English 中文(简体)
Python SocketServer: sending to multiple clients?
原标题:

Well, I m trying to build a small python prgram with a SocketServer that is supposed to send messages it receives to all connected clients. I m stuck, I don t know how to store clients on the serverside, and I don t know how to send to multiple clients. Oh and, my program fails everytime more then 1 client connects, and everytime a client sends more then one message...

Here s my code until now:

        print str(self.client_address[0])+  connected. 
    def handle(self):
        new=1
        for client in clients:
            if client==self.request:
                new=0
        if new==1:
            clients.append(self.request)
        for client in clients:
            data=self.request.recv(1024)
            client.send(data)

class Host:
    def __init__(self):
        self.address = ( localhost , 0)
        self.server = SocketServer.TCPServer(self.address, EchoRequestHandler)
        ip, port = self.server.server_address
        self.t = threading.Thread(target=self.server.serve_forever)
        self.t.setDaemon(True)
        self.t.start()
        print   
        print  Hosted with IP:  +ip+  and port:  +str(port)+ . Clients can now connect. 
        print   
    def close(self):
        self.server.socket.close()

class Client:
    name=  
    ip=  
    port=0
    def __init__(self,ip,port,name):
        self.name=name
        self.hostIp=ip
        self.hostPort=port
        self.s=socket.socket(socket.AF_INET, socket.SOCK_STREAM)
        self.s.connect((self.hostIp, self.hostPort))
    def reco(self):
        self.s=socket.socket(socket.AF_INET, socket.SOCK_STREAM)
        self.s.connect((self.hostIp, self.hostPort))
    def nick(self,newName):
        self.name=newName
    def send(self,message):
        message=self.name+  :  +message
        len_sent=self.s.send(message)
        response=self.s.recv(len_sent)
        print response
        self.reco()
    def close(self):
        self.s.close()

Obviously I have no idea what I m doing, so any help would be great.
Thanks in advance!

Edit: I m using Python 2.7 on Windows Vista.

最佳回答

You want to look at asyncore here. The socket operations you re calling on the client side are blocking (don t return until some data is received or a timeout occurs) which makes it hard to listen for messages sent from the host and let the client instances enqueue data to send at the same time. asyncore is supposed to abstract the timeout-based polling loop away from you.

Here s a code "sample" -- let me know if anything is unclear:

from __future__ import print_function

import asyncore
import collections
import logging
import socket


MAX_MESSAGE_LENGTH = 1024


class RemoteClient(asyncore.dispatcher):

    """Wraps a remote client socket."""

    def __init__(self, host, socket, address):
        asyncore.dispatcher.__init__(self, socket)
        self.host = host
        self.outbox = collections.deque()

    def say(self, message):
        self.outbox.append(message)

    def handle_read(self):
        client_message = self.recv(MAX_MESSAGE_LENGTH)
        self.host.broadcast(client_message)

    def handle_write(self):
        if not self.outbox:
            return
        message = self.outbox.popleft()
        if len(message) > MAX_MESSAGE_LENGTH:
            raise ValueError( Message too long )
        self.send(message)


class Host(asyncore.dispatcher):

    log = logging.getLogger( Host )

    def __init__(self, address=( localhost , 0)):
        asyncore.dispatcher.__init__(self)
        self.create_socket(socket.AF_INET, socket.SOCK_STREAM)
        self.bind(address)
        self.listen(1)
        self.remote_clients = []

    def handle_accept(self):
        socket, addr = self.accept() # For the remote client.
        self.log.info( Accepted client at %s , addr)
        self.remote_clients.append(RemoteClient(self, socket, addr))

    def handle_read(self):
        self.log.info( Received message: %s , self.read())

    def broadcast(self, message):
        self.log.info( Broadcasting message: %s , message)
        for remote_client in self.remote_clients:
            remote_client.say(message)


class Client(asyncore.dispatcher):

    def __init__(self, host_address, name):
        asyncore.dispatcher.__init__(self)
        self.log = logging.getLogger( Client (%7s)  % name)
        self.create_socket(socket.AF_INET, socket.SOCK_STREAM)
        self.name = name
        self.log.info( Connecting to host at %s , host_address)
        self.connect(host_address)
        self.outbox = collections.deque()

    def say(self, message):
        self.outbox.append(message)
        self.log.info( Enqueued message: %s , message)

    def handle_write(self):
        if not self.outbox:
            return
        message = self.outbox.popleft()
        if len(message) > MAX_MESSAGE_LENGTH:
            raise ValueError( Message too long )
        self.send(message)

    def handle_read(self):
        message = self.recv(MAX_MESSAGE_LENGTH)
        self.log.info( Received message: %s , message)


if __name__ ==  __main__ :
    logging.basicConfig(level=logging.INFO)
    logging.info( Creating host )
    host = Host()
    logging.info( Creating clients )
    alice = Client(host.getsockname(),  Alice )
    bob = Client(host.getsockname(),  Bob )
    alice.say( Hello, everybody! )
    logging.info( Looping )
    asyncore.loop()

