From 6accf2be0000f2ce6e1c6a5091fd3ea5d32620d9 Mon Sep 17 00:00:00 2001
From: Patrick Hoefler <patrick_hoefler@gmx.net>
Date: Sun, 22 Oct 2023 23:33:54 +0200
Subject: [PATCH 01/30] BUG: setitem casting object Index to arrow strings

---
 doc/source/whatsnew/v2.1.2.rst              |  2 ++
 pandas/core/construction.py                 | 12 ++++++++++++
 pandas/core/indexes/base.py                 | 10 +++++++++-
 pandas/tests/frame/indexing/test_setitem.py |  9 +++++++++
 4 files changed, 32 insertions(+), 1 deletion(-)

diff --git a/doc/source/whatsnew/v2.1.2.rst b/doc/source/whatsnew/v2.1.2.rst
index 97a718dd496e9..0aff745d1c7f0 100644
--- a/doc/source/whatsnew/v2.1.2.rst
+++ b/doc/source/whatsnew/v2.1.2.rst
@@ -24,9 +24,11 @@ Fixed regressions
 Bug fixes
 ~~~~~~~~~
 - Fixed bug in :meth:`Categorical.equals` if other has arrow backed string dtype (:issue:`55364`)
+- Fixed bug in :meth:`DataFrame.__setitem__` casting :class:`Index` with object-dtype to PyArrow backed strings when ``infer_string`` option is set (:issue:`55638`)
 - Fixed bug in :meth:`DataFrame.__setitem__` not inferring string dtype for zero-dimensional array with ``infer_string=True`` (:issue:`55366`)
 - Fixed bug in :meth:`DataFrame.idxmin` and :meth:`DataFrame.idxmax` raising for arrow dtypes (:issue:`55368`)
 - Fixed bug in :meth:`DataFrame.interpolate` raising incorrect error message (:issue:`55347`)
+- Fixed bug in :meth:`Index.insert` casting object-dtype to PyArrow backed strings when ``infer_string`` option is set (:issue:`55638`)
 - Fixed bug in :meth:`Index.insert` raising when inserting ``None`` into :class:`Index` with ``dtype="string[pyarrow_numpy]"`` (:issue:`55365`)
 - Fixed bug in :meth:`Series.all`  and :meth:`Series.any` not treating missing values correctly for ``dtype="string[pyarrow_numpy]"`` (:issue:`55367`)
 - Fixed bug in :meth:`Series.floordiv` for :class:`ArrowDtype` (:issue:`55561`)
diff --git a/pandas/core/construction.py b/pandas/core/construction.py
index b4b9a4176472d..b0241f950662d 100644
--- a/pandas/core/construction.py
+++ b/pandas/core/construction.py
@@ -49,6 +49,7 @@
 from pandas.core.dtypes.common import (
     is_list_like,
     is_object_dtype,
+    is_string_dtype,
     pandas_dtype,
 )
 from pandas.core.dtypes.dtypes import NumpyEADtype
@@ -547,6 +548,10 @@ def sanitize_array(
         # Avoid ending up with a NumpyExtensionArray
         dtype = dtype.numpy_dtype
 
+    object_index = False
+    if isinstance(data, ABCIndex) and data.dtype == object and dtype is None:
+        object_index = True
+
     # extract ndarray or ExtensionArray, ensure we have no NumpyExtensionArray
     data = extract_array(data, extract_numpy=True, extract_range=True)
 
@@ -596,6 +601,13 @@ def sanitize_array(
             subarr = data
             if data.dtype == object:
                 subarr = maybe_infer_to_datetimelike(data)
+                if (
+                    object_index
+                    and using_pyarrow_string_dtype()
+                    and is_string_dtype(subarr)
+                ):
+                    # Avoid inference when string option is set
+                    subarr = data
             elif data.dtype.kind == "U" and using_pyarrow_string_dtype():
                 from pandas.core.arrays.string_ import StringDtype
 
diff --git a/pandas/core/indexes/base.py b/pandas/core/indexes/base.py
index 252e88d7c7d51..4414f358b39bb 100644
--- a/pandas/core/indexes/base.py
+++ b/pandas/core/indexes/base.py
@@ -23,6 +23,7 @@
 from pandas._config import (
     get_option,
     using_copy_on_write,
+    using_pyarrow_string_dtype,
 )
 
 from pandas._libs import (
@@ -6916,7 +6917,14 @@ def insert(self, loc: int, item) -> Index:
             loc = loc if loc >= 0 else loc - 1
             new_values[loc] = item
 
-        return Index._with_infer(new_values, name=self.name)
+        idx = Index._with_infer(new_values, name=self.name)
+        if (
+            using_pyarrow_string_dtype()
+            and is_string_dtype(idx.dtype)
+            and new_values.dtype == object
+        ):
+            idx = idx.astype(new_values.dtype)
+        return idx
 
     def drop(
         self,
diff --git a/pandas/tests/frame/indexing/test_setitem.py b/pandas/tests/frame/indexing/test_setitem.py
index ccc1249088f9a..fc2e817b1600e 100644
--- a/pandas/tests/frame/indexing/test_setitem.py
+++ b/pandas/tests/frame/indexing/test_setitem.py
@@ -753,6 +753,15 @@ def test_setitem_frame_overwrite_with_ea_dtype(self, any_numeric_ea_dtype):
         )
         tm.assert_frame_equal(df, expected)
 
+    def test_setitem_string_option_object_index(self):
+        # GH#55638
+        pytest.importorskip("pyarrow")
+        df = DataFrame({"a": [1, 2]})
+        with pd.option_context("future.infer_string", True):
+            df["b"] = Index(["a", "b"], dtype=object)
+        expected = DataFrame({"a": [1, 2], "b": Series(["a", "b"], dtype=object)})
+        tm.assert_frame_equal(df, expected)
+
     def test_setitem_frame_midx_columns(self):
         # GH#49121
         df = DataFrame({("a", "b"): [10]})

From 61f49034ff98d4a5014f238c11c5d5ad9a84e49b Mon Sep 17 00:00:00 2001
From: Patrick Hoefler <patrick_hoefler@gmx.net>
Date: Mon, 23 Oct 2023 00:08:28 +0200
Subject: [PATCH 02/30] Fix

---
 pandas/tests/frame/indexing/test_indexing.py | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/pandas/tests/frame/indexing/test_indexing.py b/pandas/tests/frame/indexing/test_indexing.py
index de8df15a9d747..51b0a0a13d90b 100644
--- a/pandas/tests/frame/indexing/test_indexing.py
+++ b/pandas/tests/frame/indexing/test_indexing.py
@@ -1913,7 +1913,7 @@ def test_add_new_column_infer_string():
         df.loc[df["x"] == 1, "y"] = "1"
     expected = DataFrame(
         {"x": [1], "y": Series(["1"], dtype="string[pyarrow_numpy]")},
-        columns=Index(["x", "y"], dtype="string[pyarrow_numpy]"),
+        columns=Index(["x", "y"], dtype=object),
     )
     tm.assert_frame_equal(df, expected)
 

From 2effd1f304a54ee76cb6a864e6a340c7f1934eb4 Mon Sep 17 00:00:00 2001
From: Patrick Hoefler <61934744+phofl@users.noreply.github.com>
Date: Fri, 3 Nov 2023 20:43:06 -0400
Subject: [PATCH 03/30] Start fixing index tests

---
 pandas/conftest.py                    |  8 ++++++++
 pandas/core/config_init.py            |  2 +-
 pandas/tests/indexes/test_base.py     | 26 ++++++++++++++++++--------
 pandas/tests/indexes/test_old_base.py |  1 +
 4 files changed, 28 insertions(+), 9 deletions(-)

diff --git a/pandas/conftest.py b/pandas/conftest.py
index efe263a41afe1..df2cad430bf58 100644
--- a/pandas/conftest.py
+++ b/pandas/conftest.py
@@ -2008,6 +2008,14 @@ def using_copy_on_write() -> bool:
     )
 
 
+@pytest.fixture
+def using_infer_string() -> bool:
+    """
+    Fixture to check if Copy-on-Write is enabled.
+    """
+    return pd.options.future.infer_string is True
+
+
 @pytest.fixture
 def warn_copy_on_write() -> bool:
     """
diff --git a/pandas/core/config_init.py b/pandas/core/config_init.py
index a8b63f97141c2..bdbab78a443de 100644
--- a/pandas/core/config_init.py
+++ b/pandas/core/config_init.py
@@ -905,7 +905,7 @@ def register_converter_cb(key) -> None:
 with cf.config_prefix("future"):
     cf.register_option(
         "infer_string",
-        False,
+        True,
         "Whether to infer sequence of str objects as pyarrow string "
         "dtype, which will be the default in pandas 3.0 "
         "(at which point this option will be deprecated).",
diff --git a/pandas/tests/indexes/test_base.py b/pandas/tests/indexes/test_base.py
index 0c32149124ac6..1d3379ad29750 100644
--- a/pandas/tests/indexes/test_base.py
+++ b/pandas/tests/indexes/test_base.py
@@ -70,12 +70,17 @@ def test_constructor_casting(self, index):
         tm.assert_index_equal(index, new_index)
 
     @pytest.mark.parametrize("index", ["string"], indirect=True)
-    def test_constructor_copy(self, index):
+    def test_constructor_copy(self, index, using_infer_string):
         arr = np.array(index)
         new_index = Index(arr, copy=True, name="name")
         assert isinstance(new_index, Index)
         assert new_index.name == "name"
-        tm.assert_numpy_array_equal(arr, new_index.values)
+        if using_infer_string:
+            tm.assert_extension_array_equal(
+                new_index.values, pd.array(arr, dtype="string[pyarrow_numpy]")
+            )
+        else:
+            tm.assert_numpy_array_equal(arr, new_index.values)
         arr[0] = "SOMEBIGLONGSTRING"
         assert new_index[0] != "SOMEBIGLONGSTRING"
 
@@ -143,7 +148,7 @@ def test_constructor_from_series_freq(self):
 
         tm.assert_index_equal(result, expected)
 
-    def test_constructor_from_frame_series_freq(self):
+    def test_constructor_from_frame_series_freq(self, using_infer_string):
         # GH 6273
         # create from a series, passing a freq
         dts = ["1-1-1990", "2-1-1990", "3-1-1990", "4-1-1990", "5-1-1990"]
@@ -152,8 +157,8 @@ def test_constructor_from_frame_series_freq(self):
         df = DataFrame(np.random.default_rng(2).random((5, 3)))
         df["date"] = dts
         result = DatetimeIndex(df["date"], freq="MS")
-
-        assert df["date"].dtype == object
+        dtype = object if not using_infer_string else "string"
+        assert df["date"].dtype == dtype
         expected.name = "date"
         tm.assert_index_equal(result, expected)
 
@@ -344,6 +349,9 @@ def test_view_with_args_object_array_raises(self, index):
             msg = "When changing to a larger dtype"
             with pytest.raises(ValueError, match=msg):
                 index.view("i8")
+        elif index.dtype == "string":
+            with pytest.raises(NotImplementedError, match="i8"):
+                index.view("i8")
         else:
             msg = "Cannot change data-type for object array"
             with pytest.raises(TypeError, match=msg):
@@ -477,7 +485,7 @@ def test_empty_fancy_raises(self, index):
 
         assert index[[]].identical(empty_index)
         # np.ndarray only accepts ndarray of int & bool dtypes, so should Index
-        msg = r"arrays used as indices must be of integer \(or boolean\) type"
+        msg = r"arrays used as indices must be of integer"
         with pytest.raises(IndexError, match=msg):
             index[empty_farr]
 
@@ -653,7 +661,9 @@ def test_is_numeric(self, index, expected):
         ],
         indirect=["index"],
     )
-    def test_is_object(self, index, expected):
+    def test_is_object(self, index, expected, using_infer_string):
+        if using_infer_string and index.dtype == "string" and expected:
+            expected = False
         assert is_object_dtype(index) is expected
 
     def test_summary(self, index):
@@ -773,7 +783,7 @@ def test_drop_by_numeric_label_errors_ignore(self, key, expected):
     def test_drop_tuple(self, values, to_drop):
         # GH 18304
         index = Index(values)
-        expected = Index(["b"])
+        expected = Index(["b"], dtype=object)
 
         result = index.drop(to_drop)
         tm.assert_index_equal(result, expected)
diff --git a/pandas/tests/indexes/test_old_base.py b/pandas/tests/indexes/test_old_base.py
index 32adbc693390b..cd6bb99b5e647 100644
--- a/pandas/tests/indexes/test_old_base.py
+++ b/pandas/tests/indexes/test_old_base.py
@@ -413,6 +413,7 @@ def test_insert_base(self, index):
         # test 0th element
         assert index[0:4].equals(result.insert(0, index[0]))
 
+    @pytest.mark.skip()
     def test_insert_out_of_bounds(self, index):
         # TypeError/IndexError matches what np.insert raises in these cases
 

From 6569663010e6e134dc8853452d2ef7e2b0da9ab1 Mon Sep 17 00:00:00 2001
From: Patrick Hoefler <61934744+phofl@users.noreply.github.com>
Date: Fri, 3 Nov 2023 20:46:44 -0400
Subject: [PATCH 04/30] BUG: Index.isin raising for arrow strings and null set

---
 doc/source/whatsnew/v2.1.3.rst     |  2 +-
 pandas/core/arrays/string_arrow.py |  4 +++-
 pandas/tests/indexes/test_base.py  | 12 ++++++++++--
 3 files changed, 14 insertions(+), 4 deletions(-)

diff --git a/doc/source/whatsnew/v2.1.3.rst b/doc/source/whatsnew/v2.1.3.rst
index 3b1cd1c152baa..31ab01f171b4a 100644
--- a/doc/source/whatsnew/v2.1.3.rst
+++ b/doc/source/whatsnew/v2.1.3.rst
@@ -22,7 +22,7 @@ Fixed regressions
 Bug fixes
 ~~~~~~~~~
 - Bug in :meth:`DatetimeIndex.diff` raising ``TypeError`` (:issue:`55080`)
--
+- Bug in :meth:`Index.isin` raising for Arrow backed string and ``None`` value (:issue:`55821`)
 
