GH-48254: [Python][Parquet] Support extension types in read_schema

Author: Kuinox

python/pyarrow/parquet/core.py

@@ -265,9 +265,9 @@ class ParquetFile:
     page_checksum_verification : bool, default False
         If True, verify the checksum for each page read from the file.
     arrow_extensions_enabled : bool, default True
-        If True, read Parquet logical types as Arrow extension types where possible,
-        (e.g., read JSON as the canonical `arrow.json` extension type or UUID as
-        the canonical `arrow.uuid` extension type).
+        If True, read Parquet logical types as Arrow extension types where
+        possible (e.g., read JSON as the canonical `arrow.json` extension type
+        or UUID as the canonical `arrow.uuid` extension type).
 
     Examples
     --------
@@ -2372,7 +2372,7 @@ def write_metadata(schema, where, metadata_collector=None, filesystem=None,
 
 
 def read_metadata(where, memory_map=False, decryption_properties=None,
-                  filesystem=None):
+                  filesystem=None, arrow_extensions_enabled=True):
     """
     Read FileMetaData from footer of a single Parquet file.
 
@@ -2387,6 +2387,10 @@ def read_metadata(where, memory_map=False, decryption_properties=None,
         If nothing passed, will be inferred based on path.
         Path will try to be found in the local on-disk filesystem otherwise
         it will be parsed as an URI to determine the filesystem.
+    arrow_extensions_enabled : bool, default True
+        If True, read Parquet logical types as Arrow extension types where
+        possible (e.g. UUID as the canonical `arrow.uuid` extension type).
+        If False, use the underlying storage types instead.
 
     Returns
     -------
@@ -2416,13 +2420,17 @@ def read_metadata(where, memory_map=False, decryption_properties=None,
         file_ctx = where = filesystem.open_input_file(where)
 
     with file_ctx:
-        file = ParquetFile(where, memory_map=memory_map,
-                           decryption_properties=decryption_properties)
+        file = ParquetFile(
+            where,
+            memory_map=memory_map,
+            decryption_properties=decryption_properties,
+            arrow_extensions_enabled=arrow_extensions_enabled,
+        )
         return file.metadata
 
 
 def read_schema(where, memory_map=False, decryption_properties=None,
-                filesystem=None):
+                filesystem=None, arrow_extensions_enabled=True):
     """
     Read effective Arrow schema from Parquet file metadata.
 
@@ -2437,6 +2445,9 @@ def read_schema(where, memory_map=False, decryption_properties=None,
         If nothing passed, will be inferred based on path.
         Path will try to be found in the local on-disk filesystem otherwise
         it will be parsed as an URI to determine the filesystem.
+    arrow_extensions_enabled : bool, default True
+        If True, read Parquet logical types as Arrow extension types where
+        possible (e.g., UUID as the canonical `arrow.uuid` extension type).
 
     Returns
     -------
@@ -2462,9 +2473,12 @@ def read_schema(where, memory_map=False, decryption_properties=None,
 
     with file_ctx:
         file = ParquetFile(
-            where, memory_map=memory_map,
-            decryption_properties=decryption_properties)
-        return file.schema.to_arrow_schema()
+            where,
+            memory_map=memory_map,
+            decryption_properties=decryption_properties,
+            arrow_extensions_enabled=arrow_extensions_enabled,
+        )
+        return file.schema_arrow
 
 
 __all__ = (

python/pyarrow/tests/parquet/test_metadata.py

@@ -814,3 +814,22 @@ def msg(c):
 
     with pytest.raises(TypeError, match=msg("FileMetaData")):
         pq.FileMetaData()
+
+
+def test_read_schema_uuid_extension_type(tmp_path):
+    data = [
+        b'\xe4`\xf9p\x83QGN\xac\x7f\xa4g>K\xa8\xcb',
+        b'\x1et\x14\x95\xee\xd5C\xea\x9b\xd7s\xdc\x91BK\xaf',
+        None,
+    ]
+    table = pa.table([pa.array(data, type=pa.uuid())], names=["ext"])
+
+    file_path = tmp_path / "uuid.parquet"
+    file_path_str = str(file_path)
+    pq.write_table(table, file_path_str, store_schema=False)
+
+    schema_default = pq.read_schema(file_path_str)
+    assert schema_default.field("ext").type == pa.uuid()
+
+    schema_disabled = pq.read_schema(file_path_str, arrow_extensions_enabled=False)
+    assert schema_disabled.field("ext").type == pa.binary(16)