爬虫开发python工具包介绍

Posted 2021-01-09 163yun

本文来自网易云社区

作者：王涛

此处我们给出几个常用的代码例子，包括get,post(json,表单),带证书访问：
Get 请求

@gen.coroutine
def fetch_url():
    try:
        c = CurlAsyncHTTPClient()  # 定义一个httpclient
        myheaders = {
            "Host": "weixin.sogou.com",
            "Connection": "keep-alive",
            "Cache-Control": "max-age=0",
            "Upgrade-Insecure-Requests": "1",
            "User-Agent": "Mozilla/5.0 (Windows; U; Windows NT 6.1; en-US) AppleWebKit/532.5 (Khtml, like Gecko) Chrome/4.0.249.0 Safari/532.5 ",
            "Accept": "text/html,application/xhtml+xml,application/xml;q=0.9,image/webp,image/apng,*/*;q=0.8",
            "Accept-Encoding": "gzip, deflate",
            "Accept-Language": "zh-CN,zh;q=0.9,en;q=0.8"
        }
        url = "http://weixin.sogou.com/weixin?type=1&s_from=input&query=%E4%BA%BA%E6%B0%91%E6%97%A5%E6%8A%A5&ie=utf8&_sug_=n&_sug_type_="

        req = HTTPRequest(url=url, method="GET", headers=myheaders, follow_redirects=True, request_timeout=20, connect_timeout=10,
                          proxy_host="127.0.0.1",
                          proxy_port=8888)
        response = yield c.fetch(req)  # 发起请求
        print response.code
        print response.body
        IOLoop.current().stop()  # 停止ioloop线程
    except:
        print traceback.format_exc()

Fiddler 抓到的报文请求头：

POST  JSON数据请求

@gen.coroutine
def fetch_url():
    """抓取url"""
    try:
        c = CurlAsyncHTTPClient()  # 定义一个httpclient
        myheaders = {
            "Host": "weixin.sogou.com",
            "Connection": "keep-alive",
            "Cache-Control": "max-age=0",
            "Upgrade-Insecure-Requests": "1",
            "User-Agent": "Mozilla/5.0 (Windows; U; Windows NT 6.1; en-US) AppleWebKit/532.5 (KHTML, like Gecko) Chrome/4.0.249.0 Safari/532.5 ",
            "Accept": "text/html,application/xhtml+xml,application/xml;q=0.9,image/webp,image/apng,*/*;q=0.8",
            "Accept-Encoding": "gzip, deflate",
            "Content-Type": "Application/json",
            "Accept-Language": "zh-CN,zh;q=0.9,en;q=0.8"
        }
        url = "http://127.0.0.1?type=1&s_from=input&query=%E4%BA%BA%E6%B0%91%E6%97%A5%E6%8A%A5&ie=utf8&_sug_=n&_sug_type_="
        body =json.dumps({"key1": "value1", "key2": "value2"})  # Json格式数据

        req = HTTPRequest(url=url, method="POST", headers=myheaders, follow_redirects=True, request_timeout=20, connect_timeout=10,
                          proxy_host="127.0.0.1",proxy_port=8888,body=body)
        response = yield c.fetch(req)  # 发起请求
        print response.code
        print response.body
        IOLoop.current().stop()  # 停止ioloop线程
    except:
        print traceback.format_exc()

Fiddler 抓到的报文请求头：

POST  Form表单数据请求

@gen.coroutine
def fetch_url():
    """抓取url"""
    try:
        c = CurlAsyncHTTPClient()  # 定义一个httpclient
        myheaders = {
            "Host": "weixin.sogou.com",
            "Connection": "keep-alive",
            "Cache-Control": "max-age=0",
            "Upgrade-Insecure-Requests": "1",
            "User-Agent": "Mozilla/5.0 (Windows; U; Windows NT 6.1; en-US) AppleWebKit/532.5 (KHTML, like Gecko) Chrome/4.0.249.0 Safari/532.5 ",
            "Accept": "text/html,application/xhtml+xml,application/xml;q=0.9,image/webp,image/apng,*/*;q=0.8",
            "Accept-Encoding": "gzip, deflate",
            # "Content-Type": "Application/json",
            "Accept-Language": "zh-CN,zh;q=0.9,en;q=0.8"
        }
        import urllib
        url = "http://127.0.0.1?type=1&s_from=input&query=%E4%BA%BA%E6%B0%91%E6%97%A5%E6%8A%A5&ie=utf8&_sug_=n&_sug_type_="
        body =urllib.urlencode({"key1": "value1", "key2": "value2"})  # 封装form表单

        req = HTTPRequest(url=url, method="POST", headers=myheaders, follow_redirects=True, request_timeout=20, connect_timeout=10,
                          proxy_host="127.0.0.1",proxy_port=8888,body=body)
        response = yield c.fetch(req)  # 发起请求
        print response.code
        print response.body
        IOLoop.current().stop()  # 停止ioloop线程
    except:
        print traceback.format_exc()