 .. ---------------------------------------------------------------------------
 .. _whatsnew_213.other:
diff --git a/pandas/core/arrays/string_arrow.py b/pandas/core/arrays/string_arrow.py
index ae020ec2f599f..1c2ecd4684e2f 100644
--- a/pandas/core/arrays/string_arrow.py
+++ b/pandas/core/arrays/string_arrow.py
@@ -222,7 +222,9 @@ def isin(self, values) -> npt.NDArray[np.bool_]:
         if not len(value_set):
             return np.zeros(len(self), dtype=bool)
 
-        result = pc.is_in(self._pa_array, value_set=pa.array(value_set))
+        result = pc.is_in(
+            self._pa_array, value_set=pa.array(value_set, type=self._pa_array.type)
+        )
         # pyarrow 2.0.0 returned nulls, so we explicily specify dtype to convert nulls
         # to False
         return np.array(result, dtype=np.bool_)
diff --git a/pandas/tests/indexes/test_base.py b/pandas/tests/indexes/test_base.py
index 0c32149124ac6..828700573c7ea 100644
--- a/pandas/tests/indexes/test_base.py
+++ b/pandas/tests/indexes/test_base.py
@@ -9,7 +9,7 @@
 
 from pandas.compat import IS64
 from pandas.errors import InvalidIndexError
-from pandas.util._test_decorators import async_mark
+import pandas.util._test_decorators as td
 
 from pandas.core.dtypes.common import (
     is_any_real_numeric_dtype,
@@ -921,6 +921,14 @@ def test_isin_empty(self, empty):
         result = index.isin(empty)
         tm.assert_numpy_array_equal(expected, result)
 
+    @td.skip_if_no("pyarrow")
+    def test_isin_arrow_string_null(self):
+        # GH#55821
+        index = Index(["a", "b"], dtype="string[pyarrow_numpy]")
+        result = index.isin([None])
+        expected = np.array([False, False])
+        tm.assert_numpy_array_equal(result, expected)
+
     @pytest.mark.parametrize(
         "values",
         [
@@ -1235,7 +1243,7 @@ def test_cached_properties_not_settable(self):
         with pytest.raises(AttributeError, match="Can't set attribute"):
             index.is_unique = False
 
-    @async_mark()
+    @td.async_mark()
     async def test_tab_complete_warning(self, ip):
         # https://github.com/pandas-dev/pandas/issues/16409
         pytest.importorskip("IPython", minversion="6.0.0")

From cdbdba2993ba2f633ef1b070f3434533c6b33044 Mon Sep 17 00:00:00 2001
From: Patrick Hoefler <61934744+phofl@users.noreply.github.com>
Date: Fri, 3 Nov 2023 21:00:26 -0400
Subject: [PATCH 05/30] Fix more tests

---
 pandas/tests/indexes/test_base.py     | 19 ++++++++++++++-----
 pandas/tests/indexes/test_old_base.py |  4 ++--
 2 files changed, 16 insertions(+), 7 deletions(-)

diff --git a/pandas/tests/indexes/test_base.py b/pandas/tests/indexes/test_base.py
index b8756fce3dd95..0eb9f9cdc5072 100644
--- a/pandas/tests/indexes/test_base.py
+++ b/pandas/tests/indexes/test_base.py
@@ -836,9 +836,12 @@ def test_isin(self, values, index, expected):
         result = index.isin(values)
         tm.assert_numpy_array_equal(result, expected)
 
-    def test_isin_nan_common_object(self, nulls_fixture, nulls_fixture2):
+    def test_isin_nan_common_object(
+        self, nulls_fixture, nulls_fixture2, using_infer_string
+    ):
         # Test cartesian product of null fixtures and ensure that we don't
         # mangle the various types (save a corner case with PyPy)
+        idx = Index(["a", nulls_fixture])
 
         # all nans are the same
         if (
@@ -848,19 +851,25 @@ def test_isin_nan_common_object(self, nulls_fixture, nulls_fixture2):
             and math.isnan(nulls_fixture2)
         ):
             tm.assert_numpy_array_equal(
-                Index(["a", nulls_fixture]).isin([nulls_fixture2]),
+                idx.isin([nulls_fixture2]),
                 np.array([False, True]),
             )
 
         elif nulls_fixture is nulls_fixture2:  # should preserve NA type
             tm.assert_numpy_array_equal(
-                Index(["a", nulls_fixture]).isin([nulls_fixture2]),
+                idx.isin([nulls_fixture2]),
+                np.array([False, True]),
+            )
+
+        elif using_infer_string and idx.dtype == "string":
+            tm.assert_numpy_array_equal(
+                idx.isin([nulls_fixture2]),
                 np.array([False, True]),
             )
 
         else:
             tm.assert_numpy_array_equal(
-                Index(["a", nulls_fixture]).isin([nulls_fixture2]),
+                idx.isin([nulls_fixture2]),
                 np.array([False, False]),
             )
 
@@ -1136,7 +1145,7 @@ def test_reindex_preserves_name_if_target_is_list_or_ndarray(self, name, labels)
     def test_reindex_preserves_type_if_target_is_empty_list_or_array(self, labels):
         # GH7774
         index = Index(list("abc"))
-        assert index.reindex(labels)[0].dtype.type == np.object_
+        assert index.reindex(labels)[0].dtype.type == index.dtype.type
 
     @pytest.mark.parametrize(
         "labels,dtype",
diff --git a/pandas/tests/indexes/test_old_base.py b/pandas/tests/indexes/test_old_base.py
index cd6bb99b5e647..9b945c5f7a3a1 100644
--- a/pandas/tests/indexes/test_old_base.py
+++ b/pandas/tests/indexes/test_old_base.py
@@ -209,7 +209,7 @@ def test_numeric_compat(self, simple_index):
             1 // idx
 
     def test_logical_compat(self, simple_index):
-        if simple_index.dtype == object:
+        if simple_index.dtype in (object, "string"):
             pytest.skip("Tested elsewhere.")
         idx = simple_index
         if idx.dtype.kind in "iufcbm":
@@ -295,7 +295,7 @@ def test_ensure_copied_data(self, index):
                 tm.assert_numpy_array_equal(
                     index._values._ndarray, result._values._ndarray, check_same="same"
                 )
-            elif index.dtype == "string[pyarrow]":
+            elif index.dtype in ("string[pyarrow]", "string[pyarrow_numpy]"):
                 assert tm.shares_memory(result._values, index._values)
             else:
                 raise NotImplementedError(index.dtype)

From 57cb8f43a9942199b6f4d512ecf22baedce754c2 Mon Sep 17 00:00:00 2001
From: Patrick Hoefler <61934744+phofl@users.noreply.github.com>
Date: Fri, 3 Nov 2023 21:09:40 -0400
Subject: [PATCH 06/30] TST: Fix shares_memory for arrow string dtype

---
 pandas/_testing/__init__.py             | 13 +++++++++++--
 pandas/tests/util/test_shares_memory.py | 16 ++++++++++++++++
 2 files changed, 27 insertions(+), 2 deletions(-)

diff --git a/pandas/_testing/__init__.py b/pandas/_testing/__init__.py
index 81cd504119c38..ee173c6892743 100644
--- a/pandas/_testing/__init__.py
+++ b/pandas/_testing/__init__.py
@@ -30,6 +30,7 @@
     is_float_dtype,
     is_sequence,
     is_signed_integer_dtype,
+    is_string_dtype,
     is_unsigned_integer_dtype,
     pandas_dtype,
 )
@@ -1044,10 +1045,18 @@ def shares_memory(left, right) -> bool:
     if isinstance(left, pd.core.arrays.IntervalArray):
         return shares_memory(left._left, right) or shares_memory(left._right, right)
 
-    if isinstance(left, ExtensionArray) and left.dtype == "string[pyarrow]":
+    if (
+        isinstance(left, ExtensionArray)
+        and is_string_dtype(left.dtype)
+        and left.dtype.storage in ("pyarrow", "pyarrow_numpy")
+    ):
         # https://github.com/pandas-dev/pandas/pull/43930#discussion_r736862669
         left = cast("ArrowExtensionArray", left)
-        if isinstance(right, ExtensionArray) and right.dtype == "string[pyarrow]":
+        if (
+            isinstance(right, ExtensionArray)
+            and is_string_dtype(right.dtype)
+            and right.dtype.storage in ("pyarrow", "pyarrow_numpy")
+        ):
             right = cast("ArrowExtensionArray", right)
             left_pa_data = left._pa_array
             right_pa_data = right._pa_array
diff --git a/pandas/tests/util/test_shares_memory.py b/pandas/tests/util/test_shares_memory.py
index ed8227a5c4307..e4dca0f36b111 100644
--- a/pandas/tests/util/test_shares_memory.py
+++ b/pandas/tests/util/test_shares_memory.py
@@ -1,3 +1,5 @@
+import pandas.util._test_decorators as td
+
 import pandas as pd
 import pandas._testing as tm
 
