Coerce pd.StringDtype to np.dtypes.StringDType

doc/whats-new.rst

@@ -71,6 +71,9 @@ Bug Fixes
 - Ensure the :py:class:`~xarray.groupers.SeasonResampler` preserves the datetime
   unit of the underlying time index when resampling (:issue:`11048`,
   :pull:`11049`). By `Spencer Clark <https://github.com/spencerkclark>`_.
+- Partially support pandas 3 default string indexes by coercing ``pd.StringDtype``
+  to ``np.dtypes.StringDType`` in ``PandasIndexingAdapter`` (:issue:`11098`, :pull:`11102`).
+  By `Julia Signell <https://github.com/jsignell>`_.
 
 Documentation
 ~~~~~~~~~~~~~

xarray/core/indexing.py

@@ -16,6 +16,7 @@
 from numpy.typing import DTypeLike
 from packaging.version import Version
 
+from xarray.compat.npcompat import HAS_STRING_DTYPE
 from xarray.core import duck_array_ops
 from xarray.core.coordinate_transform import CoordinateTransform
 from xarray.core.nputils import NumpyVIndexAdapter
@@ -1916,6 +1917,8 @@ def __init__(
                 self._dtype = get_valid_numpy_dtype(array)
         elif is_allowed_extension_array_dtype(dtype):
             self._dtype = cast(pd.api.extensions.ExtensionDtype, dtype)
+        elif HAS_STRING_DTYPE and isinstance(dtype, pd.StringDtype):
+            self._dtype = np.dtypes.StringDType(na_object=dtype.na_value)
         else:
             self._dtype = np.dtype(cast(DTypeLike, dtype))
 

xarray/tests/__init__.py

@@ -172,7 +172,7 @@ def _importorskip(
 has_flox, requires_flox = _importorskip("flox")
 has_netcdf, requires_netcdf = _importorskip("netcdf")
 has_pandas_ge_2_2, requires_pandas_ge_2_2 = _importorskip("pandas", "2.2")
-has_pandas_3, requires_pandas_3 = _importorskip("pandas", "3.0.0.dev0")
+has_pandas_3, requires_pandas_3 = _importorskip("pandas", "3.0.0")
 
 
 # some special cases

xarray/tests/test_indexing.py

@@ -7,7 +7,7 @@
 import pandas as pd
 import pytest
 
-from xarray import DataArray, Dataset, Variable
+from xarray import DataArray, Dataset, Variable, concat
 from xarray.core import indexing, nputils
 from xarray.core.indexes import PandasIndex, PandasMultiIndex
 from xarray.core.types import T_Xarray
@@ -18,6 +18,7 @@
     assert_identical,
     raise_if_dask_computes,
     requires_dask,
+    requires_pandas_3,
 )
 from xarray.tests.arrays import DuckArrayWrapper
 
@@ -1192,3 +1193,10 @@ def test_backend_indexing_non_numpy() -> None:
         raw_indexing_method=array.__getitem__,
     )
     np.testing.assert_array_equal(indexed.array, np.array([1]))
+
+
+@requires_pandas_3
+def test_pandas_StringDtype_index_coerces_to_numpy() -> None:
+    da = DataArray([0, 1], coords={"x": ["x1", "x2"]})
+    actual = concat([da, da], dim=pd.Index(["y1", "y2"], name="y"))
+    assert isinstance(actual["y"].dtype, np.dtypes.StringDType)