接口并发性能测试开发之：从测试方案设计、测试策略、指标分析到代码编写，这一篇全搞定。

# -*- coding: utf-8 -*-
"""
@ auth : carl_DJ
@ time : 2021-3-6
"""
import hashlib
import threading
from time import *
from datetime import datetime,timedelta
import requests
import json
'''初始化全局变量'''
#自定义全局变量需要的线程数,10
thread_num = 10
#自定义全局变量每个线程需要循环的数量,10
one_worker_num = 10
#设定最开始的总时间
sum_time = 0.00
#设定最开始的成功连接数
success_count = 0
''' 后台登录常规操作'''
username = 'carldj'
password = hashlib.md5(b'123456').hexdigest()  #设置密码，且是md5加密方式
url = "http://www.xxx.com/energy/user/login/"
form_data = {"username":username,"password":password}
login_response = requests.post(url,data=form_data)
c = login_response.cookies
 '''订单发送请求'''
def order():
    #引用全局变量
    global c
    global sum_time
    global success_count
    #获取执行发送订单请求前时间
    t1 = time()
    #设定url、form_data进行创建订单
    url1 = "http://www.xxx.com/energy/create_order/"
    from_data1 = {"restaurant_id":1136,
                  "menu_item_total":'12.00',
                  "menu_item_data": [{'id':2667868,'p':22,'q':3}]
                  }
    make_responst = requests.post(url1,data=from_data1,cookies = c)
    #获取请求结果
    res = make_responst.text
    #结果转换成字典赋值给变量id
    id = json.loads(res)['order_id']
    #断言判断是否提交成功
    assert  id != " "
    su_time =datetime.now()+ timedelta(hours=1)
    #设定url、form_data进行生成订单
    url2 = "http://www.xxx.com/energy/place_order/"
    from_data2 = {"restaurant_id": id,
                  "customer_name": 'carl_dj',
                  "mobile_number":username,
                  "delivery_address":"address message",
                  "pay_type":'cash',
                  "preorder":su_time
                  }
    place_responst = requests.post(url2, data=from_data2, cookies=c)
    res = place_responst.text
    #追加断言，判断结果是否有"success"，有的话，说明订餐成功
    assert res == " success"
    print("订餐成功")
    #订单成功后，再次获取一下时间
    t2 = time()
    #获取订单的响应时间
    res_time = t2-t1
    #把响应时间写入txt文件
    result = open("E:\Private Folder\res.txt","a")  #路径直接写死，也可用os.path 来写路径
    result.write("成功订单响应时间：" + str(res_time)+ '\n')
    result.close()
    #也可以使用with打开文件，好处是不用关心文件是否关闭
    # with open ("E:\Private Folder\res.txt","a") as result1:
        # print(result1.read())
    #把每次成功订单数累加到全局变量sum_time中
    sum_time  = sum_time + res_time
    #把每次获取的成功订单数做累加，添加到全局变量success_count中
    success_count = success_count +1
'''嵌套指定循环次数的order()函数'''
def working()
    global one_worker_num
    for i in range(0,one_worker_num):
        order()
 '''自定义main（）函数，来执行多线程'''
def main():
    global thread_num
    #自定义一个空的数组，用来存放线程组
    threads = []
    #设置循环次数
    for i in range(thread_num):
        #将working()函数存放到线程中
        t = threading.Thread(target=working,name="T"+ str(i))
        #设定守护线程
        t.setDaemon(True)
        threads.append(t)
    #启动循环执行
    for t in threads:
        t.start()
    ##设置阻塞线程
    for t in threads:
        t.join()
if __name__ == "__main__":
    main()
    total_order = thread_num*one_worker_num
    avg_time = sum_time/success_count
    '''执行完之后，需要把数据写入到txt文件中'''
    #订单并发总数
    result.write("并发订单数："+ str(total_order)+ "\n")
    #成功并发数
    result.write("成功并发数："+ str(success_count) + "\n")
    #订单成功率
    result.write("订单成功率："+ str(success_count/total_order*100)+ "%" + "\n")
    #成功订单响应时间
    result.write("成功订单总响应时间："+ str(sum_time)+"\n")
    #成功订单平均响应时间
    result.write("成功平均响应时间："+str(sum_time/success_count)+"\n")
    #TPS事务数/秒
    result.write("TPS:"+str(success_count/avg_time) + "\n")  #tps = 并发成功数/平均响应时间
    result.close()