | Sequence[int]) -> ExtensionArray:
+    def __getitem__(
+        self: ExtensionArrayT, item: slice | np.ndarray | list[int]
+    ) -> ExtensionArrayT:
         ...
 
-    def __getitem__(self, item: PositionalIndexer2D) -> ExtensionArray | Any:
+    def __getitem__(
+        self: ExtensionArrayT, item: PositionalIndexer
+    ) -> ExtensionArrayT | Any:
         """
         Select a subset of self.
 
@@ -314,6 +318,8 @@ def __getitem__(self, item: PositionalIndexer2D) -> ExtensionArray | Any:
 
             * ndarray: A 1-d boolean NumPy ndarray the same length as 'self'
 
+            * list[int]:  A list of int
+
         Returns
         -------
         item : scalar or ExtensionArray
diff --git a/pandas/core/arrays/categorical.py b/pandas/core/arrays/categorical.py
index cb8a08f5668ac..c85ff07ee1617 100644
--- a/pandas/core/arrays/categorical.py
+++ b/pandas/core/arrays/categorical.py
@@ -11,6 +11,7 @@
     TypeVar,
     Union,
     cast,
+    overload,
 )
 from warnings import (
     catch_warnings,
@@ -34,6 +35,8 @@
     Dtype,
     NpDtype,
     Ordered,
+    PositionalIndexer2D,
+    PositionalIndexerTuple,
     Scalar,
     Shape,
     type_t,
@@ -2015,7 +2018,20 @@ def __repr__(self) -> str:
 
     # ------------------------------------------------------------------
 
-    def __getitem__(self, key):
+    @overload
+    def __getitem__(self, key: int | np.integer) -> object:
+        ...
+
+    @overload
+    def __getitem__(
+        self: CategoricalT,
+        key: slice | np.ndarray | list[int] | PositionalIndexerTuple,
+    ) -> CategoricalT:
+        ...
+
+    def __getitem__(
+        self: CategoricalT, key: PositionalIndexer2D
+    ) -> CategoricalT | object:
         """
         Return an item.
         """
diff --git a/pandas/core/arrays/datetimelike.py b/pandas/core/arrays/datetimelike.py
index 4ae63068edbcb..6b371a3531ad1 100644
--- a/pandas/core/arrays/datetimelike.py
+++ b/pandas/core/arrays/datetimelike.py
@@ -48,6 +48,7 @@
     DtypeObj,
     NpDtype,
     PositionalIndexer2D,
+    PositionalIndexerTuple,
 )
 from pandas.compat.numpy import function as nv
 from pandas.errors import (
@@ -314,7 +315,8 @@ def __getitem__(self, item: int | np.integer) -> DTScalarOrNaT:
 
     @overload
     def __getitem__(
-        self: DatetimeLikeArrayT, item: slice | np.ndarray | Sequence[int]
+        self: DatetimeLikeArrayT,
+        item: slice | np.ndarray | list[int] | PositionalIndexerTuple,
     ) -> DatetimeLikeArrayT:
         ...
 
@@ -325,11 +327,8 @@ def __getitem__(
         This getitem defers to the underlying array, which by-definition can
         only handle list-likes, slices, and integer scalars
         """
-        #  error: Invalid index type "Union[Union[int, integer[Any], slice,
-        # Sequence[int], ndarray], Tuple[Union[int, integer[Any], slice, Sequence[int],
-        # ndarray], Union[int, integer[Any], slice, Sequence[int], ndarray]]]" for
-        # "DatetimeLikeArrayMixin"; expected type "Union[int, integer[Any]]"
-        result = super().__getitem__(key)  # type: ignore[index]
+        # Use cast as we know we will get back a DatetimeLikeArray
+        result = cast(DatetimeLikeArrayT, super().__getitem__(key))
         if lib.is_scalar(result):
             return result
 
diff --git a/pandas/core/arrays/interval.py b/pandas/core/arrays/interval.py
index a99bf245a6073..2830e90ceaad7 100644
--- a/pandas/core/arrays/interval.py
+++ b/pandas/core/arrays/interval.py
@@ -10,6 +10,7 @@
     Sequence,
     TypeVar,
     cast,
+    overload,
 )
 
 import numpy as np
@@ -28,6 +29,7 @@
     ArrayLike,
     Dtype,
     NpDtype,
+    PositionalIndexer,
 )
 from pandas.compat.numpy import function as nv
 from pandas.util._decorators import Appender
@@ -629,7 +631,19 @@ def __iter__(self):
     def __len__(self) -> int:
         return len(self._left)
 
-    def __getitem__(self, key):
+    @overload
+    def __getitem__(self, key: int | np.integer) -> Interval:
+        ...
+
+    @overload
+    def __getitem__(
+        self: IntervalArrayT, key: slice | np.ndarray | list[int]
+    ) -> IntervalArrayT:
+        ...
+
+    def __getitem__(
+        self: IntervalArrayT, key: PositionalIndexer
+    ) -> IntervalArrayT | Interval:
         key = check_array_indexer(self, key)
         left = self._left[key]
         right = self._right[key]
diff --git a/pandas/core/arrays/masked.py b/pandas/core/arrays/masked.py
index 11f9f645920ec..9890543b9c05d 100644
--- a/pandas/core/arrays/masked.py
+++ b/pandas/core/arrays/masked.py
@@ -5,6 +5,7 @@
     Any,
     Sequence,
     TypeVar,
+    overload,
 )
 
 import numpy as np
@@ -135,7 +136,19 @@ def __init__(self, values: np.ndarray, mask: np.ndarray, copy: bool = False):
     def dtype(self) -> BaseMaskedDtype:
         raise AbstractMethodError(self)
 
