爬虫实例：多线程，多进程对网页的爬取

Posted 2020-08-06 上后谈爱情

采用多线程对韩寒的微博进行爬取，这个只是不需要进行模拟登陆的：

 1 #--coding:utf-8---
 2 #!/usr/bin/env python
 3 import urllib
 4 import os
 5 import re
 6 import time
 7 from threading import Thread
 8 from multiprocessing import Process
 9 
10 def downloadURL(urls,dirpath):
11     ##在之前中分装一个list
12     for url in urls:
13         if(len(url)>0):
14             content=urllib.urlopen(url).read()
15             ##采用os模块中IO接口写写html文档
16             if not os.path.exists(dirpath):
17                 os.makedirs(dirpath)
18             open(dirpath+r‘/‘+url[-26:],‘w‘).write(content)
19             
20 
21 def parseTarget(url):
22     root_url=url
23     urls=[]
24     ##这里得到的东西是每一篇文章的链接
25     content=urllib.urlopen(root_url).read()
26     
27     
28     pattern=r‘<a title="(.*?)" href="(.*?)">‘
29     
30     hrefs=re.findall(pattern,content)
31     
32     for href in hrefs:
33         #print href
34         urls.append(href[1])
35 
36     return urls
37 
38 def thread_or_process_job(n,thread_or_process,url_lists,job):
39     local_time=time.time()
40     ##args为前面函数的参数
41     Thread_or_Process=[thread_or_process(target=job,args=(url_lists[i],str(n)+thread_or_process.__name__)) for i in xrange(n)]
42         
43 
44     for t in Thread_or_Process:
45         t.start()
46         
47     for t in Thread_or_Process:
48         t.join()
49         
50     print n,thread_or_process.__name__," run job need ",time.time()-local_time
51 
52 if __name__==‘__main__‘:
53     t=time.time()
54     urls=[]
55     for i in xrange(5):
56         urls.extend(parseTarget(‘http://blog.sina.com.cn/s/articlelist_1191258123_0_‘+str(i+1)+‘.html‘))
57     url_len=len(urls)
58     print ‘urls_len:‘,len(urls)
59     
60     for n in[2,4,6,8]:
61         url_list=[]
62         url_split_len=url_len//n
63         ##将总的url进行分为多少段同时进行处理
64         for i in xrange(n):
65             if i==n-1:
66                 url_list.append(urls[i*url_split_len:url_len])
67             else:
68                 url_list.append(urls[i*url_split_len:(i+1)*url_split_len])
69             
70             thread_or_process_job(n,Thread,url_list,downloadURL)
71             thread_or_process_job(n,Process,url_list,downloadURL)
72     
73     print "All done in ",time.time()-t
74