fix: pass metadata_file_path for content detection and use monkeypatch in tests

- Pass remote_file.uri as metadata_file_path to detect_filetype for
  content-based detection, preserving the extension hint for remote files.
- Use monkeypatch.setattr for partition globals in tests to ensure
  automatic cleanup and prevent mock leakage between tests.

Co-Authored-By: AJ Steers <aj@airbyte.io>

Author: devin-ai-integration[bot]

airbyte_cdk/sources/file_based/file_types/unstructured_parser.py

@@ -424,7 +424,9 @@ def _get_filetype(self, file: IOBase, remote_file: RemoteFile) -> Optional[FileT
         if file_type and file_type != FileType.UNK:
             return file_type
 
-        type_based_on_content = detect_filetype(file=cast(IO[bytes], file))
+        type_based_on_content = detect_filetype(
+            file=cast(IO[bytes], file), metadata_file_path=remote_file.uri
+        )
         file.seek(0)  # detect_filetype is reading to read the file content, so we need to reset
 
         if type_based_on_content and type_based_on_content != FileType.UNK:

unit_tests/sources/file_based/file_types/test_unstructured_parser.py

@@ -239,13 +239,20 @@ def test_parse_records(
     raises,
     expected_records,
     parsing_error,
+    monkeypatch,
 ):
     mock_partition_pdf = MagicMock()
     mock_partition_docx = MagicMock()
     mock_partition_pptx = MagicMock()
-    unstructured_parser_module.unstructured_partition_pdf = mock_partition_pdf
-    unstructured_parser_module.unstructured_partition_docx = mock_partition_docx
-    unstructured_parser_module.unstructured_partition_pptx = mock_partition_pptx
+    monkeypatch.setattr(
+        unstructured_parser_module, "unstructured_partition_pdf", mock_partition_pdf
+    )
+    monkeypatch.setattr(
+        unstructured_parser_module, "unstructured_partition_docx", mock_partition_docx
+    )
+    monkeypatch.setattr(
+        unstructured_parser_module, "unstructured_partition_pptx", mock_partition_pptx
+    )
     stream_reader = MagicMock()
     mock_open(stream_reader.open_file, read_data=bytes(str(parse_result), "utf-8"))
     fake_file = RemoteFile(uri=FILE_URI, last_modified=datetime.now())