-    def __getitem__(self, item: PositionalIndexer) -> BaseMaskedArray | Any:
+    @overload
+    def __getitem__(self, item: int | np.integer) -> Any:
+        ...
+
+    @overload
+    def __getitem__(
+        self: BaseMaskedArrayT, item: slice | np.ndarray | list[int]
+    ) -> BaseMaskedArrayT:
+        ...
+
+    def __getitem__(
+        self: BaseMaskedArrayT, item: PositionalIndexer
+    ) -> BaseMaskedArrayT | Any:
         if is_integer(item):
             if self._mask[item]:
                 return self.dtype.na_value
diff --git a/pandas/core/arrays/sparse/array.py b/pandas/core/arrays/sparse/array.py
index 4847372f18239..fe17245fc9fb6 100644
--- a/pandas/core/arrays/sparse/array.py
+++ b/pandas/core/arrays/sparse/array.py
@@ -7,10 +7,13 @@
 import numbers
 import operator
 from typing import (
+    TYPE_CHECKING,
     Any,
     Callable,
     Sequence,
     TypeVar,
+    cast,
+    overload,
 )
 import warnings
 
@@ -27,6 +30,7 @@
 from pandas._typing import (
     Dtype,
     NpDtype,
+    PositionalIndexer,
     Scalar,
 )
 from pandas.compat.numpy import function as nv
@@ -77,6 +81,17 @@
 
 import pandas.io.formats.printing as printing
 
+# See https://github.com/python/typing/issues/684
+if TYPE_CHECKING:
+    from enum import Enum
+
+    class ellipsis(Enum):
+        Ellipsis = "..."
+
+    Ellipsis = ellipsis.Ellipsis
+else:
+    ellipsis = type(Ellipsis)
+
 # ----------------------------------------------------------------------------
 # Array
 
@@ -810,8 +825,21 @@ def value_counts(self, dropna: bool = True):
     # --------
     # Indexing
     # --------
+    @overload
+    def __getitem__(self, key: int | np.integer) -> Any:
+        ...
+
+    @overload
+    def __getitem__(
+        self: SparseArrayT,
+        key: slice | np.ndarray | list[int] | tuple[int | ellipsis, ...],
+    ) -> SparseArrayT:
+        ...
 
-    def __getitem__(self, key):
+    def __getitem__(
+        self: SparseArrayT,
+        key: PositionalIndexer | tuple[int | ellipsis, ...],
+    ) -> SparseArrayT | Any:
 
         if isinstance(key, tuple):
             if len(key) > 1:
@@ -821,6 +849,8 @@ def __getitem__(self, key):
                     key = key[:-1]
             if len(key) > 1:
                 raise IndexError("too many indices for array.")
+            if key[0] is Ellipsis:
+                raise ValueError("Cannot slice with Ellipsis")
             key = key[0]
 
         if is_integer(key):
@@ -849,7 +879,8 @@ def __getitem__(self, key):
             key = check_array_indexer(self, key)
 
             if com.is_bool_indexer(key):
-
+                # mypy doesn't know we have an array here
+                key = cast(np.ndarray, key)
                 return self.take(np.arange(len(key), dtype=np.int32)[key])
             elif hasattr(key, "__len__"):
                 return self.take(key)
diff --git a/pandas/core/arrays/string_arrow.py b/pandas/core/arrays/string_arrow.py
index d5ee28eb7017e..86e177cbd6ab1 100644
--- a/pandas/core/arrays/string_arrow.py
+++ b/pandas/core/arrays/string_arrow.py
@@ -6,6 +6,7 @@
     Any,
     Sequence,
     cast,
+    overload,
 )
 
 import numpy as np
@@ -341,7 +342,19 @@ def _concat_same_type(cls, to_concat) -> ArrowStringArray:
             )
         )
 
-    def __getitem__(self, item: PositionalIndexer) -> Any:
+    @overload
+    def __getitem__(self, item: int | np.integer) -> str:
+        ...
+
+    @overload
+    def __getitem__(
+        self: ArrowStringArray, item: slice | np.ndarray | list[int]
+    ) -> ArrowStringArray:
+        ...
+
+    def __getitem__(
+        self: ArrowStringArray, item: PositionalIndexer
+    ) -> ArrowStringArray | str:
         """Select a subset of self.
 
         Parameters
diff --git a/pandas/core/indexes/base.py b/pandas/core/indexes/base.py
index 9f18721ee575a..b263d47510b80 100644
--- a/pandas/core/indexes/base.py
+++ b/pandas/core/indexes/base.py
@@ -4915,13 +4915,19 @@ def asof_locs(self, where: Index, mask: np.ndarray) -> np.ndarray:
             which correspond to the return values of the `asof` function
             for every element in `where`.
         """
-        locs = self._values[mask].searchsorted(where._values, side="right")
+        # error: Argument 1 to "searchsorted" of "ndarray" has incompatible type
+        # "Union[ExtensionArray, ndarray]"; expected "Union[Union[int, float, complex,
+        # str, bytes, generic], Sequence[Union[int, float, complex, str, bytes,
+        # generic]], Sequence[Sequence[Any]], _SupportsArray]"
+        # TODO: will be fixed when ExtensionArray.searchsorted() is fixed
+        locs = self._values[mask].searchsorted(
+            where._values, side="right"  # type: ignore[arg-type]
+        )
         locs = np.where(locs > 0, locs - 1, 0)
 
         result = np.arange(len(self), dtype=np.intp)[mask].take(locs)
 
