文中用到了BeautifulSoup這個庫， 目的是處理html文檔分析的， 因為我只是提取了title的關鍵字，所以可以用正則表達式代替， 還有一個庫是jieba， 這個庫是中文分詞的作用， 再有一個庫是 chardet， 用來判斷字符的編碼， 本想多線程的， 但是自認為被搞糊涂了，就放棄了

代碼如下:
#coding:utf-8
import re
import urllib
import urllib2
import sys
import time
import Queue
import thread
import threading
import jieba
import chardet
from BeautifulSoup import BeautifulSoup as BS

DEEP = 1000
LOCK = threading.Lock()
PATH = "c://test//"
urlQueue = Queue.Queue()
def pachong():
 url = 'http://www.baidu.com'
 return url

def getPageUrl(html):
 reUrl = re.compile(r'</s*[Aa]{1}/s+[^>]*?[Hh][Rr][Ee][Ff]/s*=/s*[/"/']?([^>/"/']+)[/"/']?.*?>')
 urls = reUrl.findall(html)
 for url in urls:
  if len(url) > 10:
   if url.find('javascript') == -1:
    urlQueue.put(url)

def getContents(url):
 try:
  url = urllib2.quote(url.split('#')[0].encode('utf-8'), safe = "%/:=&?~#+!$,;'@()*[]")
  req = urllib2.urlopen(url)
  res = req.read()
  code = chardet.detect(res)['encoding']
  #print
  #print code
  res = res.decode(str(code), 'ignore')
  res = res.encode('gb2312', 'ignore')
  code = chardet.detect(res)['encoding']
  #print code
  #print res
  return res
 except urllib2.HTTPError, e:
  print e.code
  return None
 except urllib2.URLError, e:
  print str(e)
  return None

def writeToFile(html, url):
 fp = file(PATH + str(time.time()) + '.html', 'w')
 fp.write(html)
 fp.close()

 
def getKeyWords(html):
 code = chardet.detect(html)['encoding']
 if code == 'ISO-8859-2':
  html.decode('gbk', 'ignore').encode('gb2312', 'ignore')
 code = chardet.detect(html)['encoding']
 soup = BS(html, fromEncoding="gb2312")
 titleTag = soup.title
 titleKeyWords = titleTag.contents[0]
 cutWords(titleKeyWords)

def cutWords(contents):
 print contents
 res = jieba.cut_for_search(contents)
 res = '/n'.join(res)
 print res
 res = res.encode('gb2312')
 keyWords = file(PATH + 'cutKeyWors.txt', 'a')
 keyWords.write(res)
 keyWords.close()

def start():

 while urlQueue.empty() == False:
  url = urlQueue.get()
  html = getContents(url)
  getPageUrl(html)
  getKeyWords(html)
  #writeToFile(html, url)

  
if __name__ == '__main__':
 startUrl = pachong()
 urlQueue.put(startUrl)
 start()