74 lines
2.3 KiB
Python
74 lines
2.3 KiB
Python
|
import requests # 用来抓取网页的html源码
|
||
|
|
head='https://www.linovel.com'
|
||
|
|
# url = 'https://www.linovelib.com/novel/3207/163708.html'
|
||
|
|
url1 = 'https://www.linovelib.com/novel/2662/catalog'
|
||
|
|
url = 'https://www.linovelib.com/novel/2662/183423.html'
|
||
|
|
# url = 'https://img3.readpai.com/2/2662/183423/210186.jpg'
|
||
|
|
url0 = 'https://www.linovelib.com/'
|
||
|
|
cookie = "night=1"
|
||
|
|
header = {'User-Agent': "Mozilla/5.0 (iPhone; CPU iPhone OS 16_6 like Mac OS X) AppleWebKit/605.1.15 (KHTML, like Gecko) Version/16.6 Mobile/15E148 Safari/604.1", 'referer': head, 'Accept-Language':'zh-CN,zh;q=0.9,en;q=0.8,en-GB;q=0.7,en-US;q=0.6', 'cookie':cookie}
|
||
|
|
header = {'User-Agent': 'Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/87.0.4280.67 Safari/537.36 Edg/87.0.664.47', 'referer': head, 'Accept-Language':'zh-CN,zh;q=0.9,en;q=0.8,en-GB;q=0.7,en-US;q=0.6', 'cookie':cookie}
|
||
|
|
# session = requests.Session()
|
||
|
|
# req=session.get(url1, headers=header)
|
||
|
|
req=requests.get(url, headers=header)
|
||
|
|
print(req.text)
|
||
|
|
|
||
|
|
|
||
|
|
|
||
|
|
|
||
|
|
# from selenium import webdriver
|
||
|
|
# from selenium.webdriver.edge.options import Options
|
||
|
|
# import time
|
||
|
|
|
||
|
|
# options = Options()
|
||
|
|
# options.add_argument('--start-minimized')
|
||
|
|
|
||
|
|
# driver = webdriver.Edge(options = options)
|
||
|
|
# driver2 = webdriver.Edge(options = options)
|
||
|
|
# # exit(0)
|
||
|
|
# driver.get(url0)
|
||
|
|
# cookies = driver.get_cookies()
|
||
|
|
# print(driver.page_source)
|
||
|
|
# # driver.delete_all_cookies()
|
||
|
|
# driver2.get('https://www.linovelib.com/')
|
||
|
|
# # driver2.get(url)
|
||
|
|
# time.sleep(5)
|
||
|
|
# for cookie in cookies:
|
||
|
|
# driver2.add_cookie({'name': cookie['name'], 'value': cookie['value'], 'domain': cookie['domain']})
|
||
|
|
# driver2.get(url)
|
||
|
|
# print(driver2.page_source)
|
||
|
|
|
||
|
|
|
||
|
|
|
||
|
|
# cookies_dict = dict()
|
||
|
|
# for cookie in cookies:
|
||
|
|
# cookies_dict[cookie["name"]] = cookie["value"]
|
||
|
|
|
||
|
|
# # 将 cookie 设置到 requests 的 CookieJar 中
|
||
|
|
# requests_cookiejar = requests.utils.cookiejar_from_dict(cookies_dict)
|
||
|
|
# req=requests.get(url, headers=header, cookies=requests_cookiejar)
|
||
|
|
# print(req.text)
|
||
|
|
|
||
|
|
|
||
|
|
# driver.get(url)
|
||
|
|
# cookies = driver.get_cookies()
|
||
|
|
|
||
|
|
|
||
|
|
|
||
|
|
# 打印 Cookie
|
||
|
|
# for cookie in cookies:
|
||
|
|
# print(cookie["name"])
|
||
|
|
# print('*******************')
|
||
|
|
# print(cookie["value"])
|
||
|
|
# print('*******************')
|
||
|
|
|
||
|
|
|
||
|
|
# print(driver2.page_source)
|
||
|
|
# # driver.delete_all_cookies()
|
||
|
|
# driver.get(url)
|
||
|
|
# print(driver.page_source)
|
||
|
|
# import pdb
|
||
|
|
# pdb.set_trace()
|
||
|
|
|
||
|
|
# print(driver.element)
|