Question

I’m trying to upload a file via django forms, then send it to an API.

Here’s the encoding function:

#FYI, requestFile = request.FILES['file']
def EncodeFile(self, requestFile, fields = []):

    BOUNDARY    = '----------boundary------'
    CRLF        = '\r\n'
    body        = []
    # Add the metadata about the upload first

    for param in fields:
        body.extend(
          ['--' + BOUNDARY,
           'Content-Disposition: form-data; name="%s"' % param,
           '',
           fields[param],
           ])

    fileContent = requestFile.read()

    body.extend(
      ['--' + BOUNDARY,
       'Content-Disposition: form-data; name="file"; filename="%s"'
       % requestFile.name,
       # The upload server determines the mime-type, no need to set it.
       'Content-Type: ' + requestFile.content_type,
       '',
       fileContent,
       ])

    # Finalize the form body
    body.extend(['--' + BOUNDARY + '--', ''])
    result =  'multipart/form-data; boundary=%s' % BOUNDARY, CRLF.join(body)
    return result

The problem is that when it reaches “CRLF.join(body)” it complains about “‘utf8’ codec can’t decode byte 0xff in position 0: invalid start byte”.

The exact same piece of code works flawlessly from the command line with the exception that the requestFile is actually a path to a file, and I’m doing an open(requestFile, ‘rb’) before reading the contents.

I can not for the life of me figure out what to do next. I’ve been googling for an answer for the past 10 hours or so.

Answer 1

Apparently this line of code causes the problem:

'Content-Disposition: form-data; name="file"; filename="%s"' % requestFile.name,

The correct line would be instead:

'Content-Disposition: form-data; name="file"; filename="%s"' % smart_str(requestFile.name),