Fiddler 抓到的报文请求头：

添加证书访问  

def fetch_url():
    """抓取url"""
    try:
        c = CurlAsyncHTTPClient()  # 定义一个httpclient
        myheaders = {
            "Host": "www.amazon.com",
            "Connection": "keep-alive",
            "Cache-Control": "max-age=0",
            "Upgrade-Insecure-Requests": "1",
            "User-Agent": ("Mozilla/5.0 (Windows NT 10.0; Win64; x64) "
                   "AppleWebKit/537.36 (KHTML, like Gecko) "
                   "Chrome/68.0.3440.106 Safari/537.36"),
            "Accept": ("text/html,application/xhtml+xml,"
               "application/xml;q=0.9,image/webp,image/apng,*/*;q=0.8"),
            "Accept-Encoding": "gzip, deflate, br",
            "Accept-Language": "zh-CN,zh;q=0.9,en;q=0.8"
        }
        import urllib
        url = "https://www.amazon.com/"

        req = HTTPRequest(url=url, method="GET", headers=myheaders, follow_redirects=True, request_timeout=20, connect_timeout=10,proxy_host="127.0.0.1",
        proxy_port=8888,ca_certs="FiddlerRoot.pem")  # 绑定证书
        response = yield c.fetch(req)  # 发起请求
        print response.code
        print response.body
        IOLoop.current().stop()  # 停止ioloop线程
    except:
        print traceback.format_exc()

Fiddler抓到的报文（说明可以正常访问）

四、总结

抓取量少的时候，建议使用requests,简单易用。
  并发量大的时候，建议使用tornado，单线程高并发，高效易编程。

以上给出了requests和Fiddler中常用的接口和参数说明，能解决爬虫面对的大部分问题，包括并发抓取、日常的反爬应对，https网站的抓取。

附上一段我自己的常用抓取代码逻辑：

import randomfrom tornado.ioloop import IOLoopfrom tornado import genfrom tornado.queues import Queue


import random
from tornado.ioloop import IOLoop
from tornado import gen
from tornado.queues import Queue


TASK_QUE = Queue(maxsize=1000)


def response_handler(res):
    """ 处理应答，一般会把解析的新的url添加到任务队列中，并且解析出目标数据 """
    pass


@gen.coroutine
def url_fetcher_without_param():
    pass


@gen.coroutine
def url_fetcher(*args,**kwargs):
    global TASK_QUE
    c = CurlAsyncHTTPClient()

    while 1:
        #console_show_log("Let‘s spider")
        try: 
            param = TASK_QUE.get(time.time() + 300) # 5 分钟超时
        except tornado.util.TimeoutError::
            yield gen.sleep(random.randint(10,100))
            continue

        try:
            req = HTTPRequest(url,method=,headers=,....) # 按需配置参数
            response = yield c.fetch(req) 
            if response.coe==200:
                response_handler(response.body)
        except Exception:
            yield gen.sleep(10)
            continue
        finally:
            print "I am a slow spider"
            yield gen.sleep(random.randint(10,100))

@gen.coroutine
def period_callback():
    pass

def main():
    io_loop = IOLoop.current()
    # 添加并发逻辑1
    io_loop.spawn_callback(url_fetcher, 1)  
    io_loop.spawn_callback(url_fetcher, 2)
    io_loop.spawn_callback(url_fetcher_without_param) # 参数是可选的

    # 如果需要周期调用，调用PeriodicCallback：
    PERIOD_CALLBACK_MILSEC = 10  # 10, 单位ms
    io_loop.PeriodicCallback(period_callback,).start()
    io_loop.start()

if __name__ == "__main__":
    main()

以上，欢迎讨论交流

五、参考：

1. requests快速入门：http://docs.python-requests.org/zh_CN/latest/user/quickstart.html

2. requests高级应用：http://docs.python-requests.org/en/master/user/advanced/

3. 什么是CA_BUNDLE:https://www.namecheap.com/support/knowledgebase/article.aspx/986/69/what-is-ca-bundle

4. 如何用requests下载图片：https://stackoverflow.com/questions/13137817/how-to-download-image-using-requests

5. tornado AsyncHttpClient: https://www.tornadoweb.org/en/stable/httpclient.html

6. 100 Continue状态码：https://developer.mozilla.org/zh-CN/docs/Web/HTTP/Status/100

7. HTTP认证： https://developer.mozilla.org/en-US/docs/Web/HTTP/Authentication

8. 证书转换： https://www.alibabacloud.com/help/zh/faq-detail/40526.htm

网易云免费体验馆，0成本体验20+款云产品！ 

更多网易研发、产品、运营经验分享请访问网易云社区。

相关文章：
【推荐】 知物由学 | 广告欺诈：如何应对数字广告里分羹者？
【推荐】 分布式存储系统可靠性如何估算？
【推荐】 【工程实践】服务器数据解析