@@ -11,3 +13,17 @@ def test_shares_memory_interval():
     assert tm.shares_memory(obj, obj[:2])
 
     assert not tm.shares_memory(obj, obj._data.copy())
+
+
+@td.skip_if_no("pyarrow")
+def test_shares_memory_string():
+    import pyarrow as pa
+
+    obj = pd.array(["a", "b"], dtype="string[pyarrow]")
+    assert tm.shares_memory(obj, obj)
+
+    obj = pd.array(["a", "b"], dtype="string[pyarrow_numpy]")
+    assert tm.shares_memory(obj, obj)
+
+    obj = pd.array(["a", "b"], dtype=pd.ArrowDtype(pa.string()))
+    assert tm.shares_memory(obj, obj)

From 7697059e5b653f74ae5f3962ec02f5e1b715aa2d Mon Sep 17 00:00:00 2001
From: Patrick Hoefler <61934744+phofl@users.noreply.github.com>
Date: Fri, 3 Nov 2023 21:09:51 -0400
Subject: [PATCH 07/30] TST: Fix shares_memory for arrow string dtype

---
 pandas/tests/util/test_shares_memory.py | 1 +
 1 file changed, 1 insertion(+)

diff --git a/pandas/tests/util/test_shares_memory.py b/pandas/tests/util/test_shares_memory.py
index e4dca0f36b111..090b04d5b6b01 100644
--- a/pandas/tests/util/test_shares_memory.py
+++ b/pandas/tests/util/test_shares_memory.py
@@ -17,6 +17,7 @@ def test_shares_memory_interval():
 
 @td.skip_if_no("pyarrow")
 def test_shares_memory_string():
+    # GH#55822
     import pyarrow as pa
 
     obj = pd.array(["a", "b"], dtype="string[pyarrow]")

From 285f7bbf1436cff70fc7701bd6b172bf905e2d66 Mon Sep 17 00:00:00 2001
From: Patrick Hoefler <61934744+phofl@users.noreply.github.com>
Date: Fri, 3 Nov 2023 21:10:22 -0400
Subject: [PATCH 08/30] TST: Fix shares_memory for arrow string dtype

---
 pandas/tests/util/test_shares_memory.py | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/pandas/tests/util/test_shares_memory.py b/pandas/tests/util/test_shares_memory.py
index 090b04d5b6b01..00a897d574a07 100644
--- a/pandas/tests/util/test_shares_memory.py
+++ b/pandas/tests/util/test_shares_memory.py
@@ -17,7 +17,7 @@ def test_shares_memory_interval():
 
 @td.skip_if_no("pyarrow")
 def test_shares_memory_string():
-    # GH#55822
+    # GH#55823
     import pyarrow as pa
 
     obj = pd.array(["a", "b"], dtype="string[pyarrow]")

From f59c4bc41fe11b42a0fb88f977ddf994e48d5037 Mon Sep 17 00:00:00 2001
From: Patrick Hoefler <61934744+phofl@users.noreply.github.com>
Date: Sat, 4 Nov 2023 22:43:01 -0400
Subject: [PATCH 09/30] Fix more tests

---
 pandas/tests/indexes/base_class/test_setops.py    |  2 +-
 pandas/tests/indexes/object/test_indexing.py      |  2 +-
 .../tests/indexes/period/methods/test_astype.py   |  2 +-
 pandas/tests/indexes/test_old_base.py             | 15 +++++++++++----
 .../indexes/timedeltas/methods/test_astype.py     |  2 +-
 5 files changed, 15 insertions(+), 8 deletions(-)

diff --git a/pandas/tests/indexes/base_class/test_setops.py b/pandas/tests/indexes/base_class/test_setops.py
index 488f79eea0d11..b6c318c1b1650 100644
--- a/pandas/tests/indexes/base_class/test_setops.py
+++ b/pandas/tests/indexes/base_class/test_setops.py
@@ -147,7 +147,7 @@ def test_intersection_str_dates(self, sort):
     def test_intersection_non_monotonic_non_unique(self, index2, expected_arr, sort):
         # non-monotonic non-unique
         index1 = Index(["A", "B", "A", "C"])
-        expected = Index(expected_arr, dtype="object")
+        expected = Index(expected_arr)
         result = index1.intersection(index2, sort=sort)
         if sort is None:
             expected = expected.sort_values()
diff --git a/pandas/tests/indexes/object/test_indexing.py b/pandas/tests/indexes/object/test_indexing.py
index 87d3afc77d556..a6f48cf041060 100644
--- a/pandas/tests/indexes/object/test_indexing.py
+++ b/pandas/tests/indexes/object/test_indexing.py
@@ -172,7 +172,7 @@ def test_slice_locs_negative_step(self, in_slice, expected):
 
         s_start, s_stop = index.slice_locs(in_slice.start, in_slice.stop, in_slice.step)
         result = index[s_start : s_stop : in_slice.step]
-        expected = Index(list(expected))
+        expected = Index(list(expected), dtype=result.dtype)
         tm.assert_index_equal(result, expected)
 
     def test_slice_locs_dup(self):
diff --git a/pandas/tests/indexes/period/methods/test_astype.py b/pandas/tests/indexes/period/methods/test_astype.py
index 07595b6b8c1dd..e886f923f9656 100644
--- a/pandas/tests/indexes/period/methods/test_astype.py
+++ b/pandas/tests/indexes/period/methods/test_astype.py
@@ -41,7 +41,7 @@ def test_astype_conversion(self):
         tm.assert_index_equal(result, expected)
 
         result = idx.astype(str)
-        expected = Index([str(x) for x in idx], name="idx")
+        expected = Index([str(x) for x in idx], name="idx", dtype=object)
         tm.assert_index_equal(result, expected)
 
         idx = period_range("1990", "2009", freq="Y", name="idx")
diff --git a/pandas/tests/indexes/test_old_base.py b/pandas/tests/indexes/test_old_base.py
index 9b945c5f7a3a1..bd3bdfe4bbce0 100644
--- a/pandas/tests/indexes/test_old_base.py
+++ b/pandas/tests/indexes/test_old_base.py
@@ -691,7 +691,7 @@ def test_map_str(self, simple_index):
             pytest.skip("See test_map.py")
         idx = simple_index
         result = idx.map(str)
-        expected = Index([str(x) for x in idx], dtype=object)
+        expected = Index([str(x) for x in idx])
         tm.assert_index_equal(result, expected)
 
     @pytest.mark.parametrize("copy", [True, False])
@@ -827,7 +827,7 @@ def test_append_preserves_dtype(self, simple_index):
         alt = index.take(list(range(N)) * 2)
         tm.assert_index_equal(result, alt, check_exact=True)
 
-    def test_inv(self, simple_index):
+    def test_inv(self, simple_index, using_infer_string):
         idx = simple_index
 
         if idx.dtype.kind in ["i", "u"]:
@@ -840,14 +840,21 @@ def test_inv(self, simple_index):
             tm.assert_series_equal(res2, Series(expected))
         else:
             if idx.dtype.kind == "f":
+                err = TypeError
                 msg = "ufunc 'invert' not supported for the input types"
+            elif using_infer_string and idx.dtype == "string":
+                import pyarrow as pa
+
+                err = pa.lib.ArrowNotImplementedError
+                msg = "has no kernel"
             else:
+                err = TypeError
                 msg = "bad operand"
-            with pytest.raises(TypeError, match=msg):
+            with pytest.raises(err, match=msg):
                 ~idx
 
             # check that we get the same behavior with Series
-            with pytest.raises(TypeError, match=msg):
+            with pytest.raises(err, match=msg):
                 ~Series(idx)
 
     def test_is_boolean_is_deprecated(self, simple_index):
diff --git a/pandas/tests/indexes/timedeltas/methods/test_astype.py b/pandas/tests/indexes/timedeltas/methods/test_astype.py
index 6aeb1fc23dae9..311f2b5c9aa59 100644
--- a/pandas/tests/indexes/timedeltas/methods/test_astype.py
+++ b/pandas/tests/indexes/timedeltas/methods/test_astype.py
@@ -61,7 +61,7 @@ def test_astype(self):
         tm.assert_index_equal(result, expected)
 
         result = idx.astype(str)
-        expected = Index([str(x) for x in idx], name="idx")
+        expected = Index([str(x) for x in idx], name="idx", dtype=object)
         tm.assert_index_equal(result, expected)
 
         rng = timedelta_range("1 days", periods=10)

From ff6e271961434063b277e723e918cc7f3cbc5dea Mon Sep 17 00:00:00 2001
From: Patrick Hoefler <61934744+phofl@users.noreply.github.com>
Date: Sat, 4 Nov 2023 22:47:43 -0400
Subject: [PATCH 10/30] BUG: Index.getitem returning wrong result with negative
 step for arrow

---
 doc/source/whatsnew/v2.1.3.rst               |  2 +-
 pandas/core/arrays/arrow/array.py            |  7 +++++++
 pandas/tests/indexes/object/test_indexing.py | 14 +++++++++++---
 3 files changed, 19 insertions(+), 4 deletions(-)

