selenium实现ip代理http认证

  • 原理
    selenium本身支持的ip不支持用户认证,想要认证用户代理ip, 本文的思想就是给浏览器装插件,也就是代理插件。
    其中插件的包来源于https://github.com/RobinDev/S...
    注意:文件路径自己修改
  • 代码如下:
import osimport reimport timeimport zipfilefrom selenium import webdriver# Chrome代理模板插件(https://github.com/RobinDev/Selenium-Chrome-HTTP-Private-Proxy)目录CHROME_PROXY_HELPER_DIR = '../cookies/Chrome-proxy-helper'# 存储自定义Chrome代理扩展文件的目录CUSTOM_CHROME_PROXY_EXTENSIONS_DIR = '../cookies/chrome-proxy-extensions'def get_chrome_proxy_extension(proxy):    """获取一个Chrome代理扩展,里面配置有指定的代理(带用户名密码认证)    proxy - 指定的代理,格式: username:password@ip:port    """    m = re.compile('([^:]+):([^\@]+)\@([\d\.]+):(\d+)').search(proxy)    if m:        # 提取代理的各项参数        username = m.groups()[0]        password = m.groups()[1]        ip = m.groups()[2]        port = m.groups()[3]        # 创建一个定制Chrome代理扩展(zip文件)        if not os.path.exists(CUSTOM_CHROME_PROXY_EXTENSIONS_DIR):            os.mkdir(CUSTOM_CHROME_PROXY_EXTENSIONS_DIR)        extension_file_path = os.path.join(CUSTOM_CHROME_PROXY_EXTENSIONS_DIR, '{}.zip'.format(proxy.replace(':', '_')))        if not os.path.exists(extension_file_path):            # 扩展文件不存在,创建            zf = zipfile.ZipFile(extension_file_path, mode='w')            zf.write(os.path.join(CHROME_PROXY_HELPER_DIR, 'manifest.json'), 'manifest.json')            # 替换模板中的代理参数            background_content = open(os.path.join(CHROME_PROXY_HELPER_DIR, 'background.js')).read()            background_content = background_content.replace('%proxy_host', ip)            background_content = background_content.replace('%proxy_port', port)            background_content = background_content.replace('%username', username)            background_content = background_content.replace('%password', password)            zf.writestr('background.js', background_content)            zf.close()        return extension_file_path    else:        raise Exception('Invalid proxy format. Should be username:password@ip:port')if __name__ == '__main__':    # 测试    options = webdriver.ChromeOptions()    # 添加一个自定义的代理插件(配置特定的代理,含用户名密码认证)    options.add_extension(get_chrome_proxy_extension(proxy='cj422w:cj422w@117.41.187.145:888'))    print(options)    driver = webdriver.Chrome(chrome_options=options)    # 访问一个IP回显网站,查看代理配置是否生效了    driver.get('http://httpbin.org/ip')    print(driver.page_source)    time.sleep(60)    driver.quit()