BUG: Fix regression in to_hdf with MultiIndex and StringDtype (GH#63412) (#63414)

Author: zacharym-collins

pandas/io/pytables.py

@@ -3147,7 +3147,9 @@ def write_multi_index(self, key: str, index: MultiIndex) -> None:
             zip(index.levels, index.codes, index.names, strict=True)
         ):
             # write the level
-            if isinstance(lev.dtype, ExtensionDtype):
+            if isinstance(lev.dtype, ExtensionDtype) and not isinstance(
+                lev.dtype, StringDtype
+            ):
                 raise NotImplementedError(
                     "Saving a MultiIndex with an extension dtype is not supported."
                 )

pandas/tests/io/pytables/test_put.py

@@ -314,12 +314,6 @@ def test_column_multiindex(tmp_path, setup_path, using_infer_string):
 
     path = tmp_path / setup_path
     with HDFStore(path) as store:
-        if using_infer_string:
-            # TODO(infer_string) make this work for string dtype
-            msg = "Saving a MultiIndex with an extension dtype is not supported."
-            with pytest.raises(NotImplementedError, match=msg):
-                store.put("df", df)
-            return
         store.put("df", df)
         tm.assert_frame_equal(
             store["df"], expected, check_index_type=True, check_column_type=True

pandas/tests/io/pytables/test_read.py

@@ -181,12 +181,6 @@ def test_read_hdf_open_store(tmp_path, setup_path, using_infer_string):
     df = df.set_index(keys="E", append=True)
 
     path = tmp_path / setup_path
-    if using_infer_string:
-        # TODO(infer_string) make this work for string dtype
-        msg = "Saving a MultiIndex with an extension dtype is not supported."
-        with pytest.raises(NotImplementedError, match=msg):
-            df.to_hdf(path, key="df", mode="w")
-        return
     df.to_hdf(path, key="df", mode="w")
     direct = read_hdf(path, "df")
     with HDFStore(path, mode="r") as store:

pandas/tests/io/pytables/test_round_trip.py

@@ -434,12 +434,6 @@ def test_store_hierarchical(
 ):
     frame = multiindex_dataframe_random_data
 
-    if using_infer_string:
-        # TODO(infer_string) make this work for string dtype
-        msg = "Saving a MultiIndex with an extension dtype is not supported."
-        with pytest.raises(NotImplementedError, match=msg):
-            _check_roundtrip(frame, tm.assert_frame_equal, path=temp_file)
-        return
     _check_roundtrip(frame, tm.assert_frame_equal, path=temp_file)
     _check_roundtrip(frame.T, tm.assert_frame_equal, path=temp_file)
     _check_roundtrip(frame["A"], tm.assert_series_equal, path=temp_file)

pandas/tests/io/pytables/test_store.py

@@ -1129,3 +1129,13 @@ def test_select_categorical_string_columns(tmp_path, model):
         result = store.select("df", "modelId == model")
         expected = df[df["modelId"] == model]
         tm.assert_frame_equal(result, expected)
+
+
+def test_to_hdf_multiindex_string_dtype_crash(tmp_path):
+    # GH#63412
+    path = tmp_path / "test.h5"
+    index = MultiIndex.from_tuples([("a", "x"), ("b", "y")], names=["level1", "level2"])
+    df = DataFrame({"value": [1, 2]}, index=index)
+    df.to_hdf(path, key="test")
+    result = read_hdf(path, key="test")
+    tm.assert_frame_equal(df, result, check_dtype=False)