diff --git a/doc/source/whatsnew/v2.1.3.rst b/doc/source/whatsnew/v2.1.3.rst
index 3b1cd1c152baa..f4c32b6ecc056 100644
--- a/doc/source/whatsnew/v2.1.3.rst
+++ b/doc/source/whatsnew/v2.1.3.rst
@@ -22,7 +22,7 @@ Fixed regressions
 Bug fixes
 ~~~~~~~~~
 - Bug in :meth:`DatetimeIndex.diff` raising ``TypeError`` (:issue:`55080`)
--
+- Bug in :meth:`Index.__getitem__` returning wrong result for Arrow dtypes and negative stepsize (:issue:`55832`)
 
 .. ---------------------------------------------------------------------------
 .. _whatsnew_213.other:
diff --git a/pandas/core/arrays/arrow/array.py b/pandas/core/arrays/arrow/array.py
index 4bcc03643dac8..43927f554a875 100644
--- a/pandas/core/arrays/arrow/array.py
+++ b/pandas/core/arrays/arrow/array.py
@@ -553,6 +553,13 @@ def __getitem__(self, item: PositionalIndexer):
             )
         # We are not an array indexer, so maybe e.g. a slice or integer
         # indexer. We dispatch to pyarrow.
+        if isinstance(item, slice):
+            if item.start == item.stop:
+                pass
+            elif item.start == -len(self) - 1:
+                item = slice(None, item.stop, item.step)
+            elif item.stop == -len(self) - 1:
+                item = slice(item.start, None, item.step)
         value = self._pa_array[item]
         if isinstance(value, pa.ChunkedArray):
             return type(self)(value)
diff --git a/pandas/tests/indexes/object/test_indexing.py b/pandas/tests/indexes/object/test_indexing.py
index 87d3afc77d556..7a6e93a4605d7 100644
--- a/pandas/tests/indexes/object/test_indexing.py
+++ b/pandas/tests/indexes/object/test_indexing.py
@@ -4,6 +4,7 @@
 import pytest
 
 from pandas._libs.missing import is_matching_na
+import pandas.util._test_decorators as td
 
 import pandas as pd
 from pandas import Index
@@ -144,6 +145,13 @@ def test_get_indexer_non_unique_np_nats(self, np_nat_fixture, np_nat_fixture2):
 
 
 class TestSliceLocs:
+    @pytest.mark.parametrize(
+        "dtype",
+        [
+            "object",
+            pytest.param("string[pyarrow_numpy]", marks=td.skip_if_no("pyarrow")),
+        ],
+    )
     @pytest.mark.parametrize(
         "in_slice,expected",
         [
@@ -167,12 +175,12 @@ class TestSliceLocs:
             (pd.IndexSlice["m":"m":-1], ""),  # type: ignore[misc]
         ],
     )
-    def test_slice_locs_negative_step(self, in_slice, expected):
-        index = Index(list("bcdxy"))
+    def test_slice_locs_negative_step(self, in_slice, expected, dtype):
+        index = Index(list("bcdxy"), dtype=dtype)
 
         s_start, s_stop = index.slice_locs(in_slice.start, in_slice.stop, in_slice.step)
         result = index[s_start : s_stop : in_slice.step]
-        expected = Index(list(expected))
+        expected = Index(list(expected), dtype=dtype)
         tm.assert_index_equal(result, expected)
 
     def test_slice_locs_dup(self):

From 7684911a70609559f7efcb84c73662ba0579a169 Mon Sep 17 00:00:00 2001
From: Patrick Hoefler <61934744+phofl@users.noreply.github.com>
Date: Sun, 5 Nov 2023 09:18:25 -0500
Subject: [PATCH 11/30] Update

---
 pandas/core/arrays/arrow/array.py            |  5 +++--
 pandas/tests/indexes/object/test_indexing.py | 11 +++++++++++
 2 files changed, 14 insertions(+), 2 deletions(-)

diff --git a/pandas/core/arrays/arrow/array.py b/pandas/core/arrays/arrow/array.py
index 43927f554a875..e073b8c20acbe 100644
--- a/pandas/core/arrays/arrow/array.py
+++ b/pandas/core/arrays/arrow/array.py
@@ -556,10 +556,11 @@ def __getitem__(self, item: PositionalIndexer):
         if isinstance(item, slice):
             if item.start == item.stop:
                 pass
-            elif item.start == -len(self) - 1:
+            elif item.start <= -len(self) - 1:
                 item = slice(None, item.stop, item.step)
-            elif item.stop == -len(self) - 1:
+            elif item.stop <= -len(self) - 1:
                 item = slice(item.start, None, item.step)
+
         value = self._pa_array[item]
         if isinstance(value, pa.ChunkedArray):
             return type(self)(value)
diff --git a/pandas/tests/indexes/object/test_indexing.py b/pandas/tests/indexes/object/test_indexing.py
index 7a6e93a4605d7..93d46ebdd0b51 100644
--- a/pandas/tests/indexes/object/test_indexing.py
+++ b/pandas/tests/indexes/object/test_indexing.py
@@ -183,6 +183,17 @@ def test_slice_locs_negative_step(self, in_slice, expected, dtype):
         expected = Index(list(expected), dtype=dtype)
         tm.assert_index_equal(result, expected)
 
+    @td.skip_if_no("pyarrow")
+    def test_slice_locs_negative_step_oob(self):
+        index = Index(list("bcdxy"), dtype="string[pyarrow_numpy]")
+
+        result = index[-10:5:1]
+        tm.assert_index_equal(result, index)
+
+        result = index[4:-10:-1]
+        expected = Index(list("yxdcb"), dtype="string[pyarrow_numpy]")
+        tm.assert_index_equal(result, expected)
+
     def test_slice_locs_dup(self):
         index = Index(["a", "a", "b", "c", "d", "d"])
         assert index.slice_locs("a", "d") == (0, 6)

From 7474cb2b70e548a6000dd797c8f5c524b08624cb Mon Sep 17 00:00:00 2001
From: Patrick Hoefler <61934744+phofl@users.noreply.github.com>
Date: Sun, 5 Nov 2023 09:21:00 -0500
Subject: [PATCH 12/30] Update

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

diff --git a/pandas/core/arrays/arrow/array.py b/pandas/core/arrays/arrow/array.py
index e073b8c20acbe..87d355ef79142 100644
--- a/pandas/core/arrays/arrow/array.py
+++ b/pandas/core/arrays/arrow/array.py
@@ -556,9 +556,9 @@ def __getitem__(self, item: PositionalIndexer):
         if isinstance(item, slice):
             if item.start == item.stop:
                 pass
-            elif item.start <= -len(self) - 1:
+            elif item.start < -len(self):
                 item = slice(None, item.stop, item.step)
-            elif item.stop <= -len(self) - 1:
+            elif item.stop < -len(self):
                 item = slice(item.start, None, item.step)
 
         value = self._pa_array[item]

From b27b0f82b85e20d0c3ae724acbece252999a8b26 Mon Sep 17 00:00:00 2001
From: Patrick Hoefler <61934744+phofl@users.noreply.github.com>
Date: Sun, 5 Nov 2023 09:26:59 -0500
Subject: [PATCH 13/30] Fix

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

diff --git a/pandas/core/arrays/arrow/array.py b/pandas/core/arrays/arrow/array.py
index 87d355ef79142..9a6b91f21d90f 100644
--- a/pandas/core/arrays/arrow/array.py
+++ b/pandas/core/arrays/arrow/array.py
@@ -556,9 +556,9 @@ def __getitem__(self, item: PositionalIndexer):
         if isinstance(item, slice):
             if item.start == item.stop:
                 pass
-            elif item.start < -len(self):
+            elif item.start is not None and item.start < -len(self):
                 item = slice(None, item.stop, item.step)
-            elif item.stop < -len(self):
+            elif item.stop is not None and item.stop < -len(self):
                 item = slice(item.start, None, item.step)
 
         value = self._pa_array[item]

From 659577e6c43dfe5125b865f707ec537f66b89cd3 Mon Sep 17 00:00:00 2001
From: Patrick Hoefler <61934744+phofl@users.noreply.github.com>
Date: Sun, 5 Nov 2023 10:13:53 -0500
Subject: [PATCH 14/30] Update array.py

---
 pandas/core/arrays/arrow/array.py | 2 --
 1 file changed, 2 deletions(-)

diff --git a/pandas/core/arrays/arrow/array.py b/pandas/core/arrays/arrow/array.py
index 9a6b91f21d90f..7e5452a21af9a 100644
--- a/pandas/core/arrays/arrow/array.py
+++ b/pandas/core/arrays/arrow/array.py
@@ -556,8 +556,6 @@ def __getitem__(self, item: PositionalIndexer):
         if isinstance(item, slice):
             if item.start == item.stop:
                 pass
-            elif item.start is not None and item.start < -len(self):
-                item = slice(None, item.stop, item.step)
             elif item.stop is not None and item.stop < -len(self):
                 item = slice(item.start, None, item.step)
 

From 11c0e866c4ca22d2458f6858fe544663cc1b7758 Mon Sep 17 00:00:00 2001
From: Patrick Hoefler <61934744+phofl@users.noreply.github.com>
Date: Wed, 15 Nov 2023 20:59:52 +0100
Subject: [PATCH 15/30] Fix

---
 pandas/tests/indexes/object/test_astype.py   | 2 +-
 pandas/tests/indexes/object/test_indexing.py | 3 +++
 2 files changed, 4 insertions(+), 1 deletion(-)

diff --git a/pandas/tests/indexes/object/test_astype.py b/pandas/tests/indexes/object/test_astype.py
index 273b39b5e319d..9c1ef302c5b51 100644
--- a/pandas/tests/indexes/object/test_astype.py
+++ b/pandas/tests/indexes/object/test_astype.py
@@ -20,7 +20,7 @@ def test_astype_str_from_bytes():
 
     # while we're here, check that Series.astype behaves the same
     result = Series(idx).astype(str)
-    expected = Series(expected)
+    expected = Series(expected, dtype=object)
     tm.assert_series_equal(result, expected)
 
 
diff --git a/pandas/tests/indexes/object/test_indexing.py b/pandas/tests/indexes/object/test_indexing.py
index 93d46ebdd0b51..68fc707e5c816 100644
--- a/pandas/tests/indexes/object/test_indexing.py
+++ b/pandas/tests/indexes/object/test_indexing.py
@@ -3,6 +3,8 @@
 import numpy as np
 import pytest
 
+from pandas._config import using_pyarrow_string_dtype
+
 from pandas._libs.missing import is_matching_na
 import pandas.util._test_decorators as td
 
@@ -58,6 +60,7 @@ def test_get_indexer_with_NA_values(
 
 
 class TestGetIndexerNonUnique:
+    @pytest.mark.xfail(using_pyarrow_string_dtype(), reason="NAs are cast to NaN")
     def test_get_indexer_non_unique_nas(self, nulls_fixture):
         # even though this isn't non-unique, this should still work
         index = Index(["a", "b", nulls_fixture])

From f55b62fd0e38cc3b5076becff6d31e0c3a8c17d1 Mon Sep 17 00:00:00 2001
From: Patrick Hoefler <patrick_hoefler@gmx.net>
Date: Fri, 17 Nov 2023 00:18:44 +0100
Subject: [PATCH 16/30] Move

---
 doc/source/whatsnew/v2.1.2.rst | 1 -
 doc/source/whatsnew/v2.1.4.rst | 2 +-
 2 files changed, 1 insertion(+), 2 deletions(-)

diff --git a/doc/source/whatsnew/v2.1.2.rst b/doc/source/whatsnew/v2.1.2.rst
index 68ecd8713390a..f25af040900b1 100644
--- a/doc/source/whatsnew/v2.1.2.rst
+++ b/doc/source/whatsnew/v2.1.2.rst
@@ -41,7 +41,6 @@ Bug fixes
 - Fixed bug in :meth:`.DataFrameGroupBy.min()` and :meth:`.DataFrameGroupBy.max()` not preserving extension dtype for empty object (:issue:`55619`)
 - Fixed bug in :meth:`.SeriesGroupBy.value_counts` returning incorrect dtype for string columns (:issue:`55627`)
 - Fixed bug in :meth:`Categorical.equals` if other has arrow backed string dtype (:issue:`55364`)
-- Fixed bug in :meth:`DataFrame.__setitem__` casting :class:`Index` with object-dtype to PyArrow backed strings when ``infer_string`` option is set (:issue:`55638`)
 - Fixed bug in :meth:`DataFrame.__setitem__` not inferring string dtype for zero-dimensional array with ``infer_string=True`` (:issue:`55366`)
 - Fixed bug in :meth:`DataFrame.idxmin` and :meth:`DataFrame.idxmax` raising for arrow dtypes (:issue:`55368`)
 - Fixed bug in :meth:`DataFrame.interpolate` raising incorrect error message (:issue:`55347`)
diff --git a/doc/source/whatsnew/v2.1.4.rst b/doc/source/whatsnew/v2.1.4.rst
index 04bbb0f806cbd..954f04ffc621d 100644
--- a/doc/source/whatsnew/v2.1.4.rst
+++ b/doc/source/whatsnew/v2.1.4.rst
@@ -21,7 +21,7 @@ Fixed regressions
 
 Bug fixes
 ~~~~~~~~~
--
+- Fixed bug in :meth:`DataFrame.__setitem__` casting :class:`Index` with object-dtype to PyArrow backed strings when ``infer_string`` option is set (:issue:`55638`)
 -
 