Which results in the following output:

INFO:root:Creating host
INFO:root:Creating clients
INFO:Client (  Alice):Connecting to host at ( 127.0.0.1 , 51117)
INFO:Client (    Bob):Connecting to host at ( 127.0.0.1 , 51117)
INFO:Client (  Alice):Enqueued message: Hello, everybody!
INFO:root:Looping
INFO:Host:Accepted client at ( 127.0.0.1 , 55628)
INFO:Host:Accepted client at ( 127.0.0.1 , 55629)
INFO:Host:Broadcasting message: Hello, everybody!
INFO:Client (  Alice):Received message: Hello, everybody!
INFO:Client (    Bob):Received message: Hello, everybody!
问题回答

You can use socketserver to broadcast messages to all connected clients. However, the ability is not built into the code and will need to be implemented by extending some of the classes already provided. In the following example, this is implemented using the ThreadingTCPServer and StreamRequestHandler classes. They provide a foundation on which to build but still require some modifications to allow what you are trying to accomplish. The documentation should help explain what each function, class, and method are trying to do in order to get the job done.

Server

#! /usr/bin/env python3
import argparse
import pickle
import queue
import select
import socket
import socketserver


def main():
    """Start a chat server and serve clients forever."""
    parser = argparse.ArgumentParser(description= Execute a chat server demo. )
    parser.add_argument( port , type=int, help= location where server listens )
    arguments = parser.parse_args()
    server_address = socket.gethostbyname(socket.gethostname()), arguments.port
    server = CustomServer(server_address, CustomHandler)
    server.serve_forever()


class CustomServer(socketserver.ThreadingTCPServer):

    """Provide server support for the management of connected clients."""

    def __init__(self, server_address, request_handler_class):
        """Initialize the server and keep a set of registered clients."""
        super().__init__(server_address, request_handler_class, True)
        self.clients = set()

    def add_client(self, client):
        """Register a client with the internal store of clients."""
        self.clients.add(client)

    def broadcast(self, source, data):
        """Resend data to all clients except for the data s source."""
        for client in tuple(self.clients):
            if client is not source:
                client.schedule((source.name, data))

    def remove_client(self, client):
        """Take a client off the register to disable broadcasts to it."""
        self.clients.remove(client)


class CustomHandler(socketserver.StreamRequestHandler):

    """Allow forwarding of data to all other registered clients."""

    def __init__(self, request, client_address, server):
        """Initialize the handler with a store for future date streams."""
        self.buffer = queue.Queue()
        super().__init__(request, client_address, server)

    def setup(self):
        """Register self with the clients the server has available."""
        super().setup()
        self.server.add_client(self)

    def handle(self):
        """Run a continuous message pump to broadcast all client data."""
        try:
            while True:
                self.empty_buffers()
        except (ConnectionResetError, EOFError):
            pass

    def empty_buffers(self):
        """Transfer data to other clients and write out all waiting data."""
        if self.readable:
            self.server.broadcast(self, pickle.load(self.rfile))
        while not self.buffer.empty():
            pickle.dump(self.buffer.get_nowait(), self.wfile)

    @property
    def readable(self):
        """Check if the client s connection can be read without blocking."""
        return self.connection in select.select(
            (self.connection,), (), (), 0.1)[0]

    @property
    def name(self):
        """Get the client s address to which the server is connected."""
        return self.connection.getpeername()

    def schedule(self, data):
        """Arrange for a data packet to be transmitted to the client."""
        self.buffer.put_nowait(data)

    def finish(self):
        """Remove the client s registration from the server before closing."""
        self.server.remove_client(self)
        super().finish()


if __name__ ==  __main__ :
    main()

Of course, you also need a client that can communicate with your server and use the same protocol the server speaks. Since this is Python, the decision was made to utilize the pickle module to facilitate data transfer among server and clients. Other data transfer methods could have been used (such as JSON, XML, et cetera), but being able to pickle and unpickle data serves the needs of this program well enough. Documentation is included yet again, so it should not be too difficult to figure out what is going on. Note that server commands can interrupt user data entry.

