免费代理ip爬虫分享

liuguangbiao 2021-12-14 原文

分享一个某代理网站的免费代理ip的爬虫，直接复制到pycharm运行就可以了。

注意：爬取的代理ip有点坑，因为是免费的所以过期时间很快，可能1分钟后就会失效。并且在scrapy使用这些代理ip还会给你打印一堆广告。且用且珍惜。

import requests
from lxml import etree
import json

class XiciProxiesSpider(object):

    def __init__(self):
        self.start_url = \'http://www.xicidaili.com/nn\'
        self.headers = {
            \'User-Agent\': \'Mozilla/5.0 (Macintosh; Intel Mac OS X 10_13_6) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/67.0.3396.62 Safari/537.36\'
        }

    def get_page_from_url(self, url):
        response = requests.get(url,headers=self.headers)
        return response.content.decode()

    def get_data_from_page(self, page):
        # print(page)
        # 把page转换为Element对象
        html = etree.HTML(page)
        # 获取包含代理信息的tr列表
        trs = html.xpath(\'//*[@id="ip_list"]/tr\')[1:]

        # 遍历trs, 获取数据信息

        data = {
            \'http\':[],
            \'https\':[]
        }

        for tr in trs:
            try:
                ip = tr.xpath(\'./td[2]/text()\')[0]
                port = tr.xpath(\'./td[3]/text()\')[0]
                ip_type = tr.xpath(\'./td[6]/text()\')[0].lower()
                # 如果ip不是http或https直接返回
                if ip_type not in data.keys():
                    return
                # 构建代理数据
                item = {ip_type: \'{}://{}:{}\'.format(ip_type, ip, port)}
                # 检查代理IP是否可用, 如果可用添加到列表中
                if self.validate_ip(item, ip_type):
                    print(item[ip_type])
                    data[ip_type].append(item[ip_type])
            except Exception as ex:
                print(ex)
                print(etree.tostring(tr))

        # print(data)
        return data

    def validate_ip(self, item, ip_type):

        try:
            test_url = "{}://baidu.com".format(ip_type)
            response = requests.get(\'http://baidu.com\', proxies=item, timeout=2)
            if response.status_code == 200:
                return True
            return False
        except Exception as ex:
            return False

    def save_data(self, data):
        with open(\'proxies.json\', \'w\') as f:
            json.dump(data, f, indent=2)


    def run(self):
        # 获取页面内宽容
        page = self.get_page_from_url(self.start_url)
        # 获取可用代理IP
        data = self.get_data_from_page(page)
        # 保存数据
        self.save_data(data)

if __name__ == \'__main__\':
    fps = XiciProxiesSpider()
    fps.run()