python实现多线程采集的2个代码例子

代码一：

#!/usr/bin/python
# -*- coding: utf-8 -*-
#encoding=utf-8
 
import threading
import queue
import sys
import urllib2
import re
import mysqldb
 
#
# 数据库变量设置
#
db_host = '127.0.0.1'
db_user = "xxxx"
db_passwd = "xxxxxxxx"
db_name = "xxxx"
 
#
# 变量设置
#
thread_limit = 3
jobs = queue.queue(5)
singlelock = threading.lock()
info = queue.queue()
 
def workerbee(inputlist):
    for x in xrange(thread_limit):
        print 'thead {0} started.'.format(x)
        t = spider()
        t.start()
    for i in inputlist:
        try:
            jobs.put(i, block=true, timeout=5)
        except:
            singlelock.acquire()
            print "the queue is full !"
            singlelock.release()
 
    # wait for the threads to finish
    singlelock.acquire()        # acquire the lock so we can print
    print "waiting for threads to finish."
    singlelock.release()        # release the lock
    jobs.join()              # this command waits for all threads to finish.
    # while not jobs.empty():
    #   print jobs.get()
 
def gettitle(url,time=10):
    response = urllib2.urlopen(url,timeout=time)
    html = response.read()
    response.close()
    reg = r'<title>(.*?)</title>'
    title = re.compile(reg).findall(html)
    # title = title[0].decode('gb2312','replace').encode('utf-8')
    title = title[0]
    return title
 
class spider(threading.thread):
    def run(self):
        while 1:
            try:
                job = jobs.get(true,1)
                singlelock.acquire()
                title = gettitle(job[1])
                info.put([job[0],title], block=true, timeout=5)
                # print 'this {0} is {1}'.format(job[1],title)
                singlelock.release()
                jobs.task_done()
            except:
                break;
 
if __name__ == '__main__':
    con = none
    urls = []
    try:
        con = mysqldb.connect(db_host,db_user,db_passwd,db_name)
        cur = con.cursor()
        cur.execute('select id,url from `table_name` where `status`=0 limit 10')
        rows = cur.fetchall()
        for row in rows:
            # print row
            urls.append([row[0],row[1]])
        workerbee(urls)
        while not info.empty():
            print info.get()
    finally:
        if con:
            con.close()

代码二：

#!/usr/bin/python
# -*- coding: utf-8 -*-
#encoding=utf-8
#filename:robot.py
 
import threading,queue,sys,urllib2,re
#
# 变量设置
#
thread_limit = 3        #设置线程数
jobs = queue.queue(5)      #设置队列长度
singlelock = threading.lock()    #设置一个线程锁,避免重复调用
 
urls = ['http://games.sina.com.cn/w/n/2013-04-28/1634703505.shtml','http://games.sina.com.cn/w/n/2013-04-28/1246703487.shtml','http://games.sina.com.cn/w/n/2013-04-28/1028703471.shtml','http://games.sina.com.cn/w/n/2013-04-27/1015703426.shtml','http://games.sina.com.cn/w/n/2013-04-26/1554703373.shtml','http://games.sina.com.cn/w/n/2013-04-26/1512703346.shtml','http://games.sina.com.cn/w/n/2013-04-26/1453703334.shtml','http://games.sina.com.cn/w/n/2013-04-26/1451703333.shtml','http://games.sina.com.cn/w/n/2013-04-26/1445703329.shtml','http://games.sina.com.cn/w/n/2013-04-26/1434703322.shtml','http://games.sina.com.cn/w/n/2013-04-26/1433703321.shtml','http://games.sina.com.cn/w/n/2013-04-26/1433703320.shtml','http://games.sina.com.cn/w/n/2013-04-26/1429703318.shtml','http://games.sina.com.cn/w/n/2013-04-26/1429703317.shtml','http://games.sina.com.cn/w/n/2013-04-26/1409703297.shtml','http://games.sina.com.cn/w/n/2013-04-26/1406703296.shtml','http://games.sina.com.cn/w/n/2013-04-26/1402703292.shtml','http://games.sina.com.cn/w/n/2013-04-26/1353703286.shtml','http://games.sina.com.cn/w/n/2013-04-26/1348703284.shtml','http://games.sina.com.cn/w/n/2013-04-26/1327703275.shtml','http://games.sina.com.cn/w/n/2013-04-26/1239703265.shtml','http://games.sina.com.cn/w/n/2013-04-26/1238703264.shtml','http://games.sina.com.cn/w/n/2013-04-26/1231703262.shtml','http://games.sina.com.cn/w/n/2013-04-26/1229703261.shtml','http://games.sina.com.cn/w/n/2013-04-26/1228703260.shtml','http://games.sina.com.cn/w/n/2013-04-26/1223703259.shtml','http://games.sina.com.cn/w/n/2013-04-26/1218703258.shtml','http://games.sina.com.cn/w/n/2013-04-26/1202703254.shtml','http://games.sina.com.cn/w/n/2013-04-26/1159703251.shtml','http://games.sina.com.cn/w/n/2013-04-26/1139703233.shtml']
 
def workerbee(inputlist):
  for x in xrange(thread_limit):
    print 'thead {0} started.'.format(x)
    t = spider()
    t.start()
  for i in inputlist:
    try:
      jobs.put(i, block=true, timeout=5)
    except:
      singlelock.acquire()
      print "the queue is full !"
      singlelock.release()
 
  # wait for the threads to finish
  singlelock.acquire()    # acquire the lock so we can print
  print "waiting for threads to finish."
  singlelock.release()    # release the lock
  jobs.join()       # this command waits for all threads to finish.
  # while not jobs.empty():
  #  print jobs.get()
 
def gettitle(url,time=10):
  response = urllib2.urlopen(url,timeout=time)
  html = response.read()
  response.close()
  reg = r'<title>(.*?)</title>'
  title = re.compile(reg).findall(html)
  title = title[0].decode('gb2312','replace').encode('utf-8')
  return title
 
class spider(threading.thread):
  def run(self):
    while 1:
      try:
        job = jobs.get(true,1)
        singlelock.acquire()
        title = gettitle(job)
        print 'this {0} is {1}'.format(job,title)
        singlelock.release()
        jobs.task_done()
      except:
        break;
 
if __name__ == '__main__':
  workerbee(urls)