网络爬虫（爬取网站图片，自动保存本地）

Posted 2020-11-10 glacial_water

事先申明一点，这个人品没有什么问题，只是朋友发一段python源码，再这里分享大家。

 1 import requests
 2 from lxml import html
 3 import os
 4 from multiprocessing.dummy import Pool as ThreadPool
 5 
 6 def header(referer):
 7     headers = {
 8         ‘Host‘: ‘i.meizitu.net‘,
 9         ‘Pragma‘: ‘no-cache‘,
10         ‘Accept-Encoding‘: ‘gzip, deflate‘,
11         ‘Accept-Language‘: ‘zh-CN,zh;q=0.8,en;q=0.6‘,
12         ‘Cache-Control‘: ‘no-cache‘,
13         ‘Connection‘: ‘keep-alive‘,
14         ‘User-Agent‘: ‘Mozilla/5.0 (Macintosh; Intel Mac OS X 10_12_5) AppleWebKit/537.36 (KHTML, like Gecko) ‘
15                       ‘Chrome/59.0.3071.115 Safari/537.36‘,
16         ‘Accept‘: ‘image/webp,image/apng,image/*,*/*;q=0.8‘,
17         ‘Referer‘: ‘{}‘.format(referer),
18     }
19     return headers
20 
21 # 获取主页列表
22 def getPage(pageNum):
23     baseUrl = ‘http://www.mzitu.com/page/{}‘.format(pageNum)
24     selector = html.fromstring(requests.get(baseUrl).content)
25     urls = []
26     for i in selector.xpath(‘//ul[@id="pins"]/li/a/@href‘):
27         urls.append(i)
28         print(i)
29     return urls
30 
31 
32 # 图片链接列表， 标题
33 # url是详情页链接
34 def getPiclink(url):
35     sel = html.fromstring(requests.get(url).content)
36     # 图片总数
37     total = sel.xpath(‘//div[@class="pagenavi"]/a[last()-1]/span/text()‘)[0]
38     # 标题
39     title = sel.xpath(‘//h2[@class="main-title"]/text()‘)[0]
40     # 文件夹格式
41     dirName = u"【{}P】{}".format(total, title)
42     # 新建文件夹
43     os.mkdir(dirName)
44 
45     n = 1
46     for i in range(int(total)):
47         # 每一页
48         try:
49             link = ‘{}/{}‘.format(url, i+1)
50             s = html.fromstring(requests.get(link).content)
51             # 图片地址在src标签中
52             jpgLink = s.xpath(‘//div[@class="main-image"]/p/a/img/@src‘)[0]
53             # print(jpgLink)
54             # 文件写入的名称：当前路径／文件夹／文件名
55             filename = ‘%s/%s/%s.jpg‘ % (os.path.abspath(‘.‘), dirName, n)
56             print(u‘开始下载图片:%s 第%s张‘ % (dirName, n))
57             with open(filename, "wb+") as jpg:
58                 jpg.write(requests.get(jpgLink, headers=header(jpgLink)).content)
59             n += 1
60         except:
61             pass
62 
63 
64 if __name__ == ‘__main__‘:
65     pageNum = input(u‘请输入页码：‘)
66     p = getPage(pageNum)
67     with ThreadPool(4) as pool:
68         pool.map(getPiclink, p)    

至于爬出出来的效果图，我就不发布了