Python爬虫 Day 1.1

'''
POST 请求自动登陆github:
     请求URL：
     https://github.com/session
     请求方式：
     POST
     请求头:
     Cookie
     User-Agent: Mozilla/5.0 (Windows NT 10.0; WOW64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/70.0.3538.25 Safari/537.36 Core/1.70.3704.400 QQBrowser/10.4.3587.400
     请求体：
        commit: Sign in
utf8: ✓
authenticity_token: SFckS0VpuFLaTQ2oEb3YkoeZBymkzqgumq4IJfRODp8Syrk/u+KruBFS6SbwnrKhaF0E6YBZ1IUs+aVWsZ8lWg==
login: *******
password: *******
webauthn-support: supported
'''

# 1.获取token随机字符串
'''
    请求URL

    请求方式
        GET

    请求头
        COOKIES
        User-Agent:
2.解析并提取token字符串
#正则
<input type="hidden" name="authenticity_token" vale="(x*?)"/>
'''
import requests
import re
login_url = 'https://github.com/login'
# login页面的请求头信息
login_header = {'User-Agent':'Mozilla/5.0 (Windows NT 10.0; WOW64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/70.0.3538.25 Safari/537.36 Core/1.70.3704.400 QQBrowser/10.4.3587.400'}
login_res = requests.get(url=login_url, headers=login_header)
# print(login_res.text)
# 解析提取token字符串
authenticity_token = re.findall('<input type="hidden" name="authenticity_token" value="(.*?)" /> ',login_res.text,re.S)[0]
print(authenticity_token)

# 获取login页面的cookies信息
print(type(login_res.cookies))
print(type(login_res.cookies.get_dict()))
login_cookies = login_res.cookies.get_dict()
# 2.开始登录github
'''
POST请求自动登录github:
    请求URL：
    https://github.com/session
    请求方式：
        POST
    请求头：
        Cookies
        User-Agent:Mozilla/5.0 (Windows NT 10.0; WOW64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/70.0.3538.25 Safari/537.36 Core/1.70.3704.400 QQBrowser/10.4.3587.400
    请求体：
        commit: Sign in
utf8: ✓
authenticity_token: SFckS0VpuFLaTQ2oEb3YkoeZBymkzqgumq4IJfRODp8Syrk/u+KruBFS6SbwnrKhaF0E6YBZ1IUs+aVWsZ8lWg==
login: *******
password: *******
webauthn-support: supported
'''
# session登录url
session_url = 'https://github.com/session'
# 请求头信息
session_headers = {'User-Agent':'Mozilla/5.0 (Windows NT 10.0; WOW64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/70.0.3538.25 Safari/537.36 Core/1.70.3704.400 QQBrowser/10.4.3587.400'}
# 请求体信息
form_data = {"commit": "Sign in",
"utf8": "✓",
"authenticity_token": authenticity_token,
"login": "******",
"password": "******",
"webauthn-support": "supported"}
session_res = requests.post(url=session_url,
                            headers=session_headers,
                            cookies=login_cookies,
                            data=form_data
                            )
with open('github3.html','w',encoding='utf-8') as f:
    f.write(session_res.text)