同样在services.py 中自定义需要实现消息协议、传输控制,并且实现客户端存根clientStub和服务器端存根serverStub,服务器定义以及channel的定义。
此时,添加DistributedChannel分布式的channel,并在ThreadServer中添加了registry方法。
```
import threading import random import struct import time from io import BytesIO import socket import json from kazoo.client import KazooClient class InvalidOperation(BaseException): def __init__(self, message = None): self.message = message or 'involid operation' class MethodProtocol(object): '''' 解读方法名 ''' def __init__(self, connection): self.conn = connection def _read_all(self, size): """ 帮助我们读取二进制数据 :param size: 想要读取的二进制数据大小 :return: 二进制数据bytes """ # self.conn if isinstance(self.conn, BytesIO): buff = self.conn.read(size) return buff else: # 有时候长度大于每次读取的长度 have = 0 buff = b'' while have < size: chunk = self.conn.recv(size - have) buff += chunk l = len(chunk) have += l if l == 0: # 表示客户端已经关闭了 raise EOFError return buff def get_method_name(self): # 读取字符串长度 buff = self._read_all(4) length = struct.unpack('!I',buff)[0] # 读取字符串 buff = self._read_all(length) name = buff.decode() return name class DivideProtocol(object): """ divide过程消息协议转换工具 """ def args_encode(self, num1, num2=1): """ 将原始调用的请求参数转换打包成二进制消息数据 :param num1: int :param num2: int :return: bytes 二进制消息数据 """ name = 'divide' # 处理函数名 buff = struct.pack('!I', 6) # 无符号int buff += name.encode() # 处理参数1 buff2 = struct.pack('!B', 1) # 无符号byte buff2 += struct.pack('!i', num1) # 处理参数2 if num2 != 1: # 没有传参的时候 buff2 += struct.pack('!B', 2) buff2 += struct.pack('!i', num2) # 处理参数边界和组合成完整数据 buff += struct.pack('!I',len(buff2)) buff += buff2 return buff def _read_all(self, size): """ 帮助我们读取二进制数据 :param size: 想要读取的二进制数据大小 :return: 二进制数据bytes """ # self.conn if isinstance(self.conn, BytesIO): buff = self.conn.read(size) return buff else: # 有时候长度大于每次读取的长度 have = 0 buff = b'' while have < size: chunk = self.conn.recv(size - have) buff += chunk l = len(chunk) have += l if l == 0: # 表示客户端已经关闭了 raise EOFError return buff def args_decode(self, connection): """ 接受调用请求数据病进行解析 :param connection: 链接请求数据 socket BytesIO :return: 因为有多个参数,定义为字典 """ param_len_map = { 1:4, 2:4, } param_fmt_map = { 1:'!i', 2:'!i', } param_name_map = { 1: 'num1', 2: 'num2', } # 保存用来返回的参数字典 args = {} self.conn = connection # 处理方法的名字,已经提前被处理,稍后处理 # 处理消息边界 # 1) 读取二进制数据----read , ------ByteIO.read # 2) 将二进制数据转换为python的数据类型 buff = self._read_all(4) length = struct.unpack('!I',buff)[0] # 记录已经读取的长度值 have = 0 # 处理第一个参数 # 解析参数序号 buff = self._read_all(1) have += 1 param_seq = struct.unpack('!B', buff)[0] # 解析参数值 param_len = param_len_map[param_seq] buff = self._read_all(param_len) have += param_len param_fmt = param_fmt_map[param_seq] param = struct.unpack(param_fmt,buff)[0] # 设置解析后的字典 param_name = param_name_map[param_seq] args[param_name] = param if have >= length: return args # 处理第二个参数 # 解析参数序号 buff = self._read_all(1) param_seq = struct.unpack('!B', buff)[0] # 解析参数值 param_len = param_len_map[param_seq] buff = self._read_all(param_len) param_fmt = param_fmt_map[param_seq] param = struct.unpack(param_fmt, buff)[0] # 设置解析后的字典 param_name = param_name_map[param_seq] args[param_name] = param return args def result_encode(self, result): """ 将原始结果数据转换为消息协议二进制数据 :param result: :return: """ if isinstance(result,float): # 处理返回值类型 buff = struct.pack('!B', 1) buff += struct.pack('!f', result) return buff else: buff = struct.pack('!B', 2) # 处理返回值 length = len(result.message) # 处理字符串长度 buff += struct.pack('!I', length) buff += result.message.encode() return buff def result_decode(self, connection): """ 将返回值消息数据转换为原始返回值 :param