disallow pyarrow_numpy as option + fix more cases of checking storage to be pyarrow_numpy

Author: jorisvandenbossche

pandas/core/arrays/string_arrow.py

@@ -188,10 +188,7 @@ def _from_sequence(
 
         if dtype and not (isinstance(dtype, str) and dtype == "string"):
             dtype = pandas_dtype(dtype)
-            assert isinstance(dtype, StringDtype) and dtype.storage in (
-                "pyarrow",
-                "pyarrow_numpy",
-            )
+            assert isinstance(dtype, StringDtype) and dtype.storage == "pyarrow"
 
         if isinstance(scalars, BaseMaskedArray):
             # avoid costly conversion to object dtype in ensure_string_array and

pandas/core/config_init.py

@@ -460,7 +460,7 @@ def is_terminal() -> bool:
         "string_storage",
         "python",
         string_storage_doc,
-        validator=is_one_of_factory(["python", "pyarrow", "pyarrow_numpy"]),
+        validator=is_one_of_factory(["python", "pyarrow"]),
     )
 
 

pandas/tests/arrays/string_/test_string.py

@@ -483,7 +483,7 @@ def test_arrow_array(dtype):
     data = pd.array(["a", "b", "c"], dtype=dtype)
     arr = pa.array(data)
     expected = pa.array(list(data), type=pa.large_string(), from_pandas=True)
-    if dtype.storage in ("pyarrow", "pyarrow_numpy") and pa_version_under12p0:
+    if dtype.storage == "pyarrow" and pa_version_under12p0:
         expected = pa.chunked_array(expected)
     if dtype.storage == "python":
         expected = pc.cast(expected, pa.string())
@@ -501,6 +501,10 @@ def test_arrow_roundtrip(dtype, string_storage2, request, using_infer_string):
                 reason="infer_string takes precedence over string storage"
             )
         )
+    if string_storage2 == "pyarrow_numpy":
+        # we cannot set "pyarrow_numpy" as storage option anymore, need to
+        # update the tests for this
+        request.applymarker(pytest.mark.xfail(reason="TODO(infer_string)"))
 
     data = pd.array(["a", "b", None], dtype=dtype)
     df = pd.DataFrame({"a": data})
@@ -531,6 +535,8 @@ def test_arrow_load_from_zero_chunks(
                 reason="infer_string takes precedence over string storage"
             )
         )
+    if string_storage2 == "pyarrow_numpy":
+        request.applymarker(pytest.mark.xfail(reason="TODO(infer_string)"))
 
     data = pd.array([], dtype=dtype)
     df = pd.DataFrame({"a": data})

pandas/tests/arrays/string_/test_string_arrow.py

@@ -29,6 +29,8 @@ def test_eq_all_na():
 def test_config(string_storage, request, using_infer_string):
     if using_infer_string and string_storage != "pyarrow_numpy":
         request.applymarker(pytest.mark.xfail(reason="infer string takes precedence"))
+    if string_storage == "pyarrow_numpy":
+        request.applymarker(pytest.mark.xfail(reason="TODO(infer_string)"))
     with pd.option_context("string_storage", string_storage):
         assert StringDtype().storage == string_storage
         result = pd.array(["a", "b"])

pandas/tests/extension/test_string.py

@@ -140,28 +140,21 @@ def _get_expected_exception(
         self, op_name: str, obj, other
     ) -> type[Exception] | None:
         if op_name in ["__divmod__", "__rdivmod__"]:
-            if isinstance(obj, pd.Series) and cast(
-                StringDtype, tm.get_dtype(obj)
-            ).storage in [
-                "pyarrow",
-                "pyarrow_numpy",
-            ]:
+            if (
+                isinstance(obj, pd.Series)
+                and cast(StringDtype, tm.get_dtype(obj)).storage == "pyarrow"
+            ):
                 # TODO: re-raise as TypeError?
                 return NotImplementedError
-            elif isinstance(other, pd.Series) and cast(
-                StringDtype, tm.get_dtype(other)
-            ).storage in [
-                "pyarrow",
-                "pyarrow_numpy",
-            ]:
+            elif (
+                isinstance(other, pd.Series)
+                and cast(StringDtype, tm.get_dtype(other)).storage == "pyarrow"
+            ):
                 # TODO: re-raise as TypeError?
                 return NotImplementedError
             return TypeError
         elif op_name in ["__mod__", "__rmod__", "__pow__", "__rpow__"]:
-            if cast(StringDtype, tm.get_dtype(obj)).storage in [
-                "pyarrow",
-                "pyarrow_numpy",
-            ]:
+            if cast(StringDtype, tm.get_dtype(obj)).storage == "pyarrow":
                 return NotImplementedError
             return TypeError
         elif op_name in ["__mul__", "__rmul__"]:
@@ -175,10 +168,7 @@ def _get_expected_exception(
             "__sub__",
             "__rsub__",
         ]:
-            if cast(StringDtype, tm.get_dtype(obj)).storage in [
-                "pyarrow",
-                "pyarrow_numpy",
-            ]:
+            if cast(StringDtype, tm.get_dtype(obj)).storage == "pyarrow":
                 import pyarrow as pa
 
                 # TODO: better to re-raise as TypeError?

pandas/tests/frame/methods/test_convert_dtypes.py

@@ -18,13 +18,17 @@ def test_convert_dtypes(
         # Just check that it works for DataFrame here
         if using_infer_string:
             string_storage = "pyarrow_numpy"
+
+        string_storage_option = string_storage
+        if string_storage == "pyarrow_numpy":
+            string_storage_option = "pyarrow"
         df = pd.DataFrame(
             {
                 "a": pd.Series([1, 2, 3], dtype=np.dtype("int32")),
                 "b": pd.Series(["x", "y", "z"], dtype=np.dtype("O")),
             }
         )
-        with pd.option_context("string_storage", string_storage):
+        with pd.option_context("string_storage", string_storage_option):
             result = df.convert_dtypes(True, True, convert_integer, False)
         expected = pd.DataFrame(
             {