▓▒▓▒▓▒▓▒▓▒▓▒▓▒▓▒▓▒▓▒▓▒▓▒▓▒▓▒▓▒▓▒▓▒▓▒▓▒▓▒▓▒▓▒▓▒▓▒ ▒▒▒▒▒▒▒▒▒▒▒▒▒▒▒▒▒▒▒▒▒▒▒▒▒▒▒▒▒▒▒▒▒▒▒▒▒▒▒▒▒▒▒▒▒▒▒▒ ░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░ ░████████╗░██████╗░██████╗░░█████╗░██╗░██████╗░░ ░╚══██╔══╝██╔═══██╗██╔══██╗██╔══██╗██║██╔═══██╗░ ░░░░██║░░░██║░░░██║██████╔╝███████║██║██║░░░██║░ ░░░░██║░░░██║░░░██║██╔══██╗██╔══██║██║██║░░░██║░ ░░░░██║░░░╚██████╔╝██║░░██║██║░░██║██║╚██████╔╝░ ░░░░╚═╝░░░░╚═════╝░╚═╝░░╚═╝╚═╝░░╚═╝╚═╝░╚═════╝░░ ░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░ ▒▒▒▒▒▒▒▒▒▒▒▒▒▒▒▒▒▒▒▒▒▒▒▒▒▒▒▒▒▒▒▒▒▒▒▒▒▒▒▒▒▒▒▒▒▒▒▒ ▓▒▓▒▓▒▓▒▓▒▓▒▓▒▓▒▓▒▓▒▓▒▓▒▓▒▓▒▓▒▓▒▓▒▓▒▓▒▓▒▓▒▓▒▓▒▓▒
using toraio you can easily create a small pool of tor proxies so you can take advantage of having multiple circuits at your disposal and multiple ip addresses, while running just 1 process. It also brings a ClientSession, which is the ClientSession from aiohttp with a small modification: it shifts proxies at every request.
yes! toraio.Pool() is a context manager which, while in context, patches the low level socket library, so it's usage is easy and compatible for almost all setups/libraries.
Pool will automatically cleanup the temp folder after your program ends. the folder is needed for tor to store circuit and consensus data. if you specify a folder explicitly, like Pool(user_data_dir="c:\some\folder\tordata"), it is assumed you want a dedicated folder. in that case cleanup will not happen. using a dedicated folder will greatly improve loading time on next runs.
only once! toraio should ideally be started on your mainthread. it will block to bootstrap the tor connections, and this only happens once during the full lifecycle of your program. while subprocesses are supported by asyncio, it would render useless as proxies can't be used anyway, until initialization is finished. It also brings a lot of other unexpected behavior.
toraio.Pool is a true singleton. Every Pool() you create will refer to the same instance. If you need more proxies, just use the amount parameter (defaults to 10, which is more than sufficient, and actually too much).
proxies in the pool are refreshed at least once 15 minutes, getting new circuits from different countries, and new ip's.
for most people, setting up tor is already quite a task, implementing them as proxies in their programs requires much more work. the purpose of toraio is to make it as easy as it can possibly get.
import logging
import asyncio
import toraio
async def main():
async with toraio.ClientSession() as session:
for _ in range(10):
async with session.get('http://httpbin.org/ip') as response:
print ( await response.json() )
logging.basicConfig(level=10)
asyncio.run(main())this will output
DEBUG:asyncio:Using proactor: IocpProactor
INFO:pool:starting proxy pool 0x2a214dc9be0
INFO:pool:tor bootstrap 0.0 completed
INFO:pool:tor bootstrap 5.0 completed
INFO:pool:tor bootstrap 10.0 completed
INFO:pool:tor bootstrap 14.0 completed
INFO:pool:tor bootstrap 15.0 completed
INFO:pool:tor bootstrap 20.0 completed
INFO:pool:tor bootstrap 25.0 completed
INFO:pool:tor bootstrap 30.0 completed
INFO:pool:tor bootstrap 40.0 completed
INFO:pool:tor bootstrap 45.0 completed
INFO:pool:tor bootstrap 50.0 completed
INFO:pool:tor bootstrap 55.0 completed
INFO:pool:tor bootstrap 60.0 completed
INFO:pool:tor bootstrap 68.0 completed
INFO:pool:tor bootstrap 75.0 completed
INFO:pool:tor bootstrap 80.0 completed
INFO:pool:tor bootstrap 85.0 completed
INFO:pool:tor bootstrap 89.0 completed
INFO:pool:tor bootstrap 90.0 completed
INFO:pool:tor bootstrap 95.0 completed
INFO:pool:tor bootstrap 100.0 completed
DEBUG:toraio._client_session:proxy switched
{'origin': '109.70.100.32'}
DEBUG:toraio._client_session:proxy switched
{'origin': '107.189.31.241'}
DEBUG:toraio._client_session:proxy switched
{'origin': '89.163.143.8'}
DEBUG:toraio._client_session:proxy switched
{'origin': '45.153.160.133'}
DEBUG:toraio._client_session:proxy switched
{'origin': '104.244.75.33'}
DEBUG:toraio._client_session:proxy switched
{'origin': '199.249.230.187'}
DEBUG:toraio._client_session:proxy switched
{'origin': '185.220.102.245'}
DEBUG:toraio._client_session:proxy switched
{'origin': '213.61.215.54'}
DEBUG:toraio._client_session:proxy switched
{'origin': '193.189.100.199'}
DEBUG:toraio._client_session:proxy switched
{'origin': '176.10.99.200'}
Lets run this again
asyncio.run(main())# no initialization is done
# it returns immediately
DEBUG:toraio._client_session:proxy switched
{'origin': '109.70.100.32'}
DEBUG:toraio._client_session:proxy switched
{'origin': '107.189.31.241'}
DEBUG:toraio._client_session:proxy switched
{'origin': '89.163.143.8'}
DEBUG:toraio._client_session:proxy switched
{'origin': '45.153.160.133'}
DEBUG:toraio._client_session:proxy switched
{'origin': '104.244.75.33'}
DEBUG:toraio._client_session:proxy switched
{'origin': '199.249.230.187'}
DEBUG:toraio._client_session:proxy switched
{'origin': '185.220.102.245'}
DEBUG:toraio._client_session:proxy switched
{'origin': '213.61.215.54'}
DEBUG:toraio._client_session:proxy switched
{'origin': '193.189.100.199'}
DEBUG:toraio._client_session:proxy switched
{'origin': '176.10.99.200'}
Showing different ways to get a pool and showing how to use the pool as context manager
import toraio
pool = toraio.start() # this gives you a running pool
pool2 = toraio.Pool() # this gives you the same pool (because singleton) if already bootstrapped
pool2.start() # will start and bootstrap the pool, if not already running
# start() has the pool itself as return value
pool3 = toraio.Pool.get_instance() # same story
pool4 = toraio.start() # same story
import requests
# context manager
with pool:
# in this block, all network will be handled by the proxies and it will return
# different ip's on subsequent calls
print(requests.get('http://httpbin.org/ip').text)
# not in context
# you will see your own ip
print(requests.get('http://httpbin.org/ip').text)