自己写了些关于抓取网易新闻的爬虫,发现其网页源代码与网页的评论根本就对不上,所以,
采用了抓包工具得到了其评论的隐藏地址(每个浏览器都有自己的抓包工具,都可以用来分析网站)
如果仔细观察的话就会发现,有一个特殊的,那么这个就是自己想要的了
然后打开链接就可以找到相关的评论内容了。(下图为第一页内容)
接下来就是代码了(也照着大神的改改写写了)。
#coding=utf-8
import urllib2
import re
import json
import time
class WY():
def __init__(self):
self.headers = {'User-Agent': 'Mozilla/5.0 (Windows NT 5.1) AppleWebKit/534.24 (KHTML, like '}
self.url='http://comment.news.163.com/data/news3_bbs/df/B9IBDHEH000146BE_1.html'
def getpage(self,page):
full_url='http://comment.news.163.com/cache/newlist/news3_bbs/B9IBDHEH000146BE_'+str(page)+'.html'
return full_url
def gethtml(self,page):
try:
req=urllib2.Request(page,None,self.headers)
response = urllib2.urlopen(req)
html = response.read()
return html
except urllib2.URLError,e:
if hasattr(e,'reason'):
print u"连接失败",e.reason
return None
#处理字符串
def Process(self,data,page):
if page == 1:
data=data.replace('var replyData=','')
else:
data=data.replace('var newPostList=','')
reg1=re.compile(" \[<a href=''>")
data=reg1.sub(' ',data)
reg2=re.compile('<\\\/a>\]')
data=reg2.sub('',data)
reg3=re.compile('<br>')
data=reg3.sub('',data)
return data
#解析json
def dealJSON(self):
with open("WY.txt","a") as file:
file.write('ID'+'|'+'评论'+'|'+'踩'+'|'+'顶'+'\n')
for i in range(1,12):
if i == 1:
data=self.gethtml(self.url)
data=self.Process(data,i)[:-1]
value=json.loads(data)
file=open('WY.txt','a')
for item in value['hotPosts']:
try:
file.write(item['1']['f'].encode('utf-8')+'|')
file.write(item['1']['b'].encode('utf-8')+'|')
file.write(item['1']['a'].encode('utf-8')+'|')
file.write(item['1']['v'].encode('utf-8')+'\n')
except:
continue
file.close()
print '--正在采集%d/12--'%i
time.sleep(5)
else:
page=self.getpage(i)
data = self.gethtml(page)
data = self.Process(data,i)[:-2]
# print data
value=json.loads(data)
# print value
file=open('WY.txt','a')
for item in value['newPosts']:
try:
file.write(item['1']['f'].encode('utf-8')+'|')
file.write(item['1']['b'].encode('utf-8')+'|')
file.write(item['1']['a'].encode('utf-8')+'|')
file.write(item['1']['v'].encode('utf-8')+'\n')
except:
continue
file.close()
print '--正在采集%d/12--'%i
time.sleep(5)
if __name__ == '__main__':
WY().dealJSON()
以上就是我爬取的代码了。