模拟登陆并爬取Github

import requests
from lxml import etree


class Login(object):
    def __init__(self):
        self.headers = {
            ‘Referer‘: ‘https://github.com/‘,
            ‘User-Agent‘: ‘Mozilla/5.0 (Windows NT 10.0; WOW64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/57.0.2987.133 Safari/537.36‘,
            ‘Host‘: ‘github.com‘
        }
        #登陆地址
        self.login_url = ‘https://github.com/login‘
        #POST请求地址
        self.post_url = ‘https://github.com/session‘
        #使用session保持状态，并自动处理Cookies(在访问其他子网页时，可以保持登陆，爬取网页)
        self.session = requests.Session()

    def token(self):
        #获取网页数据
        response = self.session.get(self.login_url, headers=self.headers)
        #提取网页中我们需要的authenticity_token并返回
        selector = etree.HTML(response.text)
        token = selector.xpath(‘//input[@name="authenticity_token"]/@value‘)
        return token

    def login(self, email, password):
        post_data = {
            ‘commit‘: ‘Sign in‘,
            ‘utf8‘: ‘?‘,
            ‘authenticity_token‘: self.token(),
            ‘login‘: email,
            ‘password‘: password
        }
        #使用post方法模拟登陆
        response = self.session.post(self.post_url, data=post_data, headers=self.headers)
        #登陆正常，输出登陆后的网页代码，并将它存储带D盘github.txt
        if response.status_code == 200:
            print(response.text)
            with open(‘D:/github.txt‘, ‘w‘, encoding = ‘utf-8‘) as f:
                f.write(response.text)
        else:
            print("Error!!!")

if __name__ == "__main__":
    login = Login()
    login.login(email=‘1024593536@qq.com‘, password=‘password‘)#输入你自己的账户密码