-        # TODO: overload return type of ExtensionArray.__getitem__
-        first_value = cast(Any, self._values[mask.argmax()])
+        first_value = self._values[mask.argmax()]
         result[(locs == 0) & (where._values < first_value)] = -1
 
         return result
diff --git a/pandas/core/internals/array_manager.py b/pandas/core/internals/array_manager.py
index 31e32b053367b..d984531872aa1 100644
--- a/pandas/core/internals/array_manager.py
+++ b/pandas/core/internals/array_manager.py
@@ -324,10 +324,9 @@ def apply_with_block(self: T, f, align_keys=None, swap_axis=True, **kwargs) -> T
             if self.ndim == 2 and arr.ndim == 2:
                 # 2D for np.ndarray or DatetimeArray/TimedeltaArray
                 assert len(arr) == 1
-                # error: Invalid index type "Tuple[int, slice]" for
-                # "Union[ndarray, ExtensionArray]"; expected type
-                # "Union[int, slice, ndarray]"
-                arr = arr[0, :]  # type: ignore[index]
+                # error: No overload variant of "__getitem__" of "ExtensionArray"
+                # matches argument type "Tuple[int, slice]"
+                arr = arr[0, :]  # type: ignore[call-overload]
             result_arrays.append(arr)
 
         return type(self)(result_arrays, self._axes)
@@ -843,10 +842,9 @@ def iset(self, loc: int | slice | np.ndarray, value: ArrayLike):
         assert value.shape[0] == len(self._axes[0])
 
         for value_idx, mgr_idx in enumerate(indices):
-            # error: Invalid index type "Tuple[slice, int]" for
-            # "Union[ExtensionArray, ndarray]"; expected type
-            # "Union[int, slice, ndarray]"
-            value_arr = value[:, value_idx]  # type: ignore[index]
+            # error: No overload variant of "__getitem__" of "ExtensionArray" matches
+            # argument type "Tuple[slice, int]"
+            value_arr = value[:, value_idx]  # type: ignore[call-overload]
             self.arrays[mgr_idx] = value_arr
         return
 
@@ -866,10 +864,9 @@ def insert(self, loc: int, item: Hashable, value: ArrayLike) -> None:
         value = extract_array(value, extract_numpy=True)
         if value.ndim == 2:
             if value.shape[0] == 1:
-                # error: Invalid index type "Tuple[int, slice]" for
-                # "Union[Any, ExtensionArray, ndarray]"; expected type
-                # "Union[int, slice, ndarray]"
-                value = value[0, :]  # type: ignore[index]
+                # error: No overload variant of "__getitem__" of "ExtensionArray"
+                # matches argument type "Tuple[int, slice]"
+                value = value[0, :]  # type: ignore[call-overload]
             else:
                 raise ValueError(
                     f"Expected a 1D array, got an array with shape {value.shape}"
diff --git a/pandas/core/internals/blocks.py b/pandas/core/internals/blocks.py
index 4f1b16e747394..4e1a3f19ad56d 100644
--- a/pandas/core/internals/blocks.py
+++ b/pandas/core/internals/blocks.py
@@ -316,9 +316,9 @@ def getitem_block_index(self, slicer: slice) -> Block:
 
         Assumes self.ndim == 2
         """
-        # error: Invalid index type "Tuple[ellipsis, slice]" for
-        # "Union[ndarray, ExtensionArray]"; expected type "Union[int, slice, ndarray]"
-        new_values = self.values[..., slicer]  # type: ignore[index]
+        # error: No overload variant of "__getitem__" of "ExtensionArray" matches
+        # argument type "Tuple[int, slice]"
+        new_values = self.values[..., slicer]  # type: ignore[call-overload]
         return type(self)(new_values, self._mgr_locs, ndim=self.ndim)
 
     @final
diff --git a/pandas/core/internals/concat.py b/pandas/core/internals/concat.py
index 9642b30ab91ca..842184e28ec8c 100644
--- a/pandas/core/internals/concat.py
+++ b/pandas/core/internals/concat.py
@@ -506,10 +506,10 @@ def _concatenate_join_units(
         # concatting with at least one EA means we are concatting a single column
         # the non-EA values are 2D arrays with shape (1, n)
 
-        # error: Invalid index type "Tuple[int, slice]" for
-        # "Union[ExtensionArray, ndarray]"; expected type "Union[int, slice, ndarray]"
+        # error: No overload variant of "__getitem__" of "ExtensionArray" matches
+        # argument type "Tuple[int, slice]"
         to_concat = [
-            t if is_1d_only_ea_obj(t) else t[0, :]  # type: ignore[index]
+            t if is_1d_only_ea_obj(t) else t[0, :]  # type: ignore[call-overload]
             for t in to_concat
         ]
         concat_values = concat_compat(to_concat, axis=0, ea_compat_axis=True)
diff --git a/pandas/core/internals/ops.py b/pandas/core/internals/ops.py
index 5f03d6709dfa4..35caeea9b9067 100644
--- a/pandas/core/internals/ops.py
+++ b/pandas/core/internals/ops.py
@@ -106,28 +106,28 @@ def _get_same_shape_values(
 
     # TODO(EA2D): with 2D EAs only this first clause would be needed
     if not (left_ea or right_ea):
-        # error: Invalid index type "Tuple[Any, slice]" for "Union[ndarray,
-        # ExtensionArray]"; expected type "Union[int, slice, ndarray]"
-        lvals = lvals[rblk.mgr_locs.indexer, :]  # type: ignore[index]
+        # error: No overload variant of "__getitem__" of "ExtensionArray" matches
+        # argument type "Tuple[Union[ndarray, slice], slice]"
+        lvals = lvals[rblk.mgr_locs.indexer, :]  # type: ignore[call-overload]
         assert lvals.shape == rvals.shape, (lvals.shape, rvals.shape)
     elif left_ea and right_ea:
         assert lvals.shape == rvals.shape, (lvals.shape, rvals.shape)
     elif right_ea:
         # lvals are 2D, rvals are 1D
 
