某班网页课群自动答题,爬取题库(1)

 

from selenium import webdriver
from lxml import etree
import requests
import time
import random
import json

from selenium.webdriver.remote.webelement import WebElement

headers = {
    "user-agent": "Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) "
                  "Chrome/90.0.4430.93 Safari/537.36 Edg/90.0.818.56 "
}
browser = webdriver.Edge(executable_path = "MicrosoftWebDriver.exe")
# 让浏览器发起一个指定url的请求
browser.get("https://www.yooc.me/login")
# 定位标签
account_input = browser.find_element_by_xpath('/html/body/div[2]/div/div/div[2]/div/div[1]/form/div[2]/input')
account_input.send_keys('你的账号')
password_input = browser.find_element_by_xpath('/html/body/div[2]/div/div/div[2]/div/div[1]/form/div[3]/input')
password_input.send_keys('你的密码')
# 用page_source获取当前页面的源码数据
response = browser.page_source
tree = etree.H