API triggered document ingestion on Windows causes permission error

[jaydavid]

When running locally on a Windows 11 machine, I am able to interact with the UI and upload files without issue.

However, when I attempt to connect to the server via another application using the API ingestion endpoint /v1/ingest I am seeing a permissions issue with the temporary file that's created to facilitate the ingestion. A snippet of the error:

  File "C:\repos\privateGPT\private_gpt\server\ingest\ingest_router.py", line 38, in ingest
    ingested_documents = service.ingest(file.filename, file.file.read())
                         ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "C:\repos\privateGPT\private_gpt\server\ingest\ingest_service.py", line 118, in ingest
    path_to_tmp.write_bytes(file_data)
  File "C:\Users\Jay\AppData\Local\Programs\Python\Python311\Lib\pathlib.py", line 1067, in write_bytes
    with self.open(mode='wb') as f:
         ^^^^^^^^^^^^^^^^^^^^
  File "C:\Users\Jay\AppData\Local\Programs\Python\Python311\Lib\pathlib.py", line 1044, in open
    return io.open(self, mode, buffering, encoding, errors, newline)
           ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
PermissionError: [Errno 13] Permission denied: 'C:\\Users\\Jay\\AppData\\Local\\Temp\\tmpxq8kqnc9'

Here is the demo code that I'm using to exercise the API. This is running on the node side of a next.js app.

    let file = await fileRepo.get(fileId);
    let formData = new FormData();
    formData.append("file", base64toBlob(file.data), file.name);
    console.log("ingesting file...");
    const response = await fetch("http://127.0.0.1:8001/v1/ingest", {
      method: "POST",
      body: formData,
    });

I am able to correct this issue locally if I make the following change, which I believe is due to the way that python requisitions temporary files on a Windows machine. The change is made in the /private_gpt/server/ingest/ingest_service.py file.

Current:

                with tempfile.NamedTemporaryFile() as tmp:
                    path_to_tmp = Path(tmp.name)
                    if isinstance(file_data, bytes):
                        path_to_tmp.write_bytes(file_data)
                    else:
                        path_to_tmp.write_text(str(file_data))
                    documents = reader.load_data(path_to_tmp)

Updated:

                with tempfile.NamedTemporaryFile(delete=False) as tmp:
                    tmp.close()
                    path_to_tmp = Path(tmp.name)
                    if isinstance(file_data, bytes):
                        path_to_tmp.write_bytes(file_data)
                    else:
                        path_to_tmp.write_text(str(file_data))
                    documents = reader.load_data(path_to_tmp)

I'm sure there are other implications to this change, and this is likely not a complete solution, but I wanted to at least get this issue tracked if this can help anyone else.

[pabloogc]

Thanks for suggesting the fix, I will open a PR with it.

[pabloogc]

@jaydavid can you check if #1260 solves it?

[jaydavid]

@pabloogc

Unfortunately, the branch does not resolve the issue. Here's a copy of the console output I get when I interact via the API

12:18:43.557 [INFO    ]            uvicorn.access - 127.0.0.1:53592 - "POST /v1/ingest HTTP/1.1" 500
12:18:43.557 [ERROR   ]             uvicorn.error - Exception in ASGI application
Traceback (most recent call last):
  File "C:\Users\Jay\AppData\Local\pypoetry\Cache\virtualenvs\private-gpt-9Rr4HheO-py3.11\Lib\site-packages\uvicorn\protocols\http\httptools_impl.py", line 426, in run_asgi
    result = await app(  # type: ignore[func-returns-value]
             ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "C:\Users\Jay\AppData\Local\pypoetry\Cache\virtualenvs\private-gpt-9Rr4HheO-py3.11\Lib\site-packages\uvicorn\middleware\proxy_headers.py", line 84, in __call__
    return await self.app(scope, receive, send)
           ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "C:\Users\Jay\AppData\Local\pypoetry\Cache\virtualenvs\private-gpt-9Rr4HheO-py3.11\Lib\site-packages\fastapi\applications.py", line 292, in __call__
    await super().__call__(scope, receive, send)
  File "C:\Users\Jay\AppData\Local\pypoetry\Cache\virtualenvs\private-gpt-9Rr4HheO-py3.11\Lib\site-packages\starlette\applications.py", line 122, in __call__
    await self.middleware_stack(scope, receive, send)
  File "C:\Users\Jay\AppData\Local\pypoetry\Cache\virtualenvs\private-gpt-9Rr4HheO-py3.11\Lib\site-packages\starlette\middleware\errors.py", line 184, in __call__
    raise exc
  File "C:\Users\Jay\AppData\Local\pypoetry\Cache\virtualenvs\private-gpt-9Rr4HheO-py3.11\Lib\site-packages\starlette\middleware\errors.py", line 162, in __call__
    await self.app(scope, receive, _send)
  File "C:\Users\Jay\AppData\Local\pypoetry\Cache\virtualenvs\private-gpt-9Rr4HheO-py3.11\Lib\site-packages\starlette\middleware\exceptions.py", line 79, in __call__
    raise exc
  File "C:\Users\Jay\AppData\Local\pypoetry\Cache\virtualenvs\private-gpt-9Rr4HheO-py3.11\Lib\site-packages\starlette\middleware\exceptions.py", line 68, in __call__
    await self.app(scope, receive, sender)
  File "C:\Users\Jay\AppData\Local\pypoetry\Cache\virtualenvs\private-gpt-9Rr4HheO-py3.11\Lib\site-packages\fastapi\middleware\asyncexitstack.py", line 20, in __call__
    raise e
  File "C:\Users\Jay\AppData\Local\pypoetry\Cache\virtualenvs\private-gpt-9Rr4HheO-py3.11\Lib\site-packages\fastapi\middleware\asyncexitstack.py", line 17, in __call__
    await self.app(scope, receive, send)
  File "C:\Users\Jay\AppData\Local\pypoetry\Cache\virtualenvs\private-gpt-9Rr4HheO-py3.11\Lib\site-packages\starlette\routing.py", line 718, in __call__
    await route.handle(scope, receive, send)
  File "C:\Users\Jay\AppData\Local\pypoetry\Cache\virtualenvs\private-gpt-9Rr4HheO-py3.11\Lib\site-packages\starlette\routing.py", line 276, in handle
    await self.app(scope, receive, send)
  File "C:\Users\Jay\AppData\Local\pypoetry\Cache\virtualenvs\private-gpt-9Rr4HheO-py3.11\Lib\site-packages\starlette\routing.py", line 66, in app
    response = await func(request)
               ^^^^^^^^^^^^^^^^^^^
  File "C:\Users\Jay\AppData\Local\pypoetry\Cache\virtualenvs\private-gpt-9Rr4HheO-py3.11\Lib\site-packages\fastapi\routing.py", line 273, in app
    raw_response = await run_endpoint_function(
                   ^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "C:\Users\Jay\AppData\Local\pypoetry\Cache\virtualenvs\private-gpt-9Rr4HheO-py3.11\Lib\site-packages\fastapi\routing.py", line 192, in run_endpoint_function
    return await run_in_threadpool(dependant.call, **values)
           ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "C:\Users\Jay\AppData\Local\pypoetry\Cache\virtualenvs\private-gpt-9Rr4HheO-py3.11\Lib\site-packages\starlette\concurrency.py", line 41, in run_in_threadpool
    return await anyio.to_thread.run_sync(func, *args)
           ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "C:\Users\Jay\AppData\Local\pypoetry\Cache\virtualenvs\private-gpt-9Rr4HheO-py3.11\Lib\site-packages\anyio\to_thread.py", line 33, in run_sync
    return await get_asynclib().run_sync_in_worker_thread(
           ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "C:\Users\Jay\AppData\Local\pypoetry\Cache\virtualenvs\private-gpt-9Rr4HheO-py3.11\Lib\site-packages\anyio\_backends\_asyncio.py", line 877, in run_sync_in_worker_thread
    return await future
           ^^^^^^^^^^^^
  File "C:\Users\Jay\AppData\Local\pypoetry\Cache\virtualenvs\private-gpt-9Rr4HheO-py3.11\Lib\site-packages\anyio\_backends\_asyncio.py", line 807, in run
    result = context.run(func, *args)
             ^^^^^^^^^^^^^^^^^^^^^^^^
  File "C:\strasz-repos\privateGPT\private_gpt\server\ingest\ingest_router.py", line 38, in ingest
    ingested_documents = service.ingest(file.filename, file.file.read())
                         ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "C:\strasz-repos\privateGPT\private_gpt\server\ingest\ingest_service.py", line 125, in ingest
    path_to_tmp.unlink()
  File "C:\Users\Jay\AppData\Local\Programs\Python\Python311\Lib\pathlib.py", line 1147, in unlink
    os.unlink(self)
PermissionError: [WinError 32] The process cannot access the file because it is being used by another process: 'C:\\Users\\Jay\\AppData\\Local\\Temp\\tmpe1n9k975'

If I add a tmp.close() in the beginning of the try block, it does resolve the issue - but I am not sure if this has implications outside of a Windows environment. I hope this info is helpful.

[pabloogc]

interesting, maybe the issue is caused by the path lib and works fine with a regular close