-        # error: Invalid index type "Tuple[Any, slice]" for "Union[ndarray,
-        # ExtensionArray]"; expected type "Union[int, slice, ndarray]"
-        lvals = lvals[rblk.mgr_locs.indexer, :]  # type: ignore[index]
+        # error: No overload variant of "__getitem__" of "ExtensionArray" matches
+        # argument type "Tuple[Union[ndarray, slice], slice]"
+        lvals = lvals[rblk.mgr_locs.indexer, :]  # type: ignore[call-overload]
         assert lvals.shape[0] == 1, lvals.shape
-        # error: Invalid index type "Tuple[int, slice]" for "Union[Any,
-        # ExtensionArray]"; expected type "Union[int, slice, ndarray]"
-        lvals = lvals[0, :]  # type: ignore[index]
+        # error: No overload variant of "__getitem__" of "ExtensionArray" matches
+        # argument type "Tuple[int, slice]"
+        lvals = lvals[0, :]  # type: ignore[call-overload]
     else:
         # lvals are 1D, rvals are 2D
         assert rvals.shape[0] == 1, rvals.shape
-        # error: Invalid index type "Tuple[int, slice]" for "Union[ndarray,
-        # ExtensionArray]"; expected type "Union[int, slice, ndarray]"
-        rvals = rvals[0, :]  # type: ignore[index]
+        # error: No overload variant of "__getitem__" of "ExtensionArray" matches
+        # argument type "Tuple[int, slice]"
+        rvals = rvals[0, :]  # type: ignore[call-overload]
 
     return lvals, rvals
 

From 01c0cf521a3d4da378415b0a42507ad9c958bcf8 Mon Sep 17 00:00:00 2001
From: Irv Lustig <irv@princeton.com>
Date: Mon, 31 May 2021 11:47:23 -0400
Subject: [PATCH 05/12] casting in datetimelike, allow NA in string arrow

---
 pandas/_typing.py                  |  1 +
 pandas/core/arrays/datetimelike.py | 10 +++++++---
 pandas/core/arrays/string_arrow.py |  7 +++++--
 3 files changed, 13 insertions(+), 5 deletions(-)

diff --git a/pandas/_typing.py b/pandas/_typing.py
index 2698ed696191b..904c503be9907 100644
--- a/pandas/_typing.py
+++ b/pandas/_typing.py
@@ -211,6 +211,7 @@
 # TODO: add Ellipsis, see
 # https://github.com/python/typing/issues/684#issuecomment-548203158
 # https://bugs.python.org/issue41810
+# Using List[int] here rather than Sequence[int] to disallow tuples.
 PositionalIndexer = Union[int, np.integer, slice, List[int], np.ndarray]
 PositionalIndexerTuple = Tuple[PositionalIndexer, PositionalIndexer]
 PositionalIndexer2D = Union[PositionalIndexer, PositionalIndexerTuple]
diff --git a/pandas/core/arrays/datetimelike.py b/pandas/core/arrays/datetimelike.py
index 6b371a3531ad1..e14e368ee7e1e 100644
--- a/pandas/core/arrays/datetimelike.py
+++ b/pandas/core/arrays/datetimelike.py
@@ -327,11 +327,15 @@ def __getitem__(
         This getitem defers to the underlying array, which by-definition can
         only handle list-likes, slices, and integer scalars
         """
-        # Use cast as we know we will get back a DatetimeLikeArray
-        result = cast(DatetimeLikeArrayT, super().__getitem__(key))
+        # Use cast as we know we will get back a DatetimeLikeArray or DTScalar
+        result = cast(
+            Union[DatetimeLikeArrayT, DTScalarOrNaT], super().__getitem__(key)
+        )
         if lib.is_scalar(result):
             return result
-
+        else:
+            # At this point we know the result is an array.
+            result = cast(DatetimeLikeArrayT, result)
         result._freq = self._get_getitem_freq(key)
         return result
 
diff --git a/pandas/core/arrays/string_arrow.py b/pandas/core/arrays/string_arrow.py
index a014dff23e462..076cfd08e1d04 100644
--- a/pandas/core/arrays/string_arrow.py
+++ b/pandas/core/arrays/string_arrow.py
@@ -6,6 +6,7 @@
     TYPE_CHECKING,
     Any,
     Sequence,
+    Union,
     cast,
     overload,
 )
@@ -80,6 +81,8 @@
 if TYPE_CHECKING:
     from pandas import Series
 
+ArrowStringScalarOrNAT = Union[str, libmissing.NA]
+
 
 @register_extension_dtype
 class ArrowStringDtype(StringDtype):
@@ -344,7 +347,7 @@ def _concat_same_type(cls, to_concat) -> ArrowStringArray:
         )
 
     @overload
-    def __getitem__(self, item: int | np.integer) -> str:
+    def __getitem__(self, item: int | np.integer) -> ArrowStringScalarOrNAT:
         ...
 
     @overload
@@ -355,7 +358,7 @@ def __getitem__(
 
     def __getitem__(
         self: ArrowStringArray, item: PositionalIndexer
-    ) -> ArrowStringArray | str:
+    ) -> ArrowStringArray | ArrowStringScalarOrNAT:
         """Select a subset of self.
 
         Parameters

From 3b38de218d499f6b64579bbabe018144554ad93d Mon Sep 17 00:00:00 2001
From: Irv Lustig <irv@princeton.com>
Date: Mon, 31 May 2021 13:32:12 -0400
Subject: [PATCH 06/12] fix string arrow NA type

---
 pandas/core/arrays/string_arrow.py | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/pandas/core/arrays/string_arrow.py b/pandas/core/arrays/string_arrow.py
index 076cfd08e1d04..7fcf063ca0453 100644
--- a/pandas/core/arrays/string_arrow.py
+++ b/pandas/core/arrays/string_arrow.py
@@ -81,7 +81,7 @@
 if TYPE_CHECKING:
     from pandas import Series
 
-ArrowStringScalarOrNAT = Union[str, libmissing.NA]
+ArrowStringScalarOrNAT = Union[str, libmissing.NAType]
 