connection: socket BytesIo :return: float InvalidOperation对象 """ self.conn = connection # 处理返回值类型 buff = self._read_all(1) result_type = struct.unpack('!B', buff)[0] if result_type == 1: #正常情况 buff = self._read_all(4) val = struct.unpack('!f', buff)[0] return val else: buff = self._read_all(4) length = struct.unpack('!I', buff)[0] # 读取字符串 buff = self._read_all(length) message = buff.decode() return InvalidOperation(message) class Channel(object): """ 用于客户端建立网络链接 """ def __init__(self, host, port): self.host = host self.port = port def get_connection(self): """ 获取链接对象 :return: 与服务器通讯的socket """ sock = socket.socket(socket.AF_INET, socket.SOCK_STREAM) sock.connect((self.host, self.port)) return sock class DistributedChannel(object): """ 支持分布式的zookeeper的RPC客户端链接工具 """ def __init__(self): # 创建kazoo对象,用来跟zookeeper链接,获取信息 zk = KazooClient('127.0.0.1:2181') zk.start() self.zk = zk self._servers = [] self._get_servers() # 第一次,手动开启 def _get_servers(self, event=None): """ 从zookeeper中获取所有可用的RPC服务器的地址 :return: """ self._servers = [] # 每次重新获取地址信息 # 从zookeeper中获取/rpc节点下的所有可用的rpc服务器节点 servers = self.zk.get_children('/rpc', watch=self._get_servers) # 监视的回调函数为自身 for server in servers: addr_data = self.zk.get('/rpc/' + server)[0] addr = json.loads(addr_data.decode()) self._servers.append(addr) def _get_server(self): """ 从可用的服务器列表中选出一台服务器 :return: {"host":xxx,"port":xxx} """ return random.choice(self._servers) def get_connection(self): """ 提供一个具体的与RPC服务器的链接socket :return: """ while True: addr = self._get_server() print(addr) try: sock = socket.socket(socket.AF_INET, socket.SOCK_STREAM) sock.connect((addr['host'], addr['port'])) except ConnectionRefusedError: time.sleep(1) continue else: return sock class ThreadServer(object): """ 多线成RPC服务器 """ def __init__(self, host, port, handlers): sock = socket.socket(socket.AF_INET,socket.SOCK_STREAM) # 地址复用 sock.setsockopt(socket.SOL_SOCKET, socket.SO_REUSEADDR, 1) self.host = host self.port = port # 绑定地址 sock.bind((self.host, self.port)) # 因为在启动的方法中才开启监听,所以不在此处开启 # sock.listen(128) self.sock = sock self.handlers = handlers def serve(self): """ 开启服务器运行,提供RPC服务 :return: """ # 开启服务器的监听,等待客户端的链接请求 self.sock.listen(128) print("服务器开启监听,ip地址为%s,port为%d..." % (self.host,self.port)) # 注册到zookeeper self.register_zookeeper() while True: # 不断的接收客户端的链接请求 client_sock, client_addr = self.sock.accept() print("与客户端%s建立连接" % str(client_addr)) t = threading.Thread(target= self.handle, args=(client_sock,)) t.start() def register_zookeeper(self): """ 在zookeeper中心注册本服务器的地址信息 :return: """ # 创建kazoo的客户端 zk = KazooClient('127.0.0.1:2181') # 建立与zookeeper的链接 zk.start() # 在zookeeper中创建节点保存数据 zk.ensure_path('/rpc') data = json.dumps({'host':self.host,'port':self.port}) zk.create('/rpc/server', data.encode(), ephemeral=True, sequence=True) # 子线程函数 def handle(self,client_sock): """ 子线程调用的方法,用来处理一个客户段的请求 :return: """ # 交个ServerStub,完成客户端的具体的RPC的调用请求 stub = ServerStub(client_sock, self.handlers) try: while True: # 不断的接收 stub.process() except EOFError: # 表示客户端关闭了连接 print('客户端关闭了连接') client_sock.close() class Server(object): """ RPC服务器 """ def __init__(self, host, port, handlers): sock = socket.socket(socket.AF_INET,socket.SOCK_STREAM) # 地址复用 sock.setsockopt(socket.SOL_SOCKET, socket.SO_REUSEADDR, 1) self.host = host self.port = port # 绑定地址 sock.bind((self.host, self.port)) # 因为在启动的方法中才开启监听,所以不在此处开启 # sock.listen(128) self.sock = sock self.handlers = handlers def serve(self): """ 开启服务器运行,提供RPC服务 :return: """ # 开启服务器的监听,等待客户端的链接请求 self.sock.listen(128) print("服务器开启监听,ip地址为%s,port为%d..." % (self.host,self.port)) while