爬取百思不得姐图片--请大家帮忙指正

mongoole

1. #coding:utf-8
   
2. import urllib.request
   
3. import re
   
4. import os
   
5. 
   
6. num = int(input('请输入您需要爬的页数：\n'))
   
7. 
   
8. for page_num in range(0,num):
   
9.   
   
10.     url = 'http://www.budejie.com/pic/' + str(page_num)
    
11.     headers = {'User-Agent':'Mozilla/5.0 (Windows NT 6.1; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/65.0.3325.181 Safari/537.36'}
    
12.     req = urllib.request.Request(url,headers=headers)
    
13.     response = urllib.request.urlopen(req)
    
14.     html = response.read().decode('utf-8')
    
15.     # print(html)
    
16. 
    
17.     target_url = re.findall(r'data-original="(.+?)" title=*',html)
    
18. 
    
19.     file_path = '存放图片的path'
    
20. 
    
21.     os.chdir(file_path)
    
22.     os.mkdir(str(page_num))
    
23.     os.chdir(str(page_num))
    
24.     # print(os.getcwd())
    
25.     for img_url in target_url:
    
26.         img_name = img_url.split('.')[-2].split('/')[-1]
    
27.         img_type = img_url.split('.')[-1]
    
28.         img_finall = img_name + '.' + img_type
    
29.         urllib.request.urlretrieve(img_url,img_finall)
    
30.         print('%s\t已下载完成'%img_url)

复制代码