 .. ---------------------------------------------------------------------------

From ea3fcfbe6207152558e68df2bfce24ff622c9a76 Mon Sep 17 00:00:00 2001
From: Patrick Hoefler <patrick_hoefler@gmx.net>
Date: Fri, 17 Nov 2023 00:19:11 +0100
Subject: [PATCH 17/30] Move

---
 doc/source/whatsnew/v2.1.2.rst | 1 -
 doc/source/whatsnew/v2.1.4.rst | 1 +
 2 files changed, 1 insertion(+), 1 deletion(-)

diff --git a/doc/source/whatsnew/v2.1.2.rst b/doc/source/whatsnew/v2.1.2.rst
index f25af040900b1..38416afc1c94c 100644
--- a/doc/source/whatsnew/v2.1.2.rst
+++ b/doc/source/whatsnew/v2.1.2.rst
@@ -44,7 +44,6 @@ Bug fixes
 - Fixed bug in :meth:`DataFrame.__setitem__` not inferring string dtype for zero-dimensional array with ``infer_string=True`` (:issue:`55366`)
 - Fixed bug in :meth:`DataFrame.idxmin` and :meth:`DataFrame.idxmax` raising for arrow dtypes (:issue:`55368`)
 - Fixed bug in :meth:`DataFrame.interpolate` raising incorrect error message (:issue:`55347`)
-- Fixed bug in :meth:`Index.insert` casting object-dtype to PyArrow backed strings when ``infer_string`` option is set (:issue:`55638`)
 - Fixed bug in :meth:`Index.insert` raising when inserting ``None`` into :class:`Index` with ``dtype="string[pyarrow_numpy]"`` (:issue:`55365`)
 - Fixed bug in :meth:`Series.all`  and :meth:`Series.any` not treating missing values correctly for ``dtype="string[pyarrow_numpy]"`` (:issue:`55367`)
 - Fixed bug in :meth:`Series.floordiv` for :class:`ArrowDtype` (:issue:`55561`)
diff --git a/doc/source/whatsnew/v2.1.4.rst b/doc/source/whatsnew/v2.1.4.rst
index 954f04ffc621d..400448260b0ae 100644
--- a/doc/source/whatsnew/v2.1.4.rst
+++ b/doc/source/whatsnew/v2.1.4.rst
@@ -22,6 +22,7 @@ Fixed regressions
 Bug fixes
 ~~~~~~~~~
 - Fixed bug in :meth:`DataFrame.__setitem__` casting :class:`Index` with object-dtype to PyArrow backed strings when ``infer_string`` option is set (:issue:`55638`)
+- Fixed bug in :meth:`Index.insert` casting object-dtype to PyArrow backed strings when ``infer_string`` option is set (:issue:`55638`)
 -
 
 .. ---------------------------------------------------------------------------

From 86fe4f129a648fa2f765e9f3587d8829eedbc7ff Mon Sep 17 00:00:00 2001
From: Patrick Hoefler <patrick_hoefler@gmx.net>
Date: Fri, 17 Nov 2023 00:35:46 +0100
Subject: [PATCH 18/30] Fix

---
 pandas/core/arrays/arrow/array.py | 7 ++++++-
 1 file changed, 6 insertions(+), 1 deletion(-)

diff --git a/pandas/core/arrays/arrow/array.py b/pandas/core/arrays/arrow/array.py
index 7e5452a21af9a..820a3856d48fc 100644
--- a/pandas/core/arrays/arrow/array.py
+++ b/pandas/core/arrays/arrow/array.py
@@ -556,7 +556,12 @@ def __getitem__(self, item: PositionalIndexer):
         if isinstance(item, slice):
             if item.start == item.stop:
                 pass
-            elif item.stop is not None and item.stop < -len(self):
+            elif (
+                item.stop is not None
+                and item.stop < -len(self)
+                and item.step is not None
+                and item.step < 0
+            ):
                 item = slice(item.start, None, item.step)
 
         value = self._pa_array[item]

From 242552937fe6d8e12c2af90e2d55b91a9d9bdb5c Mon Sep 17 00:00:00 2001
From: Patrick Hoefler <61934744+phofl@users.noreply.github.com>
Date: Fri, 17 Nov 2023 19:35:18 +0100
Subject: [PATCH 19/30] Add gh ref

---
 pandas/core/arrays/arrow/array.py | 1 +
 1 file changed, 1 insertion(+)

diff --git a/pandas/core/arrays/arrow/array.py b/pandas/core/arrays/arrow/array.py
index 820a3856d48fc..d162b66e5d369 100644
--- a/pandas/core/arrays/arrow/array.py
+++ b/pandas/core/arrays/arrow/array.py
@@ -554,6 +554,7 @@ def __getitem__(self, item: PositionalIndexer):
         # We are not an array indexer, so maybe e.g. a slice or integer
         # indexer. We dispatch to pyarrow.
         if isinstance(item, slice):
+            # Arrow bug https://github.com/apache/arrow/issues/38768
             if item.start == item.stop:
                 pass
             elif (

From 84e815e29e8f020b9386546616cc86c00e5f2162 Mon Sep 17 00:00:00 2001
From: Patrick Hoefler <61934744+phofl@users.noreply.github.com>
Date: Sat, 18 Nov 2023 00:47:54 +0100
Subject: [PATCH 20/30] Update v2.1.4.rst

---
 doc/source/whatsnew/v2.1.4.rst | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/doc/source/whatsnew/v2.1.4.rst b/doc/source/whatsnew/v2.1.4.rst
index bdaf5bdee03f5..2a04adf2ac7f7 100644
--- a/doc/source/whatsnew/v2.1.4.rst
+++ b/doc/source/whatsnew/v2.1.4.rst
@@ -21,9 +21,9 @@ Fixed regressions
 
 Bug fixes
 ~~~~~~~~~
+- Bug in :class:`Series` constructor raising DeprecationWarning when ``index`` is a list of :class:`Series` (:issue:`55228`)
 - Fixed bug in :meth:`DataFrame.__setitem__` casting :class:`Index` with object-dtype to PyArrow backed strings when ``infer_string`` option is set (:issue:`55638`)
 - Fixed bug in :meth:`Index.insert` casting object-dtype to PyArrow backed strings when ``infer_string`` option is set (:issue:`55638`)
-- Bug in :class:`Series` constructor raising DeprecationWarning when ``index`` is a list of :class:`Series` (:issue:`55228`)
 -
 
 .. ---------------------------------------------------------------------------

From c4d9ba93b627ee1dc0af5044cd7b403bcf45c5a5 Mon Sep 17 00:00:00 2001
From: Patrick Hoefler <61934744+phofl@users.noreply.github.com>
Date: Mon, 20 Nov 2023 01:03:12 +0100
Subject: [PATCH 21/30] Finish

---
 pandas/core/config_init.py                    |  2 +-
 pandas/core/indexes/base.py                   |  2 +-
 .../tests/indexes/base_class/test_formats.py  |  3 ++
 .../tests/indexes/base_class/test_reshape.py  | 10 ++--
 .../tests/indexes/categorical/test_astype.py  |  2 +-
 .../indexes/categorical/test_category.py      |  5 +-
 .../tests/indexes/categorical/test_formats.py |  4 ++
 .../tests/indexes/categorical/test_reindex.py |  2 +-
 .../indexes/datetimes/methods/test_map.py     |  2 +-
 pandas/tests/indexes/interval/test_formats.py |  7 ++-
 .../tests/indexes/multi/test_constructors.py  |  7 ++-
 pandas/tests/indexes/multi/test_get_set.py    | 16 +++---
 pandas/tests/indexes/multi/test_reindex.py    |  7 ++-
 pandas/tests/indexes/multi/test_setops.py     | 20 ++++---
 pandas/tests/indexes/object/test_indexing.py  | 54 +++++++++++++------
 pandas/tests/indexes/test_base.py             | 26 ++++++---
 pandas/tests/indexes/test_old_base.py         |  8 +--
 pandas/tests/indexes/test_setops.py           |  4 +-
 18 files changed, 123 insertions(+), 58 deletions(-)

diff --git a/pandas/core/config_init.py b/pandas/core/config_init.py
index bdbab78a443de..a8b63f97141c2 100644
--- a/pandas/core/config_init.py
+++ b/pandas/core/config_init.py
@@ -905,7 +905,7 @@ def register_converter_cb(key) -> None:
 with cf.config_prefix("future"):
     cf.register_option(
         "infer_string",
-        True,
+        False,
         "Whether to infer sequence of str objects as pyarrow string "
         "dtype, which will be the default in pandas 3.0 "
         "(at which point this option will be deprecated).",
diff --git a/pandas/core/indexes/base.py b/pandas/core/indexes/base.py
index c90a563899a45..5e4fe041fadf0 100644
--- a/pandas/core/indexes/base.py
+++ b/pandas/core/indexes/base.py
@@ -6916,7 +6916,7 @@ def insert(self, loc: int, item) -> Index:
             and is_string_dtype(idx.dtype)
             and new_values.dtype == object
         ):
-            idx = idx.astype(new_values.dtype)
+            idx = Index(new_values, name=self.name, dtype=new_values.dtype)
         return idx
 
     def drop(
diff --git a/pandas/tests/indexes/base_class/test_formats.py b/pandas/tests/indexes/base_class/test_formats.py
index 379aea8826414..f30b578cfcf56 100644
--- a/pandas/tests/indexes/base_class/test_formats.py
+++ b/pandas/tests/indexes/base_class/test_formats.py
@@ -1,6 +1,7 @@
 import numpy as np
 import pytest
 
+from pandas._config import using_pyarrow_string_dtype
 import pandas._config.config as cf
 
 from pandas import Index
@@ -15,6 +16,7 @@ def test_repr_is_valid_construction_code(self):
         res = eval(repr(idx))
         tm.assert_index_equal(res, idx)
 
+    @pytest.mark.xfail(using_pyarrow_string_dtype(), reason="repr different")
     @pytest.mark.parametrize(
         "index,expected",
         [
@@ -79,6 +81,7 @@ def test_string_index_repr(self, index, expected):
         result = repr(index)
         assert result == expected
 
+    @pytest.mark.xfail(using_pyarrow_string_dtype(), reason="repr different")
     @pytest.mark.parametrize(
         "index,expected",
         [
diff --git a/pandas/tests/indexes/base_class/test_reshape.py b/pandas/tests/indexes/base_class/test_reshape.py
index 6586f5f9de480..814a6a516904b 100644
--- a/pandas/tests/indexes/base_class/test_reshape.py
+++ b/pandas/tests/indexes/base_class/test_reshape.py
@@ -33,13 +33,15 @@ def test_insert(self):
 
         # test empty
         null_index = Index([])
-        tm.assert_index_equal(Index(["a"]), null_index.insert(0, "a"))
+        tm.assert_index_equal(Index(["a"], dtype=object), null_index.insert(0, "a"))
 
-    def test_insert_missing(self, nulls_fixture):
+    def test_insert_missing(self, nulls_fixture, using_infer_string):
         # GH#22295
         # test there is no mangling of NA values
-        expected = Index(["a", nulls_fixture, "b", "c"])
-        result = Index(list("abc")).insert(1, nulls_fixture)
+        expected = Index(["a", nulls_fixture, "b", "c"], dtype=object)
+        result = Index(list("abc"), dtype=object).insert(
+            1, Index([nulls_fixture], dtype=object)
+        )
         tm.assert_index_equal(result, expected)
 
     @pytest.mark.parametrize(
diff --git a/pandas/tests/indexes/categorical/test_astype.py b/pandas/tests/indexes/categorical/test_astype.py
index da1d692f9eb2d..a17627b7515b2 100644
--- a/pandas/tests/indexes/categorical/test_astype.py
+++ b/pandas/tests/indexes/categorical/test_astype.py
@@ -18,7 +18,7 @@ def test_astype(self):
         ci = CategoricalIndex(list("aabbca"), categories=list("cab"), ordered=False)
 
         result = ci.astype(object)
-        tm.assert_index_equal(result, Index(np.array(ci)))
+        tm.assert_index_equal(result, Index(np.array(ci), dtype=object))
 
         # this IS equal, but not the same class
         assert result.equals(ci)
diff --git a/pandas/tests/indexes/categorical/test_category.py b/pandas/tests/indexes/categorical/test_category.py
index 7af4f6809ec64..aa9073ed789bb 100644
--- a/pandas/tests/indexes/categorical/test_category.py
+++ b/pandas/tests/indexes/categorical/test_category.py
@@ -1,6 +1,8 @@
 import numpy as np
 import pytest
 
+from pandas._config import using_pyarrow_string_dtype
+
 from pandas._libs import index as libindex
 from pandas._libs.arrays import NDArrayBacked
 
@@ -47,7 +49,7 @@ def test_insert(self, simple_index):
 
         # invalid -> cast to object
         expected = ci.astype(object).insert(0, "d")
-        result = ci.insert(0, "d")
+        result = ci.insert(0, "d").astype(object)
         tm.assert_index_equal(result, expected, exact=True)
 
         # GH 18295 (test missing)
@@ -194,6 +196,7 @@ def test_unique(self, data, categories, expected_data, ordered):
         expected = CategoricalIndex(expected_data, dtype=dtype)
         tm.assert_index_equal(idx.unique(), expected)
 
+    @pytest.mark.xfail(using_pyarrow_string_dtype(), reason="repr doesn't roundtrip")
     def test_repr_roundtrip(self):
         ci = CategoricalIndex(["a", "b"], categories=["a", "b"], ordered=True)
         str(ci)
diff --git a/pandas/tests/indexes/categorical/test_formats.py b/pandas/tests/indexes/categorical/test_formats.py
index ea3e4ce213e67..522ca1bc2afde 100644
--- a/pandas/tests/indexes/categorical/test_formats.py
+++ b/pandas/tests/indexes/categorical/test_formats.py
@@ -1,6 +1,9 @@
 """
 Tests for CategoricalIndex.__repr__ and related methods.
 """
+import pytest
+
+from pandas._config import using_pyarrow_string_dtype
 import pandas._config.config as cf
 
 from pandas import CategoricalIndex
@@ -16,6 +19,7 @@ def test_format_different_scalar_lengths(self):
         with tm.assert_produces_warning(FutureWarning, match=msg):
             assert idx.format() == expected
 
+    @pytest.mark.xfail(using_pyarrow_string_dtype(), reason="repr different")
     def test_string_categorical_index_repr(self):
         # short
         idx = CategoricalIndex(["a", "bb", "ccc"])
diff --git a/pandas/tests/indexes/categorical/test_reindex.py b/pandas/tests/indexes/categorical/test_reindex.py
index 8ca5c6099b4e7..5b1f2b9fb159a 100644
--- a/pandas/tests/indexes/categorical/test_reindex.py
+++ b/pandas/tests/indexes/categorical/test_reindex.py
@@ -40,7 +40,7 @@ def test_reindex_duplicate_target(self):
         # See GH25459
         cat = CategoricalIndex(["a", "b", "c"], categories=["a", "b", "c", "d"])
         res, indexer = cat.reindex(["a", "c", "c"])
-        exp = Index(["a", "c", "c"], dtype="object")
+        exp = Index(["a", "c", "c"])
         tm.assert_index_equal(res, exp, exact=True)
         tm.assert_numpy_array_equal(indexer, np.array([0, 2, 2], dtype=np.intp))
 
diff --git a/pandas/tests/indexes/datetimes/methods/test_map.py b/pandas/tests/indexes/datetimes/methods/test_map.py
index c31e2407190ea..f35f07bd32068 100644
--- a/pandas/tests/indexes/datetimes/methods/test_map.py
+++ b/pandas/tests/indexes/datetimes/methods/test_map.py
@@ -16,7 +16,7 @@ def test_map(self):
 
         f = lambda x: x.strftime("%Y%m%d")
         result = rng.map(f)
-        exp = Index([f(x) for x in rng], dtype="<U8")
+        exp = Index([f(x) for x in rng])
         tm.assert_index_equal(result, exp)
 
     def test_map_fallthrough(self, capsys):
diff --git a/pandas/tests/indexes/interval/test_formats.py b/pandas/tests/indexes/interval/test_formats.py
index bf335d154e186..6b4fae3a295e3 100644
--- a/pandas/tests/indexes/interval/test_formats.py
+++ b/pandas/tests/indexes/interval/test_formats.py
@@ -1,6 +1,8 @@
 import numpy as np
 import pytest
 
+from pandas._config import using_pyarrow_string_dtype
+
 from pandas import (
     DataFrame,
     Index,
@@ -30,13 +32,16 @@ class TestIntervalIndexRendering:
             (DataFrame, ("            0\n(0.0, 1.0]  a\nNaN         b\n(2.0, 3.0]  c")),
         ],
     )
-    def test_repr_missing(self, constructor, expected):
+    def test_repr_missing(self, constructor, expected, using_infer_string, request):
         # GH 25984
+        if using_infer_string and constructor is Series:
+            request.applymarker(pytest.mark.xfail(reason="repr different"))
         index = IntervalIndex.from_tuples([(0, 1), np.nan, (2, 3)])
         obj = constructor(list("abc"), index=index)
         result = repr(obj)
         assert result == expected
 
+    @pytest.mark.xfail(using_pyarrow_string_dtype(), reason="repr different")
     def test_repr_floats(self):
         # GH 32553
 
diff --git a/pandas/tests/indexes/multi/test_constructors.py b/pandas/tests/indexes/multi/test_constructors.py
index 32659bc2f7bff..8456e6a7acba5 100644
--- a/pandas/tests/indexes/multi/test_constructors.py
+++ b/pandas/tests/indexes/multi/test_constructors.py
@@ -847,11 +847,14 @@ def test_multiindex_inference_consistency():
     assert lev.dtype == object
 
 
-def test_dtype_representation():
+def test_dtype_representation(using_infer_string):
     # GH#46900
     pmidx = MultiIndex.from_arrays([[1], ["a"]], names=[("a", "b"), ("c", "d")])
     result = pmidx.dtypes
+    exp = "object" if not using_infer_string else "string"
     expected = Series(
-        ["int64", "object"], index=MultiIndex.from_tuples([("a", "b"), ("c", "d")])
+        ["int64", exp],
+        index=MultiIndex.from_tuples([("a", "b"), ("c", "d")]),
+        dtype=object,
     )
     tm.assert_series_equal(result, expected)
diff --git a/pandas/tests/indexes/multi/test_get_set.py b/pandas/tests/indexes/multi/test_get_set.py
index 0720a1e1c648c..d8ccccc8bb01e 100644
--- a/pandas/tests/indexes/multi/test_get_set.py
+++ b/pandas/tests/indexes/multi/test_get_set.py
@@ -34,23 +34,25 @@ def test_get_level_number_integer(idx):
         idx._get_level_number("fourth")
 
 
-def test_get_dtypes():
+def test_get_dtypes(using_infer_string):
     # Test MultiIndex.dtypes (# Gh37062)
     idx_multitype = MultiIndex.from_product(
         [[1, 2, 3], ["a", "b", "c"], pd.date_range("20200101", periods=2, tz="UTC")],
         names=["int", "string", "dt"],
     )
+
+    exp = "object" if not using_infer_string else "string"
     expected = pd.Series(
         {
             "int": np.dtype("int64"),
-            "string": np.dtype("O"),
+            "string": exp,
             "dt": DatetimeTZDtype(tz="utc"),
         }
     )
     tm.assert_series_equal(expected, idx_multitype.dtypes)
 
 
-def test_get_dtypes_no_level_name():
+def test_get_dtypes_no_level_name(using_infer_string):
     # Test MultiIndex.dtypes (# GH38580 )
     idx_multitype = MultiIndex.from_product(
         [
@@ -59,17 +61,18 @@ def test_get_dtypes_no_level_name():
             pd.date_range("20200101", periods=2, tz="UTC"),
         ],
     )
+    exp = "object" if not using_infer_string else "string"
     expected = pd.Series(
         {
             "level_0": np.dtype("int64"),
-            "level_1": np.dtype("O"),
+            "level_1": exp,
             "level_2": DatetimeTZDtype(tz="utc"),
         }
     )
     tm.assert_series_equal(expected, idx_multitype.dtypes)
 
 
-def test_get_dtypes_duplicate_level_names():
+def test_get_dtypes_duplicate_level_names(using_infer_string):
     # Test MultiIndex.dtypes with non-unique level names (# GH45174)
     result = MultiIndex.from_product(
         [
@@ -79,8 +82,9 @@ def test_get_dtypes_duplicate_level_names():
         ],
         names=["A", "A", "A"],
     ).dtypes
+    exp = "object" if not using_infer_string else "string"
     expected = pd.Series(
-        [np.dtype("int64"), np.dtype("O"), DatetimeTZDtype(tz="utc")],
+        [np.dtype("int64"), exp, DatetimeTZDtype(tz="utc")],
         index=["A", "A", "A"],
     )
     tm.assert_series_equal(result, expected)
diff --git a/pandas/tests/indexes/multi/test_reindex.py b/pandas/tests/indexes/multi/test_reindex.py
index 77a527134b79f..d1b4fe8b98760 100644
--- a/pandas/tests/indexes/multi/test_reindex.py
+++ b/pandas/tests/indexes/multi/test_reindex.py
@@ -75,11 +75,14 @@ def test_reindex_lvl_preserves_names_when_target_is_list_or_array():
     assert idx.reindex([], level=1)[0].names == ["foo", "bar"]
 
 
-def test_reindex_lvl_preserves_type_if_target_is_empty_list_or_array():
+def test_reindex_lvl_preserves_type_if_target_is_empty_list_or_array(
+    using_infer_string,
+):
     # GH7774
     idx = MultiIndex.from_product([[0, 1], ["a", "b"]])
     assert idx.reindex([], level=0)[0].levels[0].dtype.type == np.int64
-    assert idx.reindex([], level=1)[0].levels[1].dtype.type == np.object_
+    exp = np.object_ if not using_infer_string else str
+    assert idx.reindex([], level=1)[0].levels[1].dtype.type == exp
 
     # case with EA levels
     cat = pd.Categorical(["foo", "bar"])
diff --git a/pandas/tests/indexes/multi/test_setops.py b/pandas/tests/indexes/multi/test_setops.py
index 2b4107acee096..23c00f36590ed 100644
--- a/pandas/tests/indexes/multi/test_setops.py
+++ b/pandas/tests/indexes/multi/test_setops.py
@@ -263,19 +263,23 @@ def test_union(idx, sort):
         assert result.equals(idx)
 
 
-def test_union_with_regular_index(idx):
+def test_union_with_regular_index(idx, using_infer_string):
     other = Index(["A", "B", "C"])
 
     result = other.union(idx)
     assert ("foo", "one") in result
     assert "B" in result
 