 
 @register_extension_dtype

From d7c545daf0637743a76660be012f8c8f770adb48 Mon Sep 17 00:00:00 2001
From: Irv Lustig <irv@princeton.com>
Date: Sun, 13 Jun 2021 14:50:39 -0400
Subject: [PATCH 07/12] change an overload in mixins to use
 NDArrayBackedExtensionArrayT

---
 pandas/core/arrays/_mixins.py | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/pandas/core/arrays/_mixins.py b/pandas/core/arrays/_mixins.py
index 5605aacd41424..61c5dd1a08371 100644
--- a/pandas/core/arrays/_mixins.py
+++ b/pandas/core/arrays/_mixins.py
@@ -193,9 +193,9 @@ def __getitem__(self, key: int | np.integer) -> Any:
 
     @overload
     def __getitem__(
-        self: NDArrayBackedExtensionArray,
+        self: NDArrayBackedExtensionArrayT,
         key: slice | np.ndarray | list[int] | PositionalIndexerTuple,
-    ) -> NDArrayBackedExtensionArray:
+    ) -> NDArrayBackedExtensionArrayT:
         ...
 
     def __getitem__(

From 076e4342eea96545632d4f4ec439382baaf3575c Mon Sep 17 00:00:00 2001
From: Irv Lustig <irv@princeton.com>
Date: Mon, 14 Jun 2021 08:03:20 -0400
Subject: [PATCH 08/12] categorical returns Any, interval for NA, put back
 libmissing in string_arrow

---
 pandas/core/arrays/categorical.py  | 7 +++----
 pandas/core/arrays/interval.py     | 6 ++++--
 pandas/core/arrays/string_arrow.py | 5 ++++-
 3 files changed, 11 insertions(+), 7 deletions(-)

diff --git a/pandas/core/arrays/categorical.py b/pandas/core/arrays/categorical.py
index f8d6ad082afce..6b5416f1893ea 100644
--- a/pandas/core/arrays/categorical.py
+++ b/pandas/core/arrays/categorical.py
@@ -6,6 +6,7 @@
 from shutil import get_terminal_size
 from typing import (
     TYPE_CHECKING,
+    Any,
     Hashable,
     Sequence,
     TypeVar,
@@ -2005,7 +2006,7 @@ def __repr__(self) -> str:
     # ------------------------------------------------------------------
 
     @overload
-    def __getitem__(self, key: int | np.integer) -> object:
+    def __getitem__(self, key: int | np.integer) -> Any:
         ...
 
     @overload
@@ -2015,9 +2016,7 @@ def __getitem__(
     ) -> CategoricalT:
         ...
 
-    def __getitem__(
-        self: CategoricalT, key: PositionalIndexer2D
-    ) -> CategoricalT | object:
+    def __getitem__(self: CategoricalT, key: PositionalIndexer2D) -> CategoricalT | Any:
         """
         Return an item.
         """
diff --git a/pandas/core/arrays/interval.py b/pandas/core/arrays/interval.py
index 1a3e601d17496..273dc1520682e 100644
--- a/pandas/core/arrays/interval.py
+++ b/pandas/core/arrays/interval.py
@@ -9,6 +9,7 @@
 from typing import (
     Sequence,
     TypeVar,
+    Union,
     cast,
     overload,
 )
@@ -87,6 +88,7 @@
 )
 
 IntervalArrayT = TypeVar("IntervalArrayT", bound="IntervalArray")
+IntervalOrNA = Union[Interval, float]
 
 _interval_shared_docs: dict[str, str] = {}
 
@@ -631,7 +633,7 @@ def __len__(self) -> int:
         return len(self._left)
 
     @overload
-    def __getitem__(self, key: int | np.integer) -> Interval:
+    def __getitem__(self, key: int | np.integer) -> IntervalOrNA:
         ...
 
     @overload
@@ -642,7 +644,7 @@ def __getitem__(
 
     def __getitem__(
         self: IntervalArrayT, key: PositionalIndexer
-    ) -> IntervalArrayT | Interval:
+    ) -> IntervalArrayT | IntervalOrNA:
         key = check_array_indexer(self, key)
         left = self._left[key]
         right = self._right[key]
diff --git a/pandas/core/arrays/string_arrow.py b/pandas/core/arrays/string_arrow.py
index 3c8d85fd0a18b..e3338abd12ade 100644
--- a/pandas/core/arrays/string_arrow.py
+++ b/pandas/core/arrays/string_arrow.py
@@ -13,7 +13,10 @@
 
 import numpy as np
 
-from pandas._libs import lib
+from pandas._libs import (
+    lib,
+    missing as libmissing,
+)
 from pandas._typing import (
     Dtype,
     NpDtype,

From 1dbb668a2a7fda370d69b0bd890ee2aeab352f21 Mon Sep 17 00:00:00 2001
From: Irv Lustig <irv@princeton.com>
Date: Thu, 8 Jul 2021 12:22:48 -0400
Subject: [PATCH 09/12] change ignore messages

---
 pandas/core/arrays/interval.py | 9 +++++----
 pandas/core/indexes/base.py    | 8 +++-----
 2 files changed, 8 insertions(+), 9 deletions(-)

diff --git a/pandas/core/arrays/interval.py b/pandas/core/arrays/interval.py
index 80eafa43a295a..ca0d2395ef9f1 100644
--- a/pandas/core/arrays/interval.py
+++ b/pandas/core/arrays/interval.py
@@ -1643,10 +1643,11 @@ def _from_combined(self, combined: np.ndarray) -> IntervalArray:
         return self._shallow_copy(left=new_left, right=new_right)
 
