Python抓取糗事百科成人版图片
1 #coding: utf-8
2 import urllib2
3 import urllib
4 import re
5 import os
6
7 #糗百成人版抓取图片
8 class QBAdult:
9
10 #类初始化
11 def __init__(self):
12 = "/"
13 #获取索引页面内容
14 def getPage(self, pageIndex):
15 url = + "index_" + str(pageIndex) + ".html"
16 req_header = {'User-Agent':'Mozilla/5.0 (Windows NT 6.1) AppleWebKit/537.11 (KHTML, like Gecko) Chrome/23.0.1271.64 Safari/537.11',
17 'Accept':'text/html;q=0.9,*/*;q=0.8',
18 'Accept-Charset':'ISO-8859-1,utf-8;q=0.7,*;q=0.3',
19 'Accept-Encoding':'gzip',
20 'Connection':'close',
21 'Referer':None #注意如果依然不能抓取的话,这里可以设置抓取网站的host
22 }
23 req_timeout = 5
24 request = (url, None, req_header)
25 reponse = (request, None, req_timeout)
26 #print ().decode('gbk')
27 return ().decode('gbk')
28
29 #获取页面所有条目信息,list格式
30 def getContents(self, pageIndex):
31 page = (pageIndex)
32 pattern = ('div class="ui-module".*?<a href=.*?>(.*?)</a>.*?<img.*?src="(.*?)"', )
33 items = (pattern, page)
34 #for item in items:
35 #print item[0], item[1]
36 return items
37
38
39 #获取页面所有图片
40 def getAllImgs(self, items):
41
42 images = []
43 for item in items:
44 (item[1])
45 return images
46
47
48 #保存多张图片
49 def saveAllImg(self, images, n, name):
50 number = 1
51 print u"--------------正在保存第", n , u"页美女图片-------------"
52 for imageURL in images:
53 splitPath = ('.')
54 fTail = ()
55 fileName = name + "/" + str(n) + "-" + str(number) + "." + fTail
56 (imageURL, fileName)
57 number += 1
58
59
60 #传入图片地址,文件名,保存单张图片
61 def saveImg(self, imageURL, fileName):
62 u = (imageURL)
63 data = ()
64 f = open(fileName, 'wb')
65 (data)
66 print u"正在保存美女图片", fileName
67 ()
68
69 #创建新目录
70 def mkdir(self, path):
71 path = ()
72 #判断是否存在
73 isExists = (path)
74 if not isExists:
75 print u"新建名字为", path, u"的文件夹"
76 (path)
77 return True
78 else:
79 #如果目录存在则不创建,并提示目录已存在
80 print u"名为", path, u"的文件夹已经创建成功"
81 return False
82
83
84 def savePageInfos(self, start, end):
85 fileName = "beautifull"
86 #获取一页内容,存入beautifull文件夹
87 (fileName)
88 for i in range(start, end+1):
89 print u"正在保存第", i, "页的美女图片"
90 contents = (i)
91 images = (contents)
92 (images, i, fileName)
93
94
95 qAdult = QBAdult()
96 (1, 301)