fix(worker): normalize encoded document filenames

apps/api/app/services/document_ingestion/creation_service.py

@@ -5,7 +5,7 @@
 from dataclasses import dataclass
 from datetime import datetime, timezone
 from typing import cast
-from urllib.parse import urlparse
+from urllib.parse import unquote, urlparse
 
 from app.repositories.job_repository import JobRepository
 from app.services.document_ingestion.command import DocumentIngestionCommand
@@ -308,7 +308,7 @@ def _build_job_response(
 
 def _resolve_url_source_file_name(*, source_url: str, file_extension: str) -> str:
     parsed_url = urlparse(source_url)
-    url_basename = str(os.path.basename(parsed_url.path))
+    url_basename = unquote(str(os.path.basename(parsed_url.path)))
     if url_basename and os.path.splitext(url_basename)[1].lower() == file_extension:
         return url_basename
     if url_basename:

apps/api/tests/unit/test_document_ingestion_filename.py

@@ -0,0 +1,18 @@
+from app.services.document_ingestion.creation_service import (
+    _resolve_url_source_file_name,
+)
+
+
+def test_resolve_url_source_file_name_decodes_percent_encoded_path() -> None:
+    source_url = (
+        "https://files.example.test/"
+        "%E4%B8%AD%E6%96%87%E7%AE%80%E5%8E%86-%E9%9B%B7%E7%BF%94-"
+        "%E4%B8%AD%E7%A7%91%E9%99%A2%285%29.doc"
+    )
+
+    source_file_name = _resolve_url_source_file_name(
+        source_url=source_url,
+        file_extension=".doc",
+    )
+
+    assert source_file_name == "中文简历-雷翔-中科院(5).doc"

apps/worker/app/services/document_parser/support/internal_parse_name.py

@@ -3,6 +3,7 @@
 
 import os
 from dataclasses import dataclass
+from urllib.parse import unquote
 
 from app.services.common.file_utils import path_handle
 from app.services.document_parser.support.filename_limits import (
@@ -27,6 +28,7 @@ def normalize_internal_parse_name(
     candidate_name = (
         os.path.basename(filename) if isinstance(filename, str) and filename else ""
     )
+    candidate_name = unquote(candidate_name)
     cleaned_name = (
         path_handle(candidate_name, mode="clean_single") if candidate_name else ""
     )

apps/worker/tests/unit/test_internal_parse_name.py

@@ -4,10 +4,22 @@
 from pathlib import Path
 
 from app.services.document_parser.support.internal_parse_name import (
+    normalize_internal_parse_name,
     prepare_internal_parse_input,
 )
 
 
+def test_normalize_internal_parse_name_decodes_url_encoded_filename() -> None:
+    encoded_filename = (
+        "%E4%B8%AD%E6%96%87%E7%AE%80%E5%8E%86-%E9%9B%B7%E7%BF%94-"
+        "%E4%B8%AD%E7%A7%91%E9%99%A2%285%29.doc"
+    )
+
+    normalized_name = normalize_internal_parse_name(encoded_filename)
+
+    assert normalized_name == "中文简历-雷翔-中科院-5-.doc"
+
+
 def test_prepare_internal_parse_input_handles_long_encoded_filename(
     tmp_path,
 ) -> None: