Files with Certain Unicode Characters Don't Upload

[TimLudwinski]

Client Code

>>> r = requests.post("http://server/upload", files={"file": (u'\u0412\u0430\u043b\u044e\u0442\u043d\u043e\u0435\u0437\u0430\u043a\u043e\u
043d\u043e\u0434\u0430\u0442\u0435\u043b\u044c\u0441\u0442\u0432\u043e.\u0420\u0438\u0441\u043a\u0438\u043a\u043e\u0440\u0440\u0435\u043a\u0442\u0438\u0440\u043e\u0432\u043a\u0438\
u0442\u0430\u043c\u043e\u0436\u0435\u043d\u043d\u043e\u0439\u0441\u0442\u043e\u0438\u043c\u043e\u0441\u0442\u0438.doc', file_bytes)})

Server Code

@route("/upload", method='POST')
def upload():
    if request.files.file != "":
        data = request.files.file
    else:
        data = None

    if data is None:
        abort(400, "No file provided")

    return {"status": "success"}

Result:

>>> r.status_code
400
>>> r.text
'...No file provided..'

[TimLudwinski]

Using UWSGI.

[eric-wieser]

To be clear, the problem is with file names, not the files themselves?

[TimLudwinski]

Well, the application is told that there is no file in this case. Maybe I should be reporting this further upstream since it may not be strictly a bottle problem.

On Tuesday, May 24, 2016, Eric Wieser notifications@github.com wrote:

To be clear, the problem is with file names, not the files themselves?

— You are receiving this because you authored the thread. Reply to this email directly or view it on GitHub https://github.com/bottlepy/bottle/issues/852#issuecomment-221321867

[eric-wieser]

Let me clarify my question:

Files with Certain Unicode Characters Don't upload

Is this:

• Files with certain unicode characters in their name don't upload
• Files with certain unicode characters in the file's contents don't upload?

[TimLudwinski]

It's the first one. "Files with certain unicode characters in their name don't upload".

[defnull]

The requests urllib3 library uses extended header attributes, which I suspect cgi.FieldStorage does not understand.

Content-Disposition: form-data; name="file"; filename*=utf-8''%D0%92%D0%B0%D0%BB%D1%8E%D1%82%D0%BD%D0%BE%D0%B5%D0%B7%D0%B0%D0%BA%D0%BE%D0%BD%D0%BE%D0%B4%D0%B0%D1%82%D0%B5%D0%BB%D1%8C%D1%81%D1%82%D0%B2%D0%BE.%D0%A0%D0%B8%D1%81%D0%BA%D0%B8%D0%BA%D0%BE%D1%80%D1%80%D0%B5%D0%BA%D1%82%D0%B8%D1%80%D0%BE%D0%B2%D0%BA%D0%B8%D1%82%D0%B0%D0%BC%D0%BE%D0%B6%D0%B5%D0%BD%D0%BD%D0%BE%D0%B9%D1%81%D1%82%D0%BE%D0%B8%D0%BC%D0%BE%D1%81%D1%82%D0%B8.doc

I'm not really sure yet.

[defnull]

Confirmed. And it seems that werkzeug or the stdlib email library don't parse these either. Support for RFC 2231 seems to be rare.

[defnull]

https://github.com/shazow/urllib3/issues/900

[defnull]

Confirmed as a bug/misfeature in urllib3, which is used by requests. They are working on a fix.

I don't see an easy way to work around this issue other than not using requests/urllib3 for the moment.