     def unique(self) -> IntervalArray:
-        # Invalid index type "Tuple[slice, int]" for "Union[ExtensionArray,
-        # ndarray[Any, Any]]"; expected type "Union[int, integer[Any], slice,
-        # Sequence[int], ndarray[Any, Any]]"
-        nc = unique(self._combined.view("complex128")[:, 0])  # type: ignore[index]
+        # No overload variant of "__getitem__" of "ExtensionArray" matches argument
+        # type "Tuple[slice, int]"
+        nc = unique(
+            self._combined.view("complex128")[:, 0]  # type: ignore[call-overload]
+        )
         nc = nc[:, None]
         return self._from_combined(nc)
 
diff --git a/pandas/core/indexes/base.py b/pandas/core/indexes/base.py
index 42a94b47017eb..581c081e43230 100644
--- a/pandas/core/indexes/base.py
+++ b/pandas/core/indexes/base.py
@@ -5024,13 +5024,11 @@ def asof_locs(self, where: Index, mask: np.ndarray) -> np.ndarray:
             which correspond to the return values of the `asof` function
             for every element in `where`.
         """
-        # error: Argument 1 to "searchsorted" of "ndarray" has incompatible type
-        # "Union[ExtensionArray, ndarray]"; expected "Union[Union[int, float, complex,
-        # str, bytes, generic], Sequence[Union[int, float, complex, str, bytes,
-        # generic]], Sequence[Sequence[Any]], _SupportsArray]"
+        # error: No overload variant of "searchsorted" of "ndarray" matches argument
+        # types "Union[ExtensionArray, ndarray[Any, Any]]", "str"
         # TODO: will be fixed when ExtensionArray.searchsorted() is fixed
         locs = self._values[mask].searchsorted(
-            where._values, side="right"  # type: ignore[arg-type]
+            where._values, side="right"  # type: ignore[call-overload]
         )
         locs = np.where(locs > 0, locs - 1, 0)
 

From adf3a7373ccf0ee3ac9b6fe5cd9ac137803e0020 Mon Sep 17 00:00:00 2001
From: Irv Lustig <irv@princeton.com>
Date: Mon, 26 Jul 2021 07:54:00 -0400
Subject: [PATCH 10/12] resolve conflicts in core/internals/blocks.py

---
 pandas/core/internals/blocks.py | 17 +++--------------
 1 file changed, 3 insertions(+), 14 deletions(-)

diff --git a/pandas/core/internals/blocks.py b/pandas/core/internals/blocks.py
index dca7dd96a469d..4d7b96100d9d8 100644
--- a/pandas/core/internals/blocks.py
+++ b/pandas/core/internals/blocks.py
@@ -312,17 +312,6 @@ def getitem_block(self, slicer) -> Block:
 
         return type(self)(new_values, new_mgr_locs, self.ndim)
 
-    def getitem_block_index(self, slicer: slice) -> Block:
-        """
-        Perform __getitem__-like specialized to slicing along index.
-
-        Assumes self.ndim == 2
-        """
-        # error: No overload variant of "__getitem__" of "ExtensionArray" matches
-        # argument type "Tuple[int, slice]"
-        new_values = self.values[..., slicer]  # type: ignore[call-overload]
-        return type(self)(new_values, self._mgr_locs, ndim=self.ndim)
-
     @final
     def getitem_block_columns(self, slicer, new_mgr_locs: BlockPlacement) -> Block:
         """
@@ -1564,9 +1553,9 @@ def getitem_block_index(self, slicer: slice) -> ExtensionBlock:
 
         Assumes self.ndim == 2
         """
-        # error: Invalid index type "Tuple[ellipsis, slice]" for
-        # "Union[ndarray, ExtensionArray]"; expected type "Union[int, slice, ndarray]"
-        new_values = self.values[..., slicer]  # type: ignore[index]
+        # error: No overload variant of "__getitem__" of "ExtensionArray" matches
+        # argument type "Tuple[ellipsis, slice]"
+        new_values = self.values[..., slicer]  # type: ignore[call-overload]
         return type(self)(new_values, self._mgr_locs, ndim=self.ndim)
 
     def fillna(

From 9a8550dbb4b4acd6a0136cc2796c1bef950bab32 Mon Sep 17 00:00:00 2001
From: Irv Lustig <irv@princeton.com>
Date: Mon, 6 Sep 2021 13:05:28 -0400
Subject: [PATCH 11/12] create types for split of getitem arguments

---
 pandas/_typing.py                  |  4 +++-
 pandas/core/arrays/_mixins.py      |  6 ++++--
 pandas/core/arrays/base.py         |  8 ++++----
 pandas/core/arrays/categorical.py  |  6 ++++--
 pandas/core/arrays/datetimelike.py |  6 ++++--
 pandas/core/arrays/interval.py     |  8 ++++----
 pandas/core/arrays/masked.py       |  8 ++++----
 pandas/core/arrays/sparse/array.py |  6 ++++--
 pandas/core/arrays/string_arrow.py |  8 ++++----
 pandas/core/groupby/groupby.py     | 15 ++++++---------
 10 files changed, 41 insertions(+), 34 deletions(-)

diff --git a/pandas/_typing.py b/pandas/_typing.py
index cea420a548b89..6e3706263e634 100644
--- a/pandas/_typing.py
+++ b/pandas/_typing.py
@@ -205,6 +205,8 @@
 # https://github.com/python/typing/issues/684#issuecomment-548203158
 # https://bugs.python.org/issue41810
 # Using List[int] here rather than Sequence[int] to disallow tuples.
-PositionalIndexer = Union[int, np.integer, slice, List[int], np.ndarray]
+ScalarIndexer = Union[int, np.integer]
+SequenceIndexer = Union[slice, List[int], np.ndarray]
+PositionalIndexer = Union[ScalarIndexer, SequenceIndexer]
 PositionalIndexerTuple = Tuple[PositionalIndexer, PositionalIndexer]
 PositionalIndexer2D = Union[PositionalIndexer, PositionalIndexerTuple]
diff --git a/pandas/core/arrays/_mixins.py b/pandas/core/arrays/_mixins.py
index 61c5dd1a08371..32533a04e2b5f 100644
--- a/pandas/core/arrays/_mixins.py
+++ b/pandas/core/arrays/_mixins.py
@@ -17,6 +17,8 @@
     F,
     PositionalIndexer2D,
     PositionalIndexerTuple,
+    ScalarIndexer,
+    SequenceIndexer,
     Shape,
     type_t,
 )
@@ -188,13 +190,13 @@ def _validate_setitem_value(self, value):
         return value
 
     @overload
-    def __getitem__(self, key: int | np.integer) -> Any:
+    def __getitem__(self, key: ScalarIndexer) -> Any:
         ...
 
