Python-StreamedRequests/streamedrequests/dodownload.py
2019-07-08 23:54:07 -05:00

42 lines
1.8 KiB
Python

'''
StreamedRequests. A simple library for streaming HTTP requests
Copyright (C) 2019 Kevin Froman https://chaoswebs.net/
This program is free software: you can redistribute it and/or modify
it under the terms of the GNU General Public License as published by
the Free Software Foundation, either version 3 of the License, or
(at your option) any later version.
This program is distributed in the hope that it will be useful,
but WITHOUT ANY WARRANTY; without even the implied warranty of
MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
GNU General Public License for more details.
You should have received a copy of the GNU General Public License
along with this program. If not, see <https://www.gnu.org/licenses/>.
'''
import threading
from . import responsesize
def __run_callback(data, sync, callback=None):
if callback is None: # Do nothing if there is no callback
return
if sync: # If synchronous (default), run callback normally
callback(data)
else: # If async, spawn a new thread (not good for CPU-bound cases)
threading.Thread(target=callback, args=(data,)).start()
def __do_download(req, max_size, chunk_size, callback, sync):
chunk_count = responsesize.SizeValidator(max_size) # Class to verify if the stream is staying within the max_size
ret_data = b''
if chunk_size == 0:
raise ValueError("Chunk size cannot be zero")
for chunk in req.iter_content(chunk_size=chunk_size):
if max_size > 0:
chunk_count.add(chunk_size)
if not callback is None:
__run_callback(chunk, sync, callback)
ret_data += chunk
if 'text' in req.headers['content-type']:
ret_data = ret_data.decode('utf-8')
return (req, ret_data)