ENH: Allow ArrowDtype(pa.string()) to be compatable with str methods

doc/source/whatsnew/v2.0.0.rst

@@ -185,6 +185,7 @@ Other enhancements
 - Added :meth:`Index.infer_objects` analogous to :meth:`Series.infer_objects` (:issue:`50034`)
 - Added ``copy`` parameter to :meth:`Series.infer_objects` and :meth:`DataFrame.infer_objects`, passing ``False`` will avoid making copies for series or columns that are already non-object or where no better dtype can be inferred (:issue:`50096`)
 - :meth:`DataFrame.plot.hist` now recognizes ``xlabel`` and ``ylabel`` arguments (:issue:`49793`)
+- Added support for ``str`` accessor methods when using ``pd.ArrowDtype(pyarrow.string())`` (:issue:`50325`)
 - :meth:`Series.drop_duplicates` has gained ``ignore_index`` keyword to reset index (:issue:`48304`)
 - :meth:`Series.dropna` and :meth:`DataFrame.dropna` has gained ``ignore_index`` keyword to reset index (:issue:`31725`)
 - Improved error message in :func:`to_datetime` for non-ISO8601 formats, informing users about the position of the first error (:issue:`50361`)

pandas/conftest.py

@@ -70,6 +70,7 @@
     Index,
     MultiIndex,
 )
+from pandas.core.strings.accessor import StringMethods
 
 try:
     import pyarrow as pa
@@ -1947,3 +1948,124 @@ def warsaw(request):
     tzinfo for Europe/Warsaw using pytz, dateutil, or zoneinfo.
     """
     return request.param
+
+
+_any_string_method = [
+    ("cat", (), {"sep": ","}),
+    ("cat", (Series(list("zyx")),), {"sep": ",", "join": "left"}),
+    ("center", (10,), {}),
+    ("contains", ("a",), {}),
+    ("count", ("a",), {}),
+    ("decode", ("UTF-8",), {}),
+    ("encode", ("UTF-8",), {}),
+    ("endswith", ("a",), {}),
+    ("endswith", ("a",), {"na": True}),
+    ("endswith", ("a",), {"na": False}),
+    ("extract", ("([a-z]*)",), {"expand": False}),
+    ("extract", ("([a-z]*)",), {"expand": True}),
+    ("extractall", ("([a-z]*)",), {}),
+    ("find", ("a",), {}),
+    ("findall", ("a",), {}),
+    ("get", (0,), {}),
+    # because "index" (and "rindex") fail intentionally
+    # if the string is not found, search only for empty string
+    ("index", ("",), {}),
+    ("join", (",",), {}),
+    ("ljust", (10,), {}),
+    ("match", ("a",), {}),
+    ("fullmatch", ("a",), {}),
+    ("normalize", ("NFC",), {}),
+    ("pad", (10,), {}),
+    ("partition", (" ",), {"expand": False}),
+    ("partition", (" ",), {"expand": True}),
+    ("repeat", (3,), {}),
+    ("replace", ("a", "z"), {}),
+    ("rfind", ("a",), {}),
+    ("rindex", ("",), {}),
+    ("rjust", (10,), {}),
+    ("rpartition", (" ",), {"expand": False}),
+    ("rpartition", (" ",), {"expand": True}),
+    ("slice", (0, 1), {}),
+    ("slice_replace", (0, 1, "z"), {}),
+    ("split", (" ",), {"expand": False}),
+    # ("split", (" ",), {"expand": True}),
+    ("startswith", ("a",), {}),
+    ("startswith", ("a",), {"na": True}),
+    ("startswith", ("a",), {"na": False}),
+    ("removeprefix", ("a",), {}),
+    ("removesuffix", ("a",), {}),
+    # translating unicode points of "a" to "d"
+    ("translate", ({97: 100},), {}),
+    ("wrap", (2,), {}),
+    ("zfill", (10,), {}),
+] + list(
+    zip(
+        [
+            # methods without positional arguments: zip with empty tuple and empty dict
+            "capitalize",
+            "cat",
+            "get_dummies",
+            "isalnum",
+            "isalpha",
+            "isdecimal",
+            "isdigit",
+            "islower",
+            "isnumeric",
+            "isspace",
+            "istitle",
+            "isupper",
+            "len",
+            "lower",
+            "lstrip",
+            "partition",
+            "rpartition",
+            "rsplit",
+            "rstrip",
+            "slice",
+            "slice_replace",
+            "split",
+            "strip",
+            "swapcase",
+            "title",
+            "upper",
+            "casefold",
+        ],
+        [()] * 100,
+        [{}] * 100,
+    )
+)
+ids, _, _ = zip(*_any_string_method)  # use method name as fixture-id
+missing_methods = {f for f in dir(StringMethods) if not f.startswith("_")} - set(ids)
+
+# test that the above list captures all methods of StringMethods
+assert not missing_methods
+
+
+@pytest.fixture(params=_any_string_method, ids=ids)
+def any_string_method(request):
+    """
+    Fixture for all public methods of `StringMethods`
+
+    This fixture returns a tuple of the method name and sample arguments
+    necessary to call the method.
+
+    Returns
+    -------
+    method_name : str
+        The name of the method in `StringMethods`
+    args : tuple
+        Sample values for the positional arguments
+    kwargs : dict
+        Sample values for the keyword arguments
+
+    Examples
+    --------
+    >>> def test_something(any_string_method):
+    ...     s = Series(['a', 'b', np.nan, 'd'])
+    ...
+    ...     method_name, args, kwargs = any_string_method
+    ...     method = getattr(s.str, method_name)
+    ...     # will not raise
+    ...     method(*args, **kwargs)
+    """
+    return request.param