1)urllib2+BeautifulSoup抓取Goolge搜索链接

近期,参与的项目需要对Google搜索结果进行处理,之前学习了Python处理网页相关的工具。实际应用中,使用了urllib2和beautifulsoup来进行网页的抓取,但是在抓取google搜索结果的时候,发现如果是直接对google搜索结果页面的源代码进行处理,会得到很多“脏”链接。

看下图为搜索“titanic james”的结果:

【后端开发】Python多线程抓取Google搜索链接网页

图中红色标记的是不需要的,蓝色标记的是需要抓取处理的。

这种“脏链接”当然可以通过规则过滤的方法来过滤掉,但是这样程序的复杂度就高了。正当自己愁眉苦脸的正在写过滤规则时。同学提醒说google应该提供相关的api,才恍然大明白。

(2)Google Web Search API+多线程

文档中给出使用Python进行搜索的例子:

图中红色标记的是不需要的,蓝色标记的是需要抓取处理的。

这种“脏链接”当然可以通过规则过滤的方法来过滤掉,但是这样程序的复杂度就高了。正当自己愁眉苦脸的正在写过滤规则时。同学提醒说google应该提供相关的api,才恍然大明白。

(2)Google Web Search API+多线程

文档中给出使用Python进行搜索的例子:

import simplejson
    
# The request also includes the userip parameter which provides the end 
# user's IP address. Doing so will help distinguish this legitimate 
# server-side traffic from traffic which doesn't come from an end-user. 
url = ('https://ajax.googleapis.com/ajax/services/search/web'
       '?v=1.0&q=Paris%20Hilton&userip=USERS-IP-ADDRESS')
    
request = urllib2.Request(
    url, None, {'Referer': /* Enter the URL of your site here */})
response = urllib2.urlopen(request)
    
# Process the JSON string. 
results = simplejson.load(response)
# now have some fun with the results...
  
import simplejson
   
# The request also includes the userip parameter which provides the end
# user's IP address. Doing so will help distinguish this legitimate
# server-side traffic from traffic which doesn't come from an end-user.
url = ('https://ajax.googleapis.com/ajax/services/search/web'
       '?v=1.0&q=Paris%20Hilton&userip=USERS-IP-ADDRESS')
   
request = urllib2.Request(
    url, None, {'Referer': /* Enter the URL of your site here */})
response = urllib2.urlopen(request)
   
# Process the JSON string.
results = simplejson.load(response)
# now have some fun with the results..

实际应用中可能需要抓取google的很多网页,所以还需要使用多线程来分担抓取任务。使用google web search api的参考详细介绍,请看此处(这里介绍了Standard URL Arguments)。另外要特别注意,url中参数rsz必须是8(包括8)以下的值,若大于8,会报错的!

(3)代码实现

代码实现还存在问题,但是能够运行,鲁棒性差,还需要进行改进,希望各路大神指出错误(初学Python),不胜感激。

#-*-coding:utf-8-*- 
import urllib2,urllib
import simplejson
import os, time,threading
   
import common, html_filter
#input the keywords 
keywords = raw_input('Enter the keywords: ')                                 
   
#define rnum_perpage, pages 
rnum_perpage=8
pages=8                       
   
#定义线程函数 
def thread_scratch(url, rnum_perpage, page):
 url_set = [] 
 try:
   request = urllib2.Request(url, None, {'Referer': 'http://www.sina.com'})
   response = urllib2.urlopen(request)
   # Process the JSON string. 
   results = simplejson.load(response)
   info = results['responseData']['results']
 except Exception,e:
   print 'error occured'
   print e
 else:
   for minfo in info:
      url_set.append(minfo['url'])
      print minfo['url']
  #处理链接 
 i = 0
 for u in url_set:
   try:
     request_url = urllib2.Request(u, None, {'Referer': 'http://www.sina.com'})
     request_url.add_header(
     'User-agent',
     'CSC'
     )
     response_data = urllib2.urlopen(request_url).read()
     #过滤文件 
     #content_data = html_filter.filter_tags(response_data) 
     #写入文件 
     filenum = i+page
     filename = dir_name+'/related_html_'+str(filenum)
     print '  write start: related_html_'+str(filenum)
     f = open(filename, 'w+', -1)
     f.write(response_data)
     #print content_data 
     f.close()
     print '  write down: related_html_'+str(filenum)
   except Exception, e:
     print 'error occured 2'
     print e
   i = i+1
 return
   
#创建文件夹 
dir_name = 'related_html_'+urllib.quote(keywords)
if os.path.exists(dir_name):
   print 'exists  file'
   common.delete_dir_or_file(dir_name)
os.makedirs(dir_name)
   
#抓取网页 
print 'start to scratch web pages:'
for x in range(pages):
  print "page:%s"%(x+1)
  page = x * rnum_perpage
  url = ('https://ajax.googleapis.com/ajax/services/search/web'
                  '?v=1.0&q=%s&rsz=%s&start=%s') % (urllib.quote(keywords), rnum_perpage,page)
  print url
  t = threading.Thread(target=thread_scratch, args=(url,rnum_perpage, page))
  t.start()
#主线程等待子线程抓取完 
main_thread = threading.currentThread()
for t in threading.enumerate():
  if t is main_thread:
    continue
  t.join()