python制作ip代理池

爬取的代理源为西刺代理。

• 用xpath解析页面
• 用telnet来验证ip是否可用
• 把有效的ip写入到本地txt中。当然也可以写入到redis、mongodb中，也可以设置检测程序当代理池中的ip数不够（如：小于20个）时，启动该脚本来重新获取ip，本脚本的代码也要做相应的改变。

# !/usr/bin/env python
# -*- coding: utf-8 -*-
# @Version      ： 1.0
# @Time         ： 2018/10/23 上午10:40
# @Author       ： Yeoman
# @Description  ：   

import urllib.request
import lxml.etree
import telnetlib
import os

headers = {
    'User-Agent': 'Mozilla/5.0 (Macintosh; Intel Mac OS X 10_12_2) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/70.0.3538.67 Safari/537.36'
}


def get_proxy(page_num):
    # 获取页面
    req = urllib.request.Request('http://www.xicidaili.com/nn/{}'.format(page_num), headers=headers) # 构造request请求
    response = urllib.request.urlopen(req) # 发送请求
    html = response.read()
    html = html.decode('utf-8')
    # print(html)

    # 解析页面
    proxy_list = []
    selector = lxml.etree.HTML(html)
    rows = selector.xpath('//*[@id="ip_list"]//tr')
    rows_total = len(rows)
    row_xpath_head = '//*[@id="ip_list"]//tr['
    row_ip_xpath_tail = ']/td[2]/text()'
    row_port_xpath_tail = ']/td[3]/text()'
    for i in range(1, rows_total):
        ip_xpath = row_xpath_head + str(i+1) + row_ip_xpath_tail
        port_xpath = row_xpath_head + str(i+1) + row_port_xpath_tail
        ip = selector.xpath(ip_xpath)[0]
        port = selector.xpath(port_xpath)[0]
        ip_port = ip + ':' + port
        proxy_list.append(ip_port)

    return proxy_list


# 检测代理ip是否可用
def test_proxy_ip_port(proxy_ip_port):
    print('当前代理ip：{}'.format(proxy_ip_port))
    ip_port = proxy_ip_port.split(':')
    ip = ip_port[0]
    port = ip_port[1]
    # 用telnet来验证ip是否可用
    try:
        telnetlib.Telnet(ip, port, timeout=10)
    except:
        return False
    else:
        return True


# 把有效的ip写入本地
def write_ip(proxy_ip):
    with open('./ip.txt', 'a') as f:
        f.write(proxy_ip + '\n')


# 删除文件
def del_file():
    file_path = './ip.txt'
    if os.path.exists(file_path):
        os.remove(file_path)


def run():
    del_file()
    proxy_ip_port_list = []
    for i in range(1, 6):  # 前5页
        proxy_ip_port_list += get_proxy(i)
    for i in range(100):  # 一页有100条
        proxy_ip_port = proxy_ip_port_list[i]
        is_valid = test_proxy_ip_port(proxy_ip_port)
        print(is_valid)
        if is_valid:
            # 写入ip到本地
            write_ip(proxy_ip_port)


if __name__ == '__main__':
    run()