python3----练习题(爬取电影天堂资源)

Posted 2020-10-22 jonm

 1 import requests
 2 import re
 3 
 4 url = ‘http://www.ygdy8.net/html/gndy/dyzz/list_23_{}.html‘
 5 for n in range(1, 2):
 6     new_url = url.format(n)
 7     html_1 = requests.get(new_url)
 8     html_1.encoding = ‘gb2312‘
 9     detil_list = re.findall(‘<a href="(.*?)" class="ulink">‘, html_1.text)
10 
11     for m in detil_list:
12         b_url = ‘http://www.ygdy8.net‘ + m
13         html_2 = requests.get(b_url)
14         html_2.encoding = ‘gb2312‘
15         ftp = re.findall(‘<a href="(.*?)">.*?</a></td>‘, html_2.text)
16         with open(‘tddy.txt‘, ‘a‘, encoding=‘utf-8‘) as f:
17             f.write(ftp[0] + ‘\n‘)