     @overload
     def __getitem__(
         self: NDArrayBackedExtensionArrayT,
-        key: slice | np.ndarray | list[int] | PositionalIndexerTuple,
+        key: SequenceIndexer | PositionalIndexerTuple,
     ) -> NDArrayBackedExtensionArrayT:
         ...
 
diff --git a/pandas/core/arrays/base.py b/pandas/core/arrays/base.py
index 9a03f79d8e802..dff0a1a38adf5 100644
--- a/pandas/core/arrays/base.py
+++ b/pandas/core/arrays/base.py
@@ -29,6 +29,8 @@
     Dtype,
     FillnaOptions,
     PositionalIndexer,
+    ScalarIndexer,
+    SequenceIndexer,
     Shape,
 )
 from pandas.compat import set_function_name
@@ -292,13 +294,11 @@ def _from_factorized(cls, values, original):
     # Must be a Sequence
     # ------------------------------------------------------------------------
     @overload
-    def __getitem__(self, item: int | np.integer) -> Any:
+    def __getitem__(self, item: ScalarIndexer) -> Any:
         ...
 
     @overload
-    def __getitem__(
-        self: ExtensionArrayT, item: slice | np.ndarray | list[int]
-    ) -> ExtensionArrayT:
+    def __getitem__(self: ExtensionArrayT, item: SequenceIndexer) -> ExtensionArrayT:
         ...
 
     def __getitem__(
diff --git a/pandas/core/arrays/categorical.py b/pandas/core/arrays/categorical.py
index 25ce3437ae550..a1e322a9a78bb 100644
--- a/pandas/core/arrays/categorical.py
+++ b/pandas/core/arrays/categorical.py
@@ -40,6 +40,8 @@
     PositionalIndexer2D,
     PositionalIndexerTuple,
     Scalar,
+    ScalarIndexer,
+    SequenceIndexer,
     Shape,
     npt,
     type_t,
@@ -2008,13 +2010,13 @@ def __repr__(self) -> str:
     # ------------------------------------------------------------------
 
     @overload
-    def __getitem__(self, key: int | np.integer) -> Any:
+    def __getitem__(self, key: ScalarIndexer) -> Any:
         ...
 
     @overload
     def __getitem__(
         self: CategoricalT,
-        key: slice | np.ndarray | list[int] | PositionalIndexerTuple,
+        key: SequenceIndexer | PositionalIndexerTuple,
     ) -> CategoricalT:
         ...
 
diff --git a/pandas/core/arrays/datetimelike.py b/pandas/core/arrays/datetimelike.py
index 559fd1813d66d..63ba9fdd59fc6 100644
--- a/pandas/core/arrays/datetimelike.py
+++ b/pandas/core/arrays/datetimelike.py
@@ -50,6 +50,8 @@
     NpDtype,
     PositionalIndexer2D,
     PositionalIndexerTuple,
+    ScalarIndexer,
+    SequenceIndexer,
     npt,
 )
 from pandas.compat.numpy import function as nv
@@ -315,13 +317,13 @@ def __array__(self, dtype: NpDtype | None = None) -> np.ndarray:
         return self._ndarray
 
     @overload
-    def __getitem__(self, item: int | np.integer) -> DTScalarOrNaT:
+    def __getitem__(self, item: ScalarIndexer) -> DTScalarOrNaT:
         ...
 
     @overload
     def __getitem__(
         self: DatetimeLikeArrayT,
-        item: slice | np.ndarray | list[int] | PositionalIndexerTuple,
+        item: SequenceIndexer | PositionalIndexerTuple,
     ) -> DatetimeLikeArrayT:
         ...
 
diff --git a/pandas/core/arrays/interval.py b/pandas/core/arrays/interval.py
index 185bb7788b991..732bdb112b8c3 100644
--- a/pandas/core/arrays/interval.py
+++ b/pandas/core/arrays/interval.py
@@ -34,6 +34,8 @@
     Dtype,
     NpDtype,
     PositionalIndexer,
+    ScalarIndexer,
+    SequenceIndexer,
 )
 from pandas.compat.numpy import function as nv
 from pandas.util._decorators import Appender
@@ -640,13 +642,11 @@ def __len__(self) -> int:
         return len(self._left)
 
     @overload
-    def __getitem__(self, key: int | np.integer) -> IntervalOrNA:
+    def __getitem__(self, key: ScalarIndexer) -> IntervalOrNA:
         ...
 
     @overload
-    def __getitem__(
-        self: IntervalArrayT, key: slice | np.ndarray | list[int]
-    ) -> IntervalArrayT:
+    def __getitem__(self: IntervalArrayT, key: SequenceIndexer) -> IntervalArrayT:
         ...
 
     def __getitem__(
diff --git a/pandas/core/arrays/masked.py b/pandas/core/arrays/masked.py
index 90930526ce4d2..dc2dbb85611ab 100644
--- a/pandas/core/arrays/masked.py
+++ b/pandas/core/arrays/masked.py
@@ -20,6 +20,8 @@
     NpDtype,
     PositionalIndexer,
     Scalar,
+    ScalarIndexer,
+    SequenceIndexer,
     type_t,
 )
 from pandas.errors import AbstractMethodError
@@ -139,13 +141,11 @@ def dtype(self) -> BaseMaskedDtype:
         raise AbstractMethodError(self)
 
     @overload
-    def __getitem__(self, item: int | np.integer) -> Any:
+    def __getitem__(self, item: ScalarIndexer) -> Any:
         ...
 
