xfail test_read_multiple_parquet_files

apache · Dec 21, 2023 · 4c89276 · 4c89276
1 parent 77b4ecb
commit 4c89276
Show file tree

Hide file tree

Showing 2 changed files with 4 additions and 6 deletions.
diff --git a/python/pyarrow/filesystem.py b/python/pyarrow/filesystem.py
@@ -223,12 +223,8 @@ def read_parquet(self, path, columns=None, metadata=None, schema=None,
         table : pyarrow.Table
         """
         from pyarrow.parquet import ParquetDataset
-
-        if metadata is not None:
-            raise ValueError(
-                "Keyword 'metadata' is not supported with the Dataset API")
-
-        dataset = ParquetDataset(path, schema=schema, filesystem=self)
+        dataset = ParquetDataset(path, schema=schema, metadata=metadata,
+                                 filesystem=self)
         return dataset.read(columns=columns, use_threads=use_threads,
                             use_pandas_metadata=use_pandas_metadata)
 

diff --git a/python/pyarrow/tests/test_hdfs.py b/python/pyarrow/tests/test_hdfs.py
@@ -309,6 +309,8 @@ def _write_multiple_hdfs_pq_files(self, tmpdir):
         expected = pa.concat_tables(test_data)
         return expected
 
+    @pytest.mark.xfail(reason="legacy.FileSystem.read_parquet used legacy ParquetDataset "
+                       "that has been removed in PyArrow 15.0.0.", raises=TypeError)
     @pytest.mark.pandas
     @pytest.mark.parquet
     def test_read_multiple_parquet_files(self):