Python Forum

Hi guys, first time posting on the Python forums

I am wondering how do asynchronous socket servers go about maintaining so many clients, specifically, how can the server have more than one client in a connected TCP stream if it is on a single port, or is there some automagical stuff happening to assign different ports behind the scenes?

for example in the doceumentation at
https://docs.python.org/3.7/library/asyn...-based-api
the server is serving on port 8888, however this expects multiple clients to have connections.

import asyncio

async def handle_echo(reader, writer):
    data = await reader.read(100)
    message = data.decode()
    addr = writer.get_extra_info('peername')

    print(f"Received {message!r} from {addr!r}")

    print(f"Send: {message!r}")
    writer.write(data)
    await writer.drain()

    print("Close the connection")
    writer.close()

async def main():
    server = await asyncio.start_server(
        handle_echo, '127.0.0.1', 8888)

    addr = server.sockets[0].getsockname()
    print(f'Serving on {addr}')

    async with server:
        await server.serve_forever()

asyncio.run(main())

Thanks

With a trick the developers made asynchronous code looking like synchronous code.
They use generators with the send function, which allows to delegate everything.

It's a combination of:

• DefaultSelector, which uses select, epoll and other os related selectors for sockets
  
• Generators (await was before yield from, async was previously a normal generator with a yield statement used with send from outside)
  
• EventLoop, which is asyncio.run_forever()
  

If you want to dig deeper in asyncio, you should watch some talks on YouTube.
Some of them are explaining it very deep.