True: # 不断的接收客户端的链接请求 client_sock, client_addr = self.sock.accept() print("与客户端%s建立连接" % str(client_addr)) # 交个ServerStub,完成客户端的具体的RPC的调用请求 stub = ServerStub(client_sock, self.handlers) try: while True: # 不断的接收 stub.process() except EOFError: # 表示客户端关闭了连接 print('客户端关闭了连接') client_sock.close() class ClientStub(object): """ 用来帮助客户端完成远程过程调用 RPC调用 stub = ClientStub() stub.divide(200, 100) """ def __init__(self, channel): self.channel = channel self.conn = self.channel.get_connection() def divide(self, num1, num2 = 1): # 将调用的参数打包成消息协议的数据 proto = DivideProtocol() args = proto.args_encode(num1, num2) # 将消息数据通过网络发送给服务器 self.conn.sendall(args) # 接受服务器返回的消息数据,并进行解析 result = proto.result_decode(self.conn) # 将结果之(正常float 或 异常InvalidOperation)返回给客户端 if isinstance(result,float): return result else: raise result class ServerStub(object): """ 服务端存根 帮助服务端完成远端过程调用 """ def __init__(self, connection, handlers): """ :param connection: 与客户端的链接 :param handlers: 真正的本地函数路由 此处不以map的形式处理,实现类的形式 class Handler: @staticmethod def divide(): pass @staticmethod def add(): pass """ self.conn = connection self.method_proto = MethodProtocol(self.conn) self.process_map = { 'divide': self._process_divide, 'add': self._process_add } self.handlers = handlers def process(self): """ 当服务端接受了客户的链接,建立好链接后,完成远端调用的处理 :return: """ # 接收消息数据,并解析方法的名字 name = self.method_proto.get_method_name() # 根据解析获得的方法名,调用相应的过程协议,接收并解析消息数据 self.process_map[name]() def _process_divide(self): """ 处理除法过程调用 :return: """ proto = DivideProtocol() args = proto.args_decode(self.conn) # args = {'num1':xxx, 'num2':xxx} # 除法过程的本地调用------------------->>>>>>>>> # 将本地调用过程的返回值(包括可能的异常)打包成消息协议的数据,通过网络返回给客户端 try: val = self.handlers.divide(**args) except InvalidOperation as e: ret_message = proto.result_encode(e) else: ret_message = proto.result_encode(val) self.conn.sendall(ret_message) def _process_add(self): """ 处理加法过程调用 此方法暂时不识闲 :return: """ pass if __name__ == '__main__': # 目的:消息协议测试,模拟网络传输 # 构造消息数据 proto = DivideProtocol() # 测试一 # divide(200,100) # message = proto.args_encode(200,100) # 测试二 message = proto.args_encode(200) conn = BytesIO() conn.write(message) conn.seek(0) # 解析消息数据 method_proto = MethodProtocal(conn) name = method_proto.get_method_name() print(name) args = proto.args_decode(conn) print(args)
```
接下来,修改server.py文件
server.py
```
from services import InvalidOperation # from services import Server from services import ThreadServer import sys class Handlers: @staticmethod def divide(num1, num2 = 1): if num2 == 0: raise InvalidOperation('ck_god_err') val = num1/num2 return val if __name__ == '__main__': # 开启服务器 # _server = Server('127.0.0.1', 8000, Handlers) # _server.serve() # 从启动命令中提取服务器运行的ip地址和端口号,启动的多线程服务器 host = sys.argv[1] port = int(sys.argv[2]) _server = ThreadServer(host, port, Handlers) _server.serve()
```
最后,将 client.py文件也稍作修改。
```
import time from services import ClientStub # from services import Channel from services import DistributedChannel from services import InvalidOperation # 创建与服务器的连接 # channel = Channel('127.0.0.1', 8000) channel = DistributedChannel() # 进行调用 for i in range(50): try: # 创建用于rpc调用的工具 stub = ClientStub(channel) # 初始化的时候才真正的创建连接了,所以放到里面 val = stub.divide(i * 100,100) # val = stub.divide(i * 100) # val = stub.divide( 100, 0) except InvalidOperation as e: print(e.message) else: print(val) time.sleep(1)
```
ctrl + shift + T在pycharm中打开多个Terminal窗口
右键运行客户端,可以看到不断地随机切换服务器。
免责声明:本站发布的内容(图片、视频和文字)以原创、转载和分享为主,文章观点不代表本网站立场,如果涉及侵权请联系站长邮箱:is@yisu.com进行举报,并提供相关证据,一经查实,将立刻删除涉嫌侵权内容。