爬图mmjpg无法保存图片

darenhp

直接上代码

1. import urllib.request
   
2. import urllib.parse
   
3. import json
   
4. import os
   
5. 
   
6. def url_open(url):
   
7.     req = urllib.request.Request(url)
   
8.     req.add_header('User-Agent','Mozilla/5.0 (Linux; Android 6.0; Nexus 5 Build/MRA58N) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/66.0.3359.181 Mobile Safari/537.36')
   
9.     response = urllib.request.urlopen(url)
   
10.     html = response.read()
    
11.     return html
    
12.    
    
13. def get_page(url):
    
14.     html = url_open(url).decode('utf-8')
    
15.     a = html.find('href="/home/')+12
    
16.     b = html.find('"',a)
    
17.     return html[a:b]
    
18. 
    
19. def find_urls(url):
    
20.     html = url_open(url).decode('utf-8')
    
21.     url_addrs = []
    
22. 
    
23.     a = html.find('http://img.mmjpg.com/small/2018/')
    
24.     while a != -1:
    
25.         b = html.find('.jpg',a,a+255)
    
26.         if b != -1:
    
27.             if b-a == 36:#返回套图第一页位置
    
28.                 url_addrs.append('http://www.mmjpg.com/mm/'+str(html[a+32:b]))
    
29.         else:
    
30.             b = a + 32
    
31.         a = html.find('http://img.mmjpg.com/small/2018/',b)
    
32.     for each in url_addrs:
    
33.         print (each)
    
34.         html = url_open(each).decode('utf-8')
    
35.         img_addrs = []
    
36.         i=1
    
37.         page6 = html.find('/6')
    
38.         page = html.find('href="/mm/',page6+4)
    
39.         strpages = str(html[page+15:page+17])
    
40.         print(page)
    
41.         print (strpages)
    
42.         pages = int(html[page+15:page+17])
    
43.         #返回套图总页数
    
44.         print (pages)
    
45.         while i != pages:
    
46.             i=i+1
    
47.             html = url_open(each + '/' + str(i)).decode('utf-8')
    
48.             a = html.find('img src=')
    
49.             b = html.find('.jpg',a,a+255)
    
50.             if b != -1:
    
51.                 img_addrs.append(html[a+9:b+4])
    
52.             else:
    
53.                 b = a + 9
    
54.             a = html.find('img src=',b)
    
55.         for each2 in img_addrs:
    
56.             print(each2)
    
57.             filename = each2.split('/')[-1]
    
58.             with open(filename,'wb') as f:
    
59.                 img = url_open(each)
    
60.                 f.write(img)
    
61.    
    
62. 
    
63. 
    
64. def download_mm(folder='ooxx',pages=1):
    
65.     os.chdir(folder)
    
66. 
    
67.     url = "http://www.mmjpg.com/"
    
68.     page_num = int(get_page(url))
    
69.     for i in range(pages):
    
70.         page_num += 1
    
71.         page_url = url + 'home/' + str(page_num)
    
72.         url_addrs = find_urls(page_url)
    
73. 
    
74. 
    
75. if __name__ == "__main__":
    
76.     download_mm()

复制代码

代码有点乱，图片地址都能获取到，但是保存下来都是同一张无关的图片。
返回结果：
http://img.mmjpg.com/2018/1323/2iqw.jpg
http://img.mmjpg.com/2018/1323/3ikb.jpg
http://img.mmjpg.com/2018/1323/4iak.jpg
http://img.mmjpg.com/2018/1323/5ijf.jpg
http://img.mmjpg.com/2018/1323/6i0f.jpg
http://img.mmjpg.com/2018/1323/7i3u.jpg
http://img.mmjpg.com/2018/1323/8iqc.jpg
http://img.mmjpg.com/2018/1323/9i9y.jpg

darenhp

59行                 img = url_open(each)
改为                img = url_open(each2)