使用python进行websocket的客户端压力测试

这个代码是从github上 找到。然后简单修改了下。大神运用了进程池，以及线程池的内容。所以保存下来，学习学习

然后需要说明的是：本次用的python2.7，也尝试用python3.6，但是老实出现websocket-client包和python3不能兼容的情况，提示没有相关的方法。所以不得已最后又采用了python2

# -*- coding:utf-8 -*-
# __author__ == 'chenmingle'

import websocket
import time
import threading
import json
import multiprocessing
import uuid
from threadpool import ThreadPool, makeRequests

# 修改成自己的websocket地址
WS_URL = "xxxx"
# 定义进程数
processes = 4
# 定义线程数（每个文件可能限制1024个，可以修改fs.file等参数）
thread_num = 700
index = 1


def on_message(ws, message):
    # print(message)
    pass


def on_error(ws, error):
    print(error)
    pass


def on_close(ws):
    # print("### closed ###")
    pass


def on_open(ws):
    global index
    index = index + 1

    def send_thread():
        # 设置你websocket的内容
        # 每隔10秒发送一下数据使链接不中断
        while True:
            ws.send(u'hello服务器')
            time.sleep(10)

    t = threading.Thread(target=send_thread)
    t.start()


def on_start(num):
    time.sleep(5)
    # websocket.enableTrace(True)
    ws = websocket.WebSocketApp(WS_URL + str(num),
                                on_message=on_message,
                                on_error=on_error,
                                on_close=on_close)
    ws.on_open = on_open
    ws.run_forever()


def thread_web_socket():
    # 线程池
    pool_list = ThreadPool(thread_num)
    num = list()
    # 设置开启线程的数量
    for ir in range(thread_num):
        num.append(ir)
    requests = makeRequests(on_start, num)
    [pool_list.putRequest(req) for req in requests]
    pool_list.wait()


if __name__ == "__main__":
    # 进程池
    pool = multiprocessing.Pool(processes=processes)
    # 设置开启进程的数量
    for i in xrange(processes):
        pool.apply_async(thread_web_socket)
    pool.close()
    pool.join()