fix: 下载文件文件名为Path的BUG

core/config.py

@@ -39,7 +39,7 @@ FILE_SUFFIX = "-审核批注"
 
 ## 关键参数**
 use_non_fastgpt_llm = False
-use_lufa = True
+use_lufa = False
 use_jp_machine = True
 
 ## 关键参数**
@@ -60,7 +60,7 @@ else:
         base_backend_url = "http://192.168.252.71:48080"
     else:
         outer_backend_url = "http://172.21.107.45:48080"
-        base_fastgpt_url = "http://172.21.107.45:38080"
+        base_fastgpt_url = "http://172.21.107.45:3030"
         base_backend_url = "http://172.21.107.45:48080"
     segment_review_api_key = (
         "fastgpt-vLu2JHAfqwEq5FUQhvATFDK0yDS6fs804v7KwWBMyU4sRrHzh4UGl89Zpa"

mock.py

+import asyncio
+import json
+import sys
+import types
+
+from fastapi import HTTPException
+
+from main import DocumentParseRequest, parse_document
+
+
+PAYLOAD = {
+    "urls": [
+        "http://192.168.252.71:48080/admin-api/infra/file/24/get/d7ade63d9599ec4a81b2671df2d8fa48a117cfda96d0396cc8e405d0b9c1a89b.docx"
+    ],
+    "conversation_id": "0edb7152426b4c298270434f2f8810cd",
+    "ruleset_id": "金盘",
+}
+
+
+async def main() -> None:
+    try:
+        result = await parse_document(DocumentParseRequest(**PAYLOAD))
+    except HTTPException as exc:
+        print(
+            json.dumps(
+                {
+                    "type": "HTTPException",
+                    "status_code": exc.status_code,
+                    "detail": exc.detail,
+                },
+                ensure_ascii=False,
+                indent=2,
+            )
+        )
+        return
+
+    print(result.model_dump_json(indent=2))
+
+
+if __name__ == "__main__":
+    asyncio.run(main())

utils/http_util.py

@@ -85,6 +85,12 @@ def upload_file(path, input_url_to_inner=True, output_url_to_inner=False) -> str
         raise Exception(f"上传{path}失败 Response text: {response.text}")
 
 
+def _download_basename(filename: str) -> str:
+    filename = unquote(filename.strip().strip('"'))
+    filename = filename.replace("\\", "/")
+    return Path(filename).name or "downloaded_file"
+
+
 def _resolve_download_filename(url: str, response: requests.Response) -> str:
     content_disposition = response.headers.get("content-disposition", "")
     if content_disposition:
@@ -92,13 +98,13 @@ def _resolve_download_filename(url: str, response: requests.Response) -> str:
             r"filename\*=(?:UTF-8''|utf-8'')?([^;]+)", content_disposition
         )
         if match:
-            return unquote(match.group(1).strip().strip('"'))
+            return _download_basename(match.group(1))
 
         match = re.search(r'filename="?([^";]+)"?', content_disposition)
         if match:
-            return unquote(match.group(1).strip().strip('"'))
+            return _download_basename(match.group(1))
 
-    url_filename = Path(urlparse(url).path).name
+    url_filename = _download_basename(urlparse(url).path)
     if url_filename:
         return url_filename