API: str.center with pyarrow-backed string dtype

Author: jbrockmendel

doc/source/whatsnew/v3.0.0.rst

@@ -277,6 +277,7 @@ Other API changes
 - Made ``dtype`` a required argument in :meth:`ExtensionArray._from_sequence_of_strings` (:issue:`56519`)
 - Passing a :class:`Series` input to :func:`json_normalize` will now retain the :class:`Series` :class:`Index`, previously output had a new :class:`RangeIndex` (:issue:`51452`)
 - Removed :meth:`Index.sort` which always raised a ``TypeError``. This attribute is not defined and will raise an ``AttributeError`` (:issue:`59283`)
+- The ``center`` method on :class:`Series` and :class:`Index` object ``str`` accessors with pyarrow-backed dtype now matches the python behavior in corner cases with an odd number of fill characters when using pyarrow versions 17.0 and above (:issue:`54792`)
 - Updated :meth:`DataFrame.to_excel` so that the output spreadsheet has no styling. Custom styling can still be done using :meth:`Styler.to_excel` (:issue:`54154`)
 - pickle and HDF (``.h5``) files created with Python 2 are no longer explicitly supported (:issue:`57387`)
 - pickled objects from pandas version less than ``1.0.0`` are no longer supported (:issue:`57155`)

pandas/core/arrays/_arrow_string_mixins.py

@@ -1,13 +1,17 @@
 from __future__ import annotations
 
+from functools import partial
 from typing import (
     TYPE_CHECKING,
     Literal,
 )
 
 import numpy as np
 
-from pandas.compat import pa_version_under10p1
+from pandas.compat import (
+    pa_version_under10p1,
+    pa_version_under17p0,
+)
 
 from pandas.core.dtypes.missing import isna
 
@@ -49,7 +53,11 @@ def _str_pad(
         elif side == "right":
             pa_pad = pc.utf8_rpad
         elif side == "both":
-            pa_pad = pc.utf8_center
+            if pa_version_under17p0:
+                pa_pad = pc.utf8_center
+            else:
+                # GH#54792
+                pa_pad = partial(pc.utf8_center, lean_left_on_odd_padding=False)
         else:
             raise ValueError(
                 f"Invalid side: {side}. Side must be one of 'left', 'right', 'both'"

pandas/core/arrays/string_arrow.py

@@ -282,6 +282,7 @@ def astype(self, dtype, copy: bool = True):
     _str_map = BaseStringArray._str_map
     _str_startswith = ArrowStringArrayMixin._str_startswith
     _str_endswith = ArrowStringArrayMixin._str_endswith
+    _str_pad = ArrowStringArrayMixin._str_pad
 
     def _str_contains(
         self, pat, case: bool = True, flags: int = 0, na=np.nan, regex: bool = True
@@ -536,7 +537,6 @@ class ArrowStringArrayNumpySemantics(ArrowStringArray):
     _str_get = ArrowStringArrayMixin._str_get
     _str_removesuffix = ArrowStringArrayMixin._str_removesuffix
     _str_capitalize = ArrowStringArrayMixin._str_capitalize
-    _str_pad = ArrowStringArrayMixin._str_pad
     _str_title = ArrowStringArrayMixin._str_title
     _str_swapcase = ArrowStringArrayMixin._str_swapcase
     _str_slice_replace = ArrowStringArrayMixin._str_slice_replace

pandas/tests/strings/test_case_justify.py

@@ -291,11 +291,7 @@ def test_center_ljust_rjust_mixed_object():
 
 
 def test_center_ljust_rjust_fillchar(any_string_dtype):
-    if any_string_dtype == "string[pyarrow_numpy]":
-        pytest.skip(
-            "Arrow logic is different, "
-            "see https://github.com/pandas-dev/pandas/pull/54533/files#r1299808126",
-        )
+    # GH#54533, GH#54792
     s = Series(["a", "bb", "cccc", "ddddd", "eeeeee"], dtype=any_string_dtype)
 
     result = s.str.center(5, fillchar="X")