asyncio + pycurl + BytesIO 异步批量调用url请求

Posted 2021-11-11

import asyncio
import pycurl
from io import BytesIO
import json

def fetch_api(url, method, header=None, data=None):
"""
url： 获取api的url
method: 请求方法
header: 请求头
data: 请求参数
"""

if method == "get":
    _Curl = pycurl.Curl()
    buf = BytesIO()
    _Curl.setopt(pycurl.WRITEFUNCTION,buf.write)
    _Curl.setopt(pycurl.URL, url)
    _Curl.setopt(pycurl.HTTPHEADER, header)
    _Curl.perform()
    result = buf.getvalue().decode("utf-8")
    http_code = _Curl.getinfo(_Curl.HTTP_CODE)
    dns_time = _Curl.getinfo(_Curl.NAMELOOKUP_TIME) * 1000 # DNS解析时间
    connect_time = _Curl.getinfo(_Curl.CONNECT_TIME) * 1000 # 建连时间
    pretransafe_time = _Curl.getinfo(_Curl.PRETRANSFER_TIME) * 1000 # 连接上后到开始传输时的时间
    starttransafe_time =_Curl.getinfo(_Curl.STARTTRANSFER_TIME) * 1000 # 接收到第一个字节的时间
    redirect_time = _Curl.getinfo(_Curl.REDIRECT_TIME) * 1000 # 重定向时间
    total_time = _Curl.getinfo(_Curl.TOTAL_TIME) * 1000 # 请求总时间
    download_size = _Curl.getinfo(_Curl.SIZE_DOWNLOAD) # 下载数据包大小
    download_speed = _Curl.getinfo(_Curl.SPEED_DOWNLOAD) # 下载速度
    buf.close()
    print(result)
    print(http_code,dns_time,connect_time,pretransafe_time,starttransafe_time,redirect_time,total_time,download_size,download_speed)
    _Curl.close()

else:
    _Curl = pycurl.Curl()
    _Curl.setopt(pycurl.HTTPHEADER, header)

    buf = BytesIO()
    _Curl.setopt(pycurl.WRITEFUNCTION, buf.write)
    _Curl.setopt(pycurl.URL, url)
    _Curl.setopt(_Curl.POSTFIELDS, data)

    _Curl.perform()
    result = buf.getvalue().decode("utf-8")
    http_code = _Curl.getinfo(_Curl.HTTP_CODE)
    dns_time = _Curl.getinfo(_Curl.NAMELOOKUP_TIME) * 1000  # DNS解析时间
    connect_time = _Curl.getinfo(_Curl.CONNECT_TIME) * 1000  # 建连时间
    pretransafe_time = _Curl.getinfo(_Curl.PRETRANSFER_TIME) * 1000  # 连接上后到开始传输时的时间
    starttransafe_time = _Curl.getinfo(_Curl.STARTTRANSFER_TIME) * 1000  # 接收到第一个字节的时间
    redirect_time = _Curl.getinfo(_Curl.REDIRECT_TIME) * 1000  # 重定向时间
    total_time = _Curl.getinfo(_Curl.TOTAL_TIME) * 1000 # 请求总时间 毫秒
    download_size = _Curl.getinfo(_Curl.SIZE_DOWNLOAD)  # 下载数据包大小
    download_speed = _Curl.getinfo(_Curl.SPEED_DOWNLOAD)  # 下载速度
    buf.close()
    print(result)
    print(http_code, dns_time, connect_time, pretransafe_time, starttransafe_time, redirect_time, total_time,
          download_size, download_speed)
    _Curl.close()

"""
get请求测试
header = [
‘Content-Type:application/json; charset=utf-8‘
]

fetch_api(‘http://xxxx‘, ‘get‘, header)

post请求测试
header = [

‘Content-Type:application/json; charset=utf-8‘

]

data = {"userid":"%s","nickname":"zxpp","sex":"1","birthday":"2015-01-01"}
json_data = json.dumps(data)
fetch_api(‘http://xxxx‘, ‘post‘, header=header,data=json_data)

"""

@asyncio.coroutine
def fetch_async(func, url, method, header=None, data=None):
loop = asyncio.get_event_loop()
future = loop.run_in_executor(None, func, url, method,header,data)
#response = yield from future # 因为 func 函数未返回数据,所以可以不用返回值
yield from future

header = [

‘Content-Type:application/json; charset=utf-8‘

]

data = {"userid":"%s","nickname":"zxpp","sex":"1","birthday":"2015-01-01"}
json_data = json.dumps(data)

tasks = [
fetch_async(fetch_api, url=‘http://xxxx‘, method=‘get‘,header=header),
fetch_async(fetch_api, url=‘http://xxxx‘, method=‘post‘, header=header,data=json_data)
]

loop = asyncio.get_event_loop()
results = loop.run_until_complete(asyncio.gather(*tasks))
loop.close()