Client

#! /usr/bin/env python3
import argparse
import cmd
import pickle
import socket
import threading


def main():
    """Connect a chat client to a server and process incoming commands."""
    parser = argparse.ArgumentParser(description= Execute a chat client demo. )
    parser.add_argument( host , type=str, help= name of server on the network )
    parser.add_argument( port , type=int, help= location where server listens )
    arguments = parser.parse_args()
    client = User(socket.create_connection((arguments.host, arguments.port)))
    client.start()


class User(cmd.Cmd, threading.Thread):

    """Provide a command interface for internal and external instructions."""

    prompt =  >>>  

    def __init__(self, connection):
        """Initialize the user interface for communicating with the server."""
        cmd.Cmd.__init__(self)
        threading.Thread.__init__(self)
        self.connection = connection
        self.reader = connection.makefile( rb , -1)
        self.writer = connection.makefile( wb , 0)
        self.handlers = dict(print=print, ping=self.ping)

    def start(self):
        """Begin execution of processor thread and user command loop."""
        super().start()
        super().cmdloop()
        self.cleanup()

    def cleanup(self):
        """Close the connection and wait for the thread to terminate."""
        self.writer.flush()
        self.connection.shutdown(socket.SHUT_RDWR)
        self.connection.close()
        self.join()

    def run(self):
        """Execute an automated message pump for client communications."""
        try:
            while True:
                self.handle_server_command()
        except (BrokenPipeError, ConnectionResetError):
            pass

    def handle_server_command(self):
        """Get an instruction from the server and execute it."""
        source, (function, args, kwargs) = pickle.load(self.reader)
        print( Host: {} Port: {} .format(*source))
        self.handlers[function](*args, **kwargs)

    def preloop(self):
        """Announce to other clients that we are connecting."""
        self.call( print , socket.gethostname(),  just entered. )

    def call(self, function, *args, **kwargs):
        """Arrange for a handler to be executed on all other clients."""
        assert function in self.handlers,  You must create a handler first! 
        pickle.dump((function, args, kwargs), self.writer)

    def do_say(self, arg):
        """Causes a message to appear to all other clients."""
        self.call( print , arg)

    def do_ping(self, arg):
        """Ask all clients to report their presence here."""
        self.call( ping )

    def ping(self):
        """Broadcast to all other clients that we are present."""
        self.call( print , socket.gethostname(),  is here. )

    def do_exit(self, arg):
        """Disconnect from the server and close the client."""
        return True

    def postloop(self):
        """Make an announcement to other clients that we are leaving."""
        self.call( print , socket.gethostname(),  just exited. )


if __name__ ==  __main__ :
    main()

why use SocketServer? a simple client doesn t meet your needs?

import socket

HOST =   
PORT = 8000
sock = socket.socket(socket.AF_INET, socket.SOCK_STREAM)
sock.bind((HOST, PORT))
sock.listen(5)
while True:
    conn, addr = sock.accept()
    print  connecting to , addr
    while True:
        data = conn.recv(1024)
        if not data:
            break
        conn.send(data)

To take multiple clients simultaneously, you will have to add SocketServer.ForkingMixIn or ThreadingMixIn.





相关问题
Can Django models use MySQL functions?

Is there a way to force Django models to pass a field to a MySQL function every time the model data is read or loaded? To clarify what I mean in SQL, I want the Django model to produce something like ...

An enterprise scheduler for python (like quartz)

I am looking for an enterprise tasks scheduler for python, like quartz is for Java. Requirements: Persistent: if the process restarts or the machine restarts, then all the jobs must stay there and ...

How to remove unique, then duplicate dictionaries in a list?

Given the following list that contains some duplicate and some unique dictionaries, what is the best method to remove unique dictionaries first, then reduce the duplicate dictionaries to single ...

What is suggested seed value to use with random.seed()?

Simple enough question: I m using python random module to generate random integers. I want to know what is the suggested value to use with the random.seed() function? Currently I am letting this ...

How can I make the PyDev editor selectively ignore errors?

I m using PyDev under Eclipse to write some Jython code. I ve got numerous instances where I need to do something like this: import com.work.project.component.client.Interface.ISubInterface as ...

How do I profile `paster serve` s startup time?

Python s paster serve app.ini is taking longer than I would like to be ready for the first request. I know how to profile requests with middleware, but how do I profile the initialization time? I ...

Pragmatically adding give-aways/freebies to an online store

Our business currently has an online store and recently we ve been offering free specials to our customers. Right now, we simply display the special and give the buyer a notice stating we will add the ...

Converting Dictionary to List? [duplicate]

I m trying to convert a Python dictionary into a Python list, in order to perform some calculations. #My dictionary dict = {} dict[ Capital ]="London" dict[ Food ]="Fish&Chips" dict[ 2012 ]="...

热门标签