Question about file upload functionality

[montao]

Hello!

I use serverless framework with AWS Lambda and a python backend. A bit surprised that there was only low-level (using the python cgi package) to receive and parse file uploads from a HTTP form post:

from cgi import parse_header, parse_multipart
from io import BytesIO
import uuid
import boto3
s3 = boto3.resource('s3')
# TODO: Use case-insensitive headers
c_type, c_data = parse_header(event['headers']['content-type'])
c_data['boundary'] = bytes(c_data['boundary'], 'utf8')
body_file = BytesIO(bytes(event['body'], 'utf8'))
if 'CONTENT-LENGTH' not in c_data:
    # TODO: Do this properly. Why is it needed?
    c_data['CONTENT-LENGTH'] = "9999999"  
form_data = parse_multipart(body_file, c_data)
# TODO: Loop here for several file uploads
data = form_data['file'][0].decode('UTF-8')  
identifier = str(uuid.uuid4())
s3.Bucket(settings.get_bucket()).put_object(Key=f"{identifier}.json",Body=data)

It looks like flask has built-in functionality to do it, e.g. a described: https://pythonhosted.org/Flask-Uploads/. My question is if I can use the flask built-in functionality for file upload if I use serverless-flask, or if something else is preferred way instead?