ENH/TST: Add BaseInterfaceTests tests for ArrowExtensionArray

pandas/core/arrays/arrow/array.py

@@ -31,6 +31,7 @@
 )
 from pandas.core.dtypes.missing import isna
 
+from pandas.core.arraylike import OpsMixin
 from pandas.core.arrays.base import ExtensionArray
 from pandas.core.indexers import (
     check_array_indexer,
@@ -45,13 +46,22 @@
     from pandas.core.arrays.arrow._arrow_utils import fallback_performancewarning
     from pandas.core.arrays.arrow.dtype import ArrowDtype
 
+    ARROW_CMP_FUNCS = {
+        "eq": pc.equal,
+        "ne": pc.not_equal,
+        "lt": pc.less,
+        "gt": pc.greater,
+        "le": pc.less_equal,
+        "ge": pc.greater_equal,
+    }
+
 if TYPE_CHECKING:
     from pandas import Series
 
 ArrowExtensionArrayT = TypeVar("ArrowExtensionArrayT", bound="ArrowExtensionArray")
 
 
-class ArrowExtensionArray(ExtensionArray):
+class ArrowExtensionArray(OpsMixin, ExtensionArray):
     """
     Base class for ExtensionArray backed by Arrow ChunkedArray.
     """
@@ -179,6 +189,34 @@ def __arrow_array__(self, type=None):
         """Convert myself to a pyarrow ChunkedArray."""
         return self._data
 
+    def _cmp_method(self, other, op):
+        from pandas.arrays import BooleanArray
+
+        pc_func = ARROW_CMP_FUNCS[op.__name__]
+        if isinstance(other, ArrowExtensionArray):
+            result = pc_func(self._data, other._data)
+        elif isinstance(other, (np.ndarray, list)):
+            result = pc_func(self._data, other)
+        elif is_scalar(other):
+            try:
+                result = pc_func(self._data, pa.scalar(other))
+            except (pa.lib.ArrowNotImplementedError, pa.lib.ArrowInvalid):
+                mask = isna(self) | isna(other)
+                valid = ~mask
+                result = np.zeros(len(self), dtype="bool")
+                result[valid] = op(np.array(self)[valid], other)
+                return BooleanArray(result, mask)
+        else:
+            return NotImplementedError(
+                f"{op.__name__} not implemented for {type(other)}"
+            )
+
+        if pa_version_under2p0:
+            result = result.to_pandas().values
+        else:
+            result = result.to_numpy()
+        return BooleanArray._from_sequence(result)
+
     def equals(self, other) -> bool:
         if not isinstance(other, ArrowExtensionArray):
             return False

pandas/core/arrays/string_arrow.py

@@ -34,7 +34,6 @@
 )
 from pandas.core.dtypes.missing import isna
 
-from pandas.core.arraylike import OpsMixin
 from pandas.core.arrays.arrow import ArrowExtensionArray
 from pandas.core.arrays.boolean import BooleanDtype
 from pandas.core.arrays.integer import Int64Dtype
@@ -51,15 +50,6 @@
 
     from pandas.core.arrays.arrow._arrow_utils import fallback_performancewarning
 
-    ARROW_CMP_FUNCS = {
-        "eq": pc.equal,
-        "ne": pc.not_equal,
-        "lt": pc.less,
-        "gt": pc.greater,
-        "le": pc.less_equal,
-        "ge": pc.greater_equal,
-    }
-
 ArrowStringScalarOrNAT = Union[str, libmissing.NAType]
 
 
@@ -74,9 +64,7 @@ def _chk_pyarrow_available() -> None:
 # fallback for the ones that pyarrow doesn't yet support
 
 
-class ArrowStringArray(
-    OpsMixin, ArrowExtensionArray, BaseStringArray, ObjectStringArrayMixin
-):
+class ArrowStringArray(ArrowExtensionArray, BaseStringArray, ObjectStringArrayMixin):
     """
     Extension array for string data in a ``pyarrow.ChunkedArray``.
 
@@ -190,32 +178,6 @@ def to_numpy(
             result[mask] = na_value
         return result
 
-    def _cmp_method(self, other, op):
-        from pandas.arrays import BooleanArray
-
-        pc_func = ARROW_CMP_FUNCS[op.__name__]
-        if isinstance(other, ArrowStringArray):
-            result = pc_func(self._data, other._data)
-        elif isinstance(other, (np.ndarray, list)):
-            result = pc_func(self._data, other)
-        elif is_scalar(other):
-            try:
-                result = pc_func(self._data, pa.scalar(other))
-            except (pa.lib.ArrowNotImplementedError, pa.lib.ArrowInvalid):
-                mask = isna(self) | isna(other)
-                valid = ~mask
-                result = np.zeros(len(self), dtype="bool")
-                result[valid] = op(np.array(self)[valid], other)
-                return BooleanArray(result, mask)
-        else:
-            return NotImplemented
-
-        if pa_version_under2p0:
-            result = result.to_pandas().values
-        else:
-            result = result.to_numpy()
-        return BooleanArray._from_sequence(result)
-
     def insert(self, loc: int, item):
         if not isinstance(item, str) and item is not libmissing.NA:
             raise TypeError("Scalar must be NA or str")

pandas/tests/extension/arrow/arrays.py

@@ -23,7 +23,6 @@
     take,
 )
 from pandas.api.types import is_scalar
-from pandas.core.arraylike import OpsMixin
 from pandas.core.arrays.arrow import ArrowExtensionArray as _ArrowExtensionArray
 from pandas.core.construction import extract_array
 
@@ -72,7 +71,7 @@ def construct_array_type(cls) -> type_t[ArrowStringArray]:
         return ArrowStringArray
 
 
-class ArrowExtensionArray(OpsMixin, _ArrowExtensionArray):
+class ArrowExtensionArray(_ArrowExtensionArray):
     _data: pa.ChunkedArray
 
     @classmethod

pandas/tests/extension/test_arrow.py

@@ -34,7 +34,7 @@
 from pandas.core.arrays.arrow.dtype import ArrowDtype  # isort:skip
 
 
-@pytest.fixture(params=tm.ALL_PYARROW_DTYPES)
+@pytest.fixture(params=tm.ALL_PYARROW_DTYPES, ids=str)
 def dtype(request):
     return ArrowDtype(pyarrow_dtype=request.param)
 
@@ -201,6 +201,26 @@ class TestBaseIndex(base.BaseIndexTests):
     pass
 
 
+class TestBaseInterface(base.BaseInterfaceTests):
+    def test_contains(self, data, data_missing, request):
+        tz = getattr(data.dtype.pyarrow_dtype, "tz", None)
+        unit = getattr(data.dtype.pyarrow_dtype, "unit", None)
+        if pa_version_under2p0 and tz not in (None, "UTC") and unit == "us":
+            request.node.add_marker(
+                pytest.mark.xfail(
+                    reason=(
+                        f"Not supported by pyarrow < 2.0 "
+                        f"with timestamp type {tz} and {unit}"
+                    )
+                )
+            )
+        super().test_contains(data, data_missing)
+
+    @pytest.mark.xfail(reason="pyarrow.ChunkedArray does not support views.")
+    def test_view(self, data):
+        super().test_view(data)
+
+
 def test_arrowdtype_construct_from_string_type_with_parameters():
     with pytest.raises(NotImplementedError, match="Passing pyarrow type"):
         ArrowDtype.construct_from_string("timestamp[s][pyarrow]")