-    msg = "The values in the array are unorderable"
-    with tm.assert_produces_warning(RuntimeWarning, match=msg):
-        result2 = idx.union(other)
-    # This is more consistent now, if sorting fails then we don't sort at all
-    # in the MultiIndex case.
-    assert not result.equals(result2)
+    if using_infer_string:
+        with pytest.raises(NotImplementedError, match="Can only union"):
+            idx.union(other)
+    else:
+        msg = "The values in the array are unorderable"
+        with tm.assert_produces_warning(RuntimeWarning, match=msg):
+            result2 = idx.union(other)
+        # This is more consistent now, if sorting fails then we don't sort at all
+        # in the MultiIndex case.
+        assert not result.equals(result2)
 
 
 def test_intersection(idx, sort):
@@ -759,4 +763,4 @@ def test_union_with_na_when_constructing_dataframe():
     series2 = Series((10, 20), index=MultiIndex.from_tuples(((None, None), ("a", "b"))))
     result = DataFrame([series1, series2])
     expected = DataFrame({(np.nan, np.nan): [1.0, 10.0], ("a", "b"): [np.nan, 20.0]})
-    tm.assert_frame_equal(result, expected)
+    tm.assert_frame_equal(result, expected, check_column_type=False)
diff --git a/pandas/tests/indexes/object/test_indexing.py b/pandas/tests/indexes/object/test_indexing.py
index 68fc707e5c816..ebf9dac715f8d 100644
--- a/pandas/tests/indexes/object/test_indexing.py
+++ b/pandas/tests/indexes/object/test_indexing.py
@@ -3,9 +3,10 @@
 import numpy as np
 import pytest
 
-from pandas._config import using_pyarrow_string_dtype
-
-from pandas._libs.missing import is_matching_na
+from pandas._libs.missing import (
+    NA,
+    is_matching_na,
+)
 import pandas.util._test_decorators as td
 
 import pandas as pd
@@ -27,20 +28,36 @@ def test_get_indexer_strings(self, method, expected):
 
         tm.assert_numpy_array_equal(actual, expected)
 
-    def test_get_indexer_strings_raises(self):
+    def test_get_indexer_strings_raises(self, using_infer_string):
         index = Index(["b", "c"])
 
-        msg = r"unsupported operand type\(s\) for -: 'str' and 'str'"
-        with pytest.raises(TypeError, match=msg):
-            index.get_indexer(["a", "b", "c", "d"], method="nearest")
+        if using_infer_string:
+            import pyarrow as pa
 
-        with pytest.raises(TypeError, match=msg):
-            index.get_indexer(["a", "b", "c", "d"], method="pad", tolerance=2)
+            msg = "has no kernel"
+            with pytest.raises(pa.lib.ArrowNotImplementedError, match=msg):
+                index.get_indexer(["a", "b", "c", "d"], method="nearest")
 
-        with pytest.raises(TypeError, match=msg):
-            index.get_indexer(
-                ["a", "b", "c", "d"], method="pad", tolerance=[2, 2, 2, 2]
-            )
+            with pytest.raises(pa.lib.ArrowNotImplementedError, match=msg):
+                index.get_indexer(["a", "b", "c", "d"], method="pad", tolerance=2)
+
+            with pytest.raises(pa.lib.ArrowNotImplementedError, match=msg):
+                index.get_indexer(
+                    ["a", "b", "c", "d"], method="pad", tolerance=[2, 2, 2, 2]
+                )
+
+        else:
+            msg = r"unsupported operand type\(s\) for -: 'str' and 'str'"
+            with pytest.raises(TypeError, match=msg):
+                index.get_indexer(["a", "b", "c", "d"], method="nearest")
+
+            with pytest.raises(TypeError, match=msg):
+                index.get_indexer(["a", "b", "c", "d"], method="pad", tolerance=2)
+
+            with pytest.raises(TypeError, match=msg):
+                index.get_indexer(
+                    ["a", "b", "c", "d"], method="pad", tolerance=[2, 2, 2, 2]
+                )
 
     def test_get_indexer_with_NA_values(
         self, unique_nulls_fixture, unique_nulls_fixture2
@@ -53,16 +70,21 @@ def test_get_indexer_with_NA_values(
         arr = np.array([unique_nulls_fixture, unique_nulls_fixture2], dtype=object)
         index = Index(arr, dtype=object)
         result = index.get_indexer(
-            [unique_nulls_fixture, unique_nulls_fixture2, "Unknown"]
+            Index(
+                [unique_nulls_fixture, unique_nulls_fixture2, "Unknown"], dtype=object
+            )
         )
         expected = np.array([0, 1, -1], dtype=np.intp)
         tm.assert_numpy_array_equal(result, expected)
 
 
 class TestGetIndexerNonUnique:
-    @pytest.mark.xfail(using_pyarrow_string_dtype(), reason="NAs are cast to NaN")
-    def test_get_indexer_non_unique_nas(self, nulls_fixture):
+    def test_get_indexer_non_unique_nas(
+        self, nulls_fixture, request, using_infer_string
+    ):
         # even though this isn't non-unique, this should still work
+        if using_infer_string and (nulls_fixture is None or nulls_fixture is NA):
+            request.applymarker(pytest.mark.xfail(reason="NAs are cast to NaN"))
         index = Index(["a", "b", nulls_fixture])
         indexer, missing = index.get_indexer_non_unique([nulls_fixture])
 
diff --git a/pandas/tests/indexes/test_base.py b/pandas/tests/indexes/test_base.py
index c9cf711b429a5..656aaf661e469 100644
--- a/pandas/tests/indexes/test_base.py
+++ b/pandas/tests/indexes/test_base.py
@@ -167,8 +167,10 @@ def test_constructor_from_frame_series_freq(self, using_infer_string):
 
         # GH 6274
         # infer freq of same
-        freq = pd.infer_freq(df["date"])
-        assert freq == "MS"
+        if not using_infer_string:
+            # Doesn't work with arrow strings
+            freq = pd.infer_freq(df["date"])
+            assert freq == "MS"
 
     def test_constructor_int_dtype_nan(self):
         # see gh-15187
@@ -465,7 +467,9 @@ def test_fancy(self, simple_index):
         indirect=True,
     )
     @pytest.mark.parametrize("dtype", [int, np.bool_])
-    def test_empty_fancy(self, index, dtype):
+    def test_empty_fancy(self, index, dtype, request, using_infer_string):
+        if dtype is np.bool_ and using_infer_string and index.dtype == "string":
+            request.applymarker(pytest.mark.xfail(reason="numpy behavior is buggy"))
         empty_arr = np.array([], dtype=dtype)
         empty_index = type(index)([], dtype=index.dtype)
 
@@ -1237,13 +1241,21 @@ def test_equals_op_mismatched_multiindex_raises(self, index):
         with pytest.raises(ValueError, match="Lengths must match"):
             df.index == index
 
-    def test_equals_op_index_vs_mi_same_length(self):
+    def test_equals_op_index_vs_mi_same_length(self, using_infer_string):
         mi = MultiIndex.from_tuples([(1, 2), (4, 5), (8, 9)])
         index = Index(["foo", "bar", "baz"])
+        if using_infer_string:
+            import pyarrow as pa
 
-        result = mi == index
-        expected = np.array([False, False, False])
-        tm.assert_numpy_array_equal(result, expected)
+            with pytest.raises(pa.lib.ArrowNotImplementedError, match="has no kernel"):
+                with tm.assert_produces_warning(
+                    DeprecationWarning, check_stacklevel=False
+                ):
+                    mi == index
+        else:
+            result = mi == index
+            expected = np.array([False, False, False])
+            tm.assert_numpy_array_equal(result, expected)
 
     @pytest.mark.parametrize(
         "dt_conv, arg",
diff --git a/pandas/tests/indexes/test_old_base.py b/pandas/tests/indexes/test_old_base.py
index d625be2a02ac6..8aadb829ffc9b 100644
--- a/pandas/tests/indexes/test_old_base.py
+++ b/pandas/tests/indexes/test_old_base.py
@@ -758,15 +758,15 @@ def test_getitem_2d_deprecated(self, simple_index):
         if isinstance(simple_index, IntervalIndex):
             pytest.skip("Tested elsewhere")
         idx = simple_index
-        msg = "Multi-dimensional indexing"
-        with pytest.raises(ValueError, match=msg):
+        msg = "Multi-dimensional indexing|too many|only"
+        with pytest.raises((ValueError, IndexError), match=msg):
             idx[:, None]
 
         if not isinstance(idx, RangeIndex):
             # GH#44051 RangeIndex already raised pre-2.0 with a different message
-            with pytest.raises(ValueError, match=msg):
+            with pytest.raises((ValueError, IndexError), match=msg):
                 idx[True]
-            with pytest.raises(ValueError, match=msg):
+            with pytest.raises((ValueError, IndexError), match=msg):
                 idx[False]
         else:
             msg = "only integers, slices"
diff --git a/pandas/tests/indexes/test_setops.py b/pandas/tests/indexes/test_setops.py
index d126d32e627cd..8d65d29ea5ecd 100644
--- a/pandas/tests/indexes/test_setops.py
+++ b/pandas/tests/indexes/test_setops.py
@@ -775,10 +775,10 @@ def test_union_identity(self, index, sort):
 
         # This should no longer be the same object, since [] is not consistent,
         # both objects will be recast to dtype('O')
-        union = first.union([], sort=sort)
+        union = first.union(Index([], dtype=first.dtype), sort=sort)
         assert (union is first) is (not sort)
 
-        union = Index([]).union(first, sort=sort)
+        union = Index([], dtype=first.dtype).union(first, sort=sort)
         assert (union is first) is (not sort)
 
     @pytest.mark.parametrize("index", ["string"], indirect=True)

From 06b4f8909fef759bdda8877602b40f4be3b9925e Mon Sep 17 00:00:00 2001
From: Patrick Hoefler <61934744+phofl@users.noreply.github.com>
Date: Tue, 21 Nov 2023 21:17:28 +0100
Subject: [PATCH 22/30] Update

---
 doc/source/whatsnew/v2.1.3.rst | 1 -
 doc/source/whatsnew/v2.1.4.rst | 1 +
 2 files changed, 1 insertion(+), 1 deletion(-)

diff --git a/doc/source/whatsnew/v2.1.3.rst b/doc/source/whatsnew/v2.1.3.rst
index a7f20aa67685f..af626895a9e0e 100644
--- a/doc/source/whatsnew/v2.1.3.rst
+++ b/doc/source/whatsnew/v2.1.3.rst
@@ -21,7 +21,6 @@ Fixed regressions
 Bug fixes
 ~~~~~~~~~
 - Bug in :meth:`DatetimeIndex.diff` raising ``TypeError`` (:issue:`55080`)
-- Bug in :meth:`Index.__getitem__` returning wrong result for Arrow dtypes and negative stepsize (:issue:`55832`)
 - Bug in :meth:`Index.isin` raising for Arrow backed string and ``None`` value (:issue:`55821`)
 - Fix :func:`read_parquet` and :func:`read_feather` for `CVE-2023-47248 <https://www.cve.org/CVERecord?id=CVE-2023-47248>`__ (:issue:`55894`)
 
diff --git a/doc/source/whatsnew/v2.1.4.rst b/doc/source/whatsnew/v2.1.4.rst
index 25afcbb3bb532..e52c42dd31211 100644
--- a/doc/source/whatsnew/v2.1.4.rst
+++ b/doc/source/whatsnew/v2.1.4.rst
@@ -22,6 +22,7 @@ Fixed regressions
 Bug fixes
 ~~~~~~~~~
 - Bug in :class:`Series` constructor raising DeprecationWarning when ``index`` is a list of :class:`Series` (:issue:`55228`)
+- Bug in :meth:`Index.__getitem__` returning wrong result for Arrow dtypes and negative stepsize (:issue:`55832`)
 -
 
 .. ---------------------------------------------------------------------------

From 02adc20e0810b0b4b555d24671cf289000a04f7a Mon Sep 17 00:00:00 2001
From: Patrick Hoefler <61934744+phofl@users.noreply.github.com>
Date: Thu, 30 Nov 2023 21:42:34 +0100
Subject: [PATCH 23/30] Update test_base.py

---
 pandas/tests/indexes/test_base.py | 14 +++-----------
 1 file changed, 3 insertions(+), 11 deletions(-)

diff --git a/pandas/tests/indexes/test_base.py b/pandas/tests/indexes/test_base.py
index 9e009bb6c08cd..eb01f75b35cea 100644
--- a/pandas/tests/indexes/test_base.py
+++ b/pandas/tests/indexes/test_base.py
@@ -1254,18 +1254,10 @@ def test_equals_op_mismatched_multiindex_raises(self, index):
     def test_equals_op_index_vs_mi_same_length(self, using_infer_string):
         mi = MultiIndex.from_tuples([(1, 2), (4, 5), (8, 9)])
         index = Index(["foo", "bar", "baz"])
-        if using_infer_string:
-            import pyarrow as pa
 
-            with pytest.raises(pa.lib.ArrowNotImplementedError, match="has no kernel"):
-                with tm.assert_produces_warning(
-                    DeprecationWarning, check_stacklevel=False
-                ):
-                    mi == index
-        else:
-            result = mi == index
-            expected = np.array([False, False, False])
-            tm.assert_numpy_array_equal(result, expected)
+        result = mi == index
+        expected = np.array([False, False, False])
+        tm.assert_numpy_array_equal(result, expected)
 
     @pytest.mark.parametrize(
         "dt_conv, arg",

From cf283691a0b019d280bc613e24d281cdd241822a Mon Sep 17 00:00:00 2001
From: Patrick Hoefler <61934744+phofl@users.noreply.github.com>
Date: Thu, 30 Nov 2023 21:44:51 +0100
Subject: [PATCH 24/30] Update test_old_base.py

---
 pandas/tests/indexes/test_old_base.py | 4 +++-
 1 file changed, 3 insertions(+), 1 deletion(-)

diff --git a/pandas/tests/indexes/test_old_base.py b/pandas/tests/indexes/test_old_base.py
index 8aadb829ffc9b..1d5c414e70dd7 100644
--- a/pandas/tests/indexes/test_old_base.py
+++ b/pandas/tests/indexes/test_old_base.py
@@ -6,6 +6,8 @@
 import numpy as np
 import pytest
 
+from pandas._config import using_pyarrow_string_dtype
+
 from pandas._libs.tslibs import Timestamp
 
 from pandas.core.dtypes.common import (
@@ -415,7 +417,7 @@ def test_insert_base(self, index):
         # test 0th element
         assert index[0:4].equals(result.insert(0, index[0]))
 
-    @pytest.mark.skip()
+    @pytest.mark.skip(using_pyarrow_string_dtype(), reason="completely different behavior, tested elsewher")
     def test_insert_out_of_bounds(self, index):
         # TypeError/IndexError matches what np.insert raises in these cases
 

From 94cf90855952ef183fc380964658b2f7dc9fd9d6 Mon Sep 17 00:00:00 2001
From: Patrick Hoefler <61934744+phofl@users.noreply.github.com>
Date: Thu, 30 Nov 2023 21:45:29 +0100
Subject: [PATCH 25/30] Update conftest.py

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

diff --git a/pandas/conftest.py b/pandas/conftest.py
index 64d1117b96df7..25a3383c97b84 100644
--- a/pandas/conftest.py
+++ b/pandas/conftest.py
@@ -1926,7 +1926,7 @@ def using_copy_on_write() -> bool:
 @pytest.fixture
 def using_infer_string() -> bool:
     """
-    Fixture to check if Copy-on-Write is enabled.
+    Fixture to check if infer string option is enabled.
     """
     return pd.options.future.infer_string is True
 
@@ -1934,7 +1934,7 @@ def using_infer_string() -> bool:
 @pytest.fixture
 def warn_copy_on_write() -> bool:
     """
-    Fixture to check if Copy-on-Write is enabled.
+    Fixture to check if Copy-on-Write is in warning mode.
     """
     return (
         pd.options.mode.copy_on_write == "warn"

From 671a3535cb095573694759508209e088979db99c Mon Sep 17 00:00:00 2001
From: Patrick Hoefler <61934744+phofl@users.noreply.github.com>
Date: Thu, 30 Nov 2023 21:46:11 +0100
Subject: [PATCH 26/30] Update conftest.py

---
 pandas/conftest.py | 16 ++++++++--------
 1 file changed, 8 insertions(+), 8 deletions(-)

diff --git a/pandas/conftest.py b/pandas/conftest.py
index 25a3383c97b84..798d485a03cca 100644
--- a/pandas/conftest.py
+++ b/pandas/conftest.py
@@ -1923,14 +1923,6 @@ def using_copy_on_write() -> bool:
     )
 
 
-@pytest.fixture
-def using_infer_string() -> bool:
-    """
-    Fixture to check if infer string option is enabled.
-    """
-    return pd.options.future.infer_string is True
-
-
 @pytest.fixture
 def warn_copy_on_write() -> bool:
     """
@@ -1942,6 +1934,14 @@ def warn_copy_on_write() -> bool:
     )
 
 
+@pytest.fixture
+def using_infer_string() -> bool:
+    """
+    Fixture to check if infer string option is enabled.
+    """
+    return pd.options.future.infer_string is True
+
+
 warsaws = ["Europe/Warsaw", "dateutil/Europe/Warsaw"]
 if zoneinfo is not None:
     warsaws.append(zoneinfo.ZoneInfo("Europe/Warsaw"))  # type: ignore[arg-type]

From 77f4711516d10905deed064daa8a7f622287b7c7 Mon Sep 17 00:00:00 2001
From: Patrick Hoefler <61934744+phofl@users.noreply.github.com>
Date: Thu, 30 Nov 2023 22:25:29 +0100
Subject: [PATCH 27/30] Update test_old_base.py

---
 pandas/tests/indexes/test_old_base.py | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/pandas/tests/indexes/test_old_base.py b/pandas/tests/indexes/test_old_base.py
index 1d5c414e70dd7..e4fae1119e121 100644
--- a/pandas/tests/indexes/test_old_base.py
+++ b/pandas/tests/indexes/test_old_base.py
@@ -417,7 +417,7 @@ def test_insert_base(self, index):
         # test 0th element
         assert index[0:4].equals(result.insert(0, index[0]))
 
-    @pytest.mark.skip(using_pyarrow_string_dtype(), reason="completely different behavior, tested elsewher")
+    @pytest.mark.skipif(using_pyarrow_string_dtype(), reason="completely different behavior, tested elsewher")
     def test_insert_out_of_bounds(self, index):
         # TypeError/IndexError matches what np.insert raises in these cases
 

From aa1a6cf37b5d84f68fdfe4338cc4d5566dd7997c Mon Sep 17 00:00:00 2001
From: Patrick Hoefler <61934744+phofl@users.noreply.github.com>
Date: Thu, 30 Nov 2023 22:28:55 +0100
Subject: [PATCH 28/30] Update

---
 pandas/tests/indexes/multi/test_setops.py | 10 ++++++++--
 1 file changed, 8 insertions(+), 2 deletions(-)

diff --git a/pandas/tests/indexes/multi/test_setops.py b/pandas/tests/indexes/multi/test_setops.py
index a447e6dbc4c9c..5238ca5cca1a3 100644
--- a/pandas/tests/indexes/multi/test_setops.py
+++ b/pandas/tests/indexes/multi/test_setops.py
@@ -760,8 +760,14 @@ def test_intersection_keep_ea_dtypes(val, any_numeric_ea_dtype):
 
 def test_union_with_na_when_constructing_dataframe():
     # GH43222
-    series1 = Series((1,), index=MultiIndex.from_tuples(((None, None),)))
+    pd.options.future.infer_string = True
+    series1 = Series(
+        (1,),
+        index=MultiIndex.from_arrays(
+            [Series([None], dtype="string"), Series([None], dtype="string")]
+        ),
+    )
     series2 = Series((10, 20), index=MultiIndex.from_tuples(((None, None), ("a", "b"))))
     result = DataFrame([series1, series2])
     expected = DataFrame({(np.nan, np.nan): [1.0, 10.0], ("a", "b"): [np.nan, 20.0]})
-    tm.assert_frame_equal(result, expected, check_column_type=False)
+    tm.assert_frame_equal(result, expected)

From bd5e3052d5b05cb005d0100af7777ddd68fe9e76 Mon Sep 17 00:00:00 2001
From: Patrick Hoefler <61934744+phofl@users.noreply.github.com>
Date: Thu, 30 Nov 2023 23:05:55 +0100
Subject: [PATCH 29/30] Update test_setops.py

---
 pandas/tests/indexes/multi/test_setops.py | 1 -
 1 file changed, 1 deletion(-)

diff --git a/pandas/tests/indexes/multi/test_setops.py b/pandas/tests/indexes/multi/test_setops.py
index 5238ca5cca1a3..0abb56ecf9de7 100644
--- a/pandas/tests/indexes/multi/test_setops.py
+++ b/pandas/tests/indexes/multi/test_setops.py
@@ -760,7 +760,6 @@ def test_intersection_keep_ea_dtypes(val, any_numeric_ea_dtype):
 
 def test_union_with_na_when_constructing_dataframe():
     # GH43222
-    pd.options.future.infer_string = True
     series1 = Series(
         (1,),
         index=MultiIndex.from_arrays(

From 746b50432915a3bac889e87585e839169cf3cc5c Mon Sep 17 00:00:00 2001
From: Patrick Hoefler <patrick_hoefler@gmx.net>
Date: Fri, 8 Dec 2023 22:48:51 +0100
Subject: [PATCH 30/30] Fix pre-commit

---
 pandas/tests/indexes/test_old_base.py | 5 ++++-
 1 file changed, 4 insertions(+), 1 deletion(-)

diff --git a/pandas/tests/indexes/test_old_base.py b/pandas/tests/indexes/test_old_base.py
index e4fae1119e121..528ef2e19aaec 100644
--- a/pandas/tests/indexes/test_old_base.py
+++ b/pandas/tests/indexes/test_old_base.py
@@ -417,7 +417,10 @@ def test_insert_base(self, index):
         # test 0th element
         assert index[0:4].equals(result.insert(0, index[0]))
 
-    @pytest.mark.skipif(using_pyarrow_string_dtype(), reason="completely different behavior, tested elsewher")
+    @pytest.mark.skipif(
+        using_pyarrow_string_dtype(),
+        reason="completely different behavior, tested elsewher",
+    )
     def test_insert_out_of_bounds(self, index):
         # TypeError/IndexError matches what np.insert raises in these cases