python函数式实现的多线程抓取示例

#encoding: gb2312
import urllib2
import threading
import logging
import re
import sys
import os
from bs4 import BeautifulSoup
reload(sys)
sys.setdefaultencoding("utf-8")
#日志初始化
FILE = os.getcwd()
logging.basicConfig(filename=os.path.join(FILE, 'log.txt'),level=logging.DEBUG)
#待抓取的任务队列
url_new = [('none','http://www.111com.net/')]
#已完成的任务
url_old = []
#已完成的状态
url_err = {200:[]}
#锁
lock = threading.Lock()
lock2= threading.Lock()

#线程执行主方法
#从任务列表中获取一条url进行抓取
#分析url，去重复，将得到的urls重新放入任务列表
#保存当前url的访问状态
def geturl():
    global url_new
    try:
        while True:
            lock.acquire()
            if len(url_new)<=0:
                lock.release()
                continue
            url_t = url_new.pop(0)
            url = url_t[1]
            try:
                req = urllib2.urlopen(url)
            except urllib2.HTTPError, e:
                #记录到对应的列表中
                if url_err.has_key(e.code):
                    url_err[e.code].append((url,url_t[0]))
                else:
                    url_err[e.code] = [(url,url_t[0])]
                with open('log.html', 'a+') as f:
                        f.write(str(e.code)+'：'+url+', 来路：'+url_t[0]+'
')
                        continue
            else:
                url_err[200].append(url)
                with open('log.html', 'a+') as f:
                        f.write('200：'+url+', 来路：'+url_t[0]+'
')
           
            #记录到已访问的列表中
            url_old.append(url)
            #开始提取页面url
            soup = BeautifulSoup(req.read().decode('UTF-8', 'ignore'))
            alink= soup.find_all('a', attrs={'href':re.compile(".*?xxxxxx.*?")})
            tmp_url = []
            for a in alink:
                href = a.get('href')
                tmp_url.append(a.get('href') if a.get('href').find('http:')>=0 else 'http://www.xxxxxx.com'+a.get('href'))
            tmp_url= {}.fromkeys(tmp_url).keys()
            for link in tmp_url:
                if link not in url_old:
                    url_new.append((url, link))
            tmp = []
            for i in xrange(len(url_new)):
                if url_new[i][1] not in tmp:
                    tmp.append(url_new[i][1])
                else:
                    del url_new[i]
               
            #url_new = {}.fromkeys(url_new).keys()

            #输出一下状态信息
            os.system('cls')
            print threading.Thread().getName()+":当前线程数："+str(threading.activeCount())+"，当前剩余任务量："+str(len(url_new))+", 已访问:"+str(len(url_old))
            for k in url_err.keys():
                print str(k)+'：'+str(len(url_err[k]))

            lock.release()
    except Exception as e:
        logging.debug(str(e))
        lock.release()

#线程数检测 死循环持续检测当前活动线程数
#不够数量时自动创建启动新线程
def threadcheck(num):
    t=threading.Thread(target=geturl)
    t.start()
    t.join()

#定义主方法
def main():
    """初始 创建200个线程
    for i in xrange(190):
        t = threading.Thread(target=geturl)
        threads.append(t)
    for i in xrange(190):
        threads[i].start()
    for i in xrange(190):
        threads[i].join()"""
    t = threading.Thread(target=threadcheck, args=(10,))
    t.start()
    t.join()
    #geturl(url_new.pop(0))

#开始
if __name__ == '__main__':
    main()
input('整站抓取已结束！')