python爬取代理IP地址

本篇以爬取爬取西刺代理为例，通过python爬取国内http代理IP及端口并存入csv文件。

抓取页面

我们通过urllib的urllib.request子模块来抓取页面信息

# 西刺国内HTTP代理第一页
url = 'https://www.xicidaili.com/wt'

user_agent = 'Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/77.0.3865.90 Safari/537.36'

# 请求头信息 `user_agent:用户代理`
headers = {
        'User-Agent': user_agent
    }

# 抓取页面信息
req = urllib.request.Request(url, headers=headers)
page = urllib.request.urlopen(req).read().decode('utf-8')

解析页面信息

我们使用BeautifulSoup来解析html，并通过正则表达式模块re来匹配我们需要的信息

• 我们查看西刺的html结构，可以发现需要获取的IP及端口都是在标签中的

# BeautifulSoup解析html
soup = BeautifulSoup(page, 'lxml')

# 正则
# 匹配ip
ips = re.findall('td>(\d+\.\d+\.\d+\.\d+)</td', str(soup))
# 匹配端口
ports = re.findall('td>(\d+)</td', str(soup))

存csv文件

最后将获取到的数据存入一个csv文件中，

#csv文件数据
csvdata = []

index = 0

while index < len(ips):
       _item = []
       _item.append(ips[index])
       _item.append(ports[index])
       csvdata.append(_item)
       index += 1

# csv文件title
headers_title = ['ip', 'port']

# 写入csv文件
csvPath = 'ipCsvData.csv'

# w模型写入文件
with open(csvPath, 'w', newline='')as f:
           f_csv = csv.writer(f)
           f_csv.writerow(headers_title)
           f_csv.writerows(csvdata)

完整代码

import urllib.request
import csv
import re
import os
from bs4 import BeautifulSoup

pageUrl = 'https://www.xicidaili.com/wt'
pageIndex = 1

# 获取前20页数据
pageCount = 20


# url爬取的网址，headers
# https://www.xicidaili.com/wt

def crawlerPage(pageUrl):
    url = pageUrl

    user_agent = 'Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/77.0.3865.90 Safari/537.36'
    headers = {
        'User-Agent': user_agent
    }

    # 抓取page信息
    req = urllib.request.Request(url, headers=headers)
    page = urllib.request.urlopen(req).read().decode('utf-8')

    # BeautifulSoup解析html
    soup = BeautifulSoup(page, 'lxml')

    # 正则
    # 匹配ip
    ips = re.findall('td>(\d+\.\d+\.\d+\.\d+)</td', str(soup))
    # 匹配端口
    ports = re.findall('td>(\d+)</td', str(soup))

    # 存csv文件数据
    csvdata = []

    index = 0

    while index < len(ips):
        _item = []
        _item.append(ips[index])
        _item.append(ports[index])
        csvdata.append(_item)
        index += 1

    headers_title = ['ip', 'port']

    # 写入csv文件

    csvPath = 'ipCsvData.csv'
    
    # 如果csv文件则原文件基础增加，否则创建新文件
    if (os.path.exists(csvPath)):
        with open(csvPath, 'a+', newline='')as f:
            f_csv = csv.writer(f)
            f_csv.writerows(csvdata)
    else:
        with open(csvPath, 'w', newline='')as f:
            f_csv = csv.writer(f)
            f_csv.writerow(headers_title)
            f_csv.writerows(csvdata)


while pageIndex <= pageCount:
    if (pageIndex == 1):
        crawlerPage(pageUrl)
        print('Page 1 get success!')
    else:
        crawlerPage(pageUrl + "/" + str(pageIndex))
        print('Page ' + str(pageIndex) + ' get success!')

    pageIndex += 1