     @overload
-    def __getitem__(
-        self: BaseMaskedArrayT, item: slice | np.ndarray | list[int]
-    ) -> BaseMaskedArrayT:
+    def __getitem__(self: BaseMaskedArrayT, item: SequenceIndexer) -> BaseMaskedArrayT:
         ...
 
     def __getitem__(
diff --git a/pandas/core/arrays/sparse/array.py b/pandas/core/arrays/sparse/array.py
index b2e739f83bc05..36688d79ed0dd 100644
--- a/pandas/core/arrays/sparse/array.py
+++ b/pandas/core/arrays/sparse/array.py
@@ -32,6 +32,8 @@
     NpDtype,
     PositionalIndexer,
     Scalar,
+    ScalarIndexer,
+    SequenceIndexer,
 )
 from pandas.compat.numpy import function as nv
 from pandas.errors import PerformanceWarning
@@ -822,13 +824,13 @@ def value_counts(self, dropna: bool = True):
     # Indexing
     # --------
     @overload
-    def __getitem__(self, key: int | np.integer) -> Any:
+    def __getitem__(self, key: ScalarIndexer) -> Any:
         ...
 
     @overload
     def __getitem__(
         self: SparseArrayT,
-        key: slice | np.ndarray | list[int] | tuple[int | ellipsis, ...],
+        key: SequenceIndexer | tuple[int | ellipsis, ...],
     ) -> SparseArrayT:
         ...
 
diff --git a/pandas/core/arrays/string_arrow.py b/pandas/core/arrays/string_arrow.py
index e3338abd12ade..4be7f4eb0c521 100644
--- a/pandas/core/arrays/string_arrow.py
+++ b/pandas/core/arrays/string_arrow.py
@@ -22,6 +22,8 @@
     NpDtype,
     PositionalIndexer,
     Scalar,
+    ScalarIndexer,
+    SequenceIndexer,
 )
 from pandas.compat import (
     pa_version_under1p0,
@@ -268,13 +270,11 @@ def _concat_same_type(cls, to_concat) -> ArrowStringArray:
         )
 
     @overload
-    def __getitem__(self, item: int | np.integer) -> ArrowStringScalarOrNAT:
+    def __getitem__(self, item: ScalarIndexer) -> ArrowStringScalarOrNAT:
         ...
 
     @overload
-    def __getitem__(
-        self: ArrowStringArray, item: slice | np.ndarray | list[int]
-    ) -> ArrowStringArray:
+    def __getitem__(self: ArrowStringArray, item: SequenceIndexer) -> ArrowStringArray:
         ...
 
     def __getitem__(
diff --git a/pandas/core/groupby/groupby.py b/pandas/core/groupby/groupby.py
index 0e358e611f418..ec8775cf78571 100644
--- a/pandas/core/groupby/groupby.py
+++ b/pandas/core/groupby/groupby.py
@@ -2984,10 +2984,9 @@ def blk_func(values: ArrayLike) -> ArrayLike:
 
             if real_2d and values.ndim == 1:
                 assert result.shape[1] == 1, result.shape
-                # error: Invalid index type "Tuple[slice, int]" for
-                # "Union[ExtensionArray, ndarray[Any, Any]]"; expected type
-                # "Union[int, integer[Any], slice, Sequence[int], ndarray[Any, Any]]"
-                result = result[:, 0]  # type: ignore[index]
+                # error: No overload variant of "__getitem__" of "ExtensionArray"
+                # matches argument type "Tuple[slice, int]"
+                result = result[:, 0]  # type: ignore[call-overload]
                 if needs_mask:
                     mask = mask[:, 0]
 
@@ -3001,11 +3000,9 @@ def blk_func(values: ArrayLike) -> ArrayLike:
             if needs_2d and not real_2d:
                 if result.ndim == 2:
                     assert result.shape[1] == 1
-                    # error: Invalid index type "Tuple[slice, int]" for
-                    # "Union[ExtensionArray, Any, ndarray[Any, Any]]"; expected
-                    # type "Union[int, integer[Any], slice, Sequence[int],
-                    # ndarray[Any, Any]]"
-                    result = result[:, 0]  # type: ignore[index]
+                    # error: No overload variant of "__getitem__" of "ExtensionArray"
+                    # matches argument type "Tuple[slice, int]"
+                    result = result[:, 0]  # type: ignore[call-overload]
 
             return result.T
 

From b9417327db6d65649ea03c47daa6d4cfbe43bfc9 Mon Sep 17 00:00:00 2001
From: Irv Lustig <irv@princeton.com>
Date: Mon, 6 Sep 2021 16:41:04 -0400
Subject: [PATCH 12/12] comments on various indexers in _typing.py

---
 pandas/_typing.py | 4 ++++
 1 file changed, 4 insertions(+)

diff --git a/pandas/_typing.py b/pandas/_typing.py
index 6defb0aabcedc..5077e659410e3 100644
--- a/pandas/_typing.py
+++ b/pandas/_typing.py
@@ -206,6 +206,10 @@
 # indexing
 # PositionalIndexer -> valid 1D positional indexer, e.g. can pass
 # to ndarray.__getitem__
+# ScalarIndexer is for a single value as the index
+# SequenceIndexer is for list like or slices (but not tuples)
+# PositionalIndexerTuple is extends the PositionalIndexer for 2D arrays
+# These are used in various __getitem__ overloads
 # TODO: add Ellipsis, see
 # https://github.com/python/typing/issues/684#issuecomment-548203158
 # https://bugs.python.org/issue41810