爬虫：模拟github登陆

根据崔庆才崔大大的又改了一下。

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/69.0.3497.100 Safari/537.36',
            'Host': 'github.com'
        }
        self.login_url = 'https://github.com/login'
        self.post_url = 'https://github.com/session'
        self.logined_url = 'https://github.com/settings/profile'
        self.session = requests.Session()

    def token(self):
        response = self.session.get(self.login_url, headers=self.headers)
        selector = etree.HTML(response.text)
        # print(response.text)
        # print((etree.tostring(selector)).decode('utf-8'))
        # 修改此处
        token = selector.xpath('//div[@id="login"]//input/@value')[0]
        # print(token)
        return token

    def login(self, email, password):
        post_data = {
             'commit': 'Sign in',
            'utf8': '✓',
            'authenticity_token': self.token(),
             'login': email,
             'password': password
        }

        response = self.session.post(self.post_url, data=post_data, headers=self.headers)
        if response.status_code == 200:
            # print(response.text)
            self.dynamics(response.text)

        response = self.session.get(self.logined_url, headers=self.headers)
        if response.status_code == 200:
            self.profile(response.text)

    def dynamics(self, html):
        selector = etree.HTML(html)
        # print((etree.tostring(selector)).decode('utf-8'))
        dynamics = selector.xpath('//div[contains(@class, "news")]//div[contains(@class, "alert")]')
        # 此处未有节点
        # print(dynamics)
        for item in dynamics:
            dynamic = ' '.join(item.xpath('.//div[@class="title"]//text()')).strip()
            # print('dynamic')
            print(dynamic)

    def profile(self, html):
        selector = etree.HTML(html)
        name = selector.xpath('//input[@id="user_profile_name"]/@value')[0]
        email = selector.xpath('//select[@id="user_profile_email"]/option[@value!=""]/text()')
        print(name, email)


if __name__ == "__main__":
    login = Login()
    login.login(email='1319144981@qq.com', password='***************')

运行结果

这个是requirements.txt

certifi==2020.6.20
chardet==3.0.4
cssselect==1.1.0
idna==2.10
lxml==4.5.2
PyMySQL==0.10.1
pyquery==1.4.1
redis==3.5.3
requests==2.24.0
urllib3==1.25.10