【原创】python多线程测试接口性能

除了使用性能测试工具进行性能测试,我们也可以直接用python多线程进行性能测试。

下面,使用这几个模块,对一个查询接口做性能测试:

requests:发送http请求

json:返回的字符串转换成json格式

threading:多线程

time:统计时间

具体实现过程见代码及注释。

import requests
import json
import threading
import time
# 定义请求基本地址
base_url = "http://127.0.0.1:8000"
success = 0
fail = 0
# 查询线程
def get_guest_list_thread(start_user,end_user):
    for i in range(start_user,end_user):
        phone = 13511001095 + i #13511001100
        r = requests.get(base_url+'/api/get_guest_list/', params={'eid':1,'phone':phone})
        # print(r.status_code) # 200
        # print(r.content) # b'{"status": 200, "message": "success", "data": {"realname": "alen", "phone": "13511001100", "email": "[email protected]", "sign": false}}'
        # print(r.json()) # {'status': 200, 'message': 'success', 'data': {'realname': 'alen', 'phone': '13511001100', 'email': '[email protected]', 'sign': False}}
        # print(type(r)) #  ,这个类型有json方法,不需要import json
        # print(r) # 
        # print(json.loads(r.content)) #需要import json,{'status': 200, 'message': 'success', 'data': {'realname': 'alen', 'phone': '13511001100', 'email': '[email protected]', 'sign': False}}
        result = r.json()
        global success,fail
        try:
            if phone=='13511001100' or phone=='13511001101':
                assert result['status'] == 20
                success +=1
            else:
                assert result['status'] == 10022
                success +=1
        except AssertionError as e:
            print('get error:'+str(phone))
            fail +=1

# 5个线程,25个数据
# lists = {1:6, 6:11, 11:16, 16:21, 21:26} # 可以这样写数据,也可以通过下面生成
data = 25
n = 5
step = int(data/n)
lists = {}
for i in range(1,n+1):
    lists[(i-1)*step+1]=i*step + 1
print(lists)


# 创建线程列表
threads = []
# 创建线程
for start_user,end_user in lists.items():
    t = threading.Thread(target=get_guest_list_thread,args=(start_user,end_user)) # args是一个元组
    threads.append(t)
if __name__ == '__main__':
    # 开始时间
    start_time = time.time()
    # 启动线程
    for i in range(len(lists)):
        threads[i].start()
    for i in range(len(lists)):
        threads[i].join()
    # 结束时间
    time.sleep(3) # 为了更明显看出用例执行耗时,加上休眠
    end_time = time.time()
    print("开始时间:"+str(start_time)+'>>>>>'+time.strftime("%Y-%m-%d %H:%M:%S", time.localtime(start_time)))
    print("结束时间:"+str(end_time)+'>>>>>'+time.strftime("%Y-%m-%d %H:%M:%S", time.localtime(end_time)))
    print('总共耗时:' + str(end_time - start_time))
    print('总共耗时:%.2f'%(end_time - start_time)) # 保留两位小数
    print('测试通过用例数:{}, 测试失败用例数:{}, 测试通过率为:{}'.format(success,fail,str(success*100/(success+fail))+'%'))

  

结果:

【原创】python多线程测试接口性能_第1张图片

 

转载于:https://www.cnblogs.com/uncleyong/p/7004415.html

你可能感兴趣的:(python)