funny_read.py

#coding:utf-8

import requests
import time
from bs4 import BeautifulSoup

class TopicResponse(object):
    title = u''
    url = ''
    reply_count = -1
    created_at = 'None'
    def __str__(self):
        return '''%i %s %s %s ''' % (self.reply_count, self.created_at, self.title, self.url)


class FunnyRead(object):
    def __init__(self):
        pass
    
    def datas(self, content):
        _topics = []
        c = BeautifulSoup(content, from_encoding='gbk')
        data = c.find_all('tr', class_ = 'tr3 t_one')
        for d in data:
            _t = TopicResponse()
            alink = d.find('h3').find('a')
            _t.title = alink.text.encode('utf-8')
            _t.url = 'http://t66y.com/' + alink.get('href',None)
            rly_count = d.find('td', class_ = 'tal f10 y-style')
            if rly_count:
                _t.reply_count = int(rly_count.text)
            else:
                _t.reply_count = -1
            _t.created_at = d.find('td', class_ = 'tal y-style').find('div', class_ = 'f10').text.encode('utf-8')
            _topics.append(_t)
        
        return _topics

def funnyOutput(url = None, limit = 25):
    print 'funny'
    _url = 'http://t66y.com/thread0806.php?fid=2'
    if url:
        _url = url
    pages = 3
    _topics = []
    funny = FunnyRead()
    for i in range(pages):
        print 'working'
        u = _url + '&page=%i'%i
        try:
            req = requests.get(u)
        except:
            continue
        rs = funny.datas(req.content)
        _topics = _topics + rs
    
    uniques = [o.url for o in _topics]

    t = [ss for ss in _topics if uniques.count(ss.url) == 1]
    t = sorted(t, key = lambda t: t.reply_count, reverse = True)
    t = t[:limit]
    for out in t:
        print out
        
if __name__ == '__main__':
    funnyOutput('http://t66y.com/thread0806.php?fid=2')