Python实现Discuz论坛的自动POST登录发贴回帖

#-*-coding:utf-8-*-
import urllib2, urllib, cookielib
import re
import getpass
import sqlite3
import random
import time

class Discuz:
    def __init__(self,user,pwd,args):
        self.username = user
        self.password = pwd
        self.args = args
        self.regex = {
            'loginreg':'',
            'replyreg':'',
            'tidreg': '[\s\S]+?'
        }
        self.conn = None
        self.cur = None
        self.islogin = False
        self.login()
        self.InitDB()

    def login(self):
        try:
            loginPage = urllib2.urlopen(self.args['loginurl']).read()
            formhash = re.search(self.regex['loginreg'], loginPage)
            formhash = formhash.group(1)
            #print 'login formhash:', formhash
            print 'start login...'
            cj = cookielib.CookieJar()
            opener = urllib2.build_opener(urllib2.HTTPCookieProcessor(cj))
            user_agent = 'Mozilla/4.0 (compatible; MSIE 6.0; Windows NT 5.1; Mozilla/4.0 \
                    (compatible; MSIE 6.0; Windows NT 5.1; SV1) ; .NET CLR 2.0.507'
            opener.addheaders = [('User-agent', user_agent)]
            urllib2.install_opener(opener)
            logindata = urllib.urlencode({
                'cookietime':    2592000,
                'formhash': formhash,
                'loginfield':'username',
                'username':    self.username,
                'password':    self.password,
                'questionid':    0,
                'referer': self.args['referer']
                })
            request = urllib2.Request(self.args['loginsubmiturl'],logindata)
            response = urllib2.urlopen(request)
            self.islogin = True
            print 'login success...'
        except Exception,e:
                print 'loggin error: %s' % e

    def PostReply(self, fid, tid, content):
        try:
            sql = "select * from post where fid='%s' and tid='%s'" % (fid,tid)
            self.cur.execute(sql)
            if self.cur.rowcount == -1:
                tidurl = self.args['tidurl'] % tid
                replysubmiturl = self.args['replysubmiturl'] % (fid,tid)
                tidPage = urllib2.urlopen(tidurl).read()
                formhash = re.search(self.regex['replyreg'], tidPage)
                formhash = formhash.group(1)
                #print 'reply formhash:', formhash
                print 'start reply...'
                replydata = urllib.urlencode({
                    'formhash': formhash,
                    'message': content,
                    'subject': '',
                    'usesig':'1'
                })
                request = urllib2.Request(replysubmiturl,replydata)
                response = urllib2.urlopen(request)
                sql = "insert into post values ('%s', '%s', '%d')" % (fid, tid, 1)
                self.cur.execute(sql)
                self.conn.commit()
                print 'reply success for [%s]' % tidurl
            else:
                print 'Skip! Thread:%s is already replied...' % tid
        except Exception, e:
                print 'reply error: %s' % e

    def GetTids(self, fid):
        if self.islogin:
            fidurl = self.args['fidurl'] % fid
            response = urllib2.urlopen(fidurl)
            content = response.read()
            tids = re.findall(self.regex['tidreg'], content)
            return tids
        else:
            print 'Error Please Login...'

    def InitDB(self):
        self.conn = sqlite3.connect('data.db')
        self.cur = self.conn.cursor()
        sql = '''create table if not exists post (
            fid text,
            tid text,
            replied integer)'''
        self.cur.execute(sql)
        self.conn.commit()

if __name__ == '__main__':
    username = raw_input('username:').strip()
    password = getpass.getpass('password:').strip()
    args = {
            'loginurl': 'http://www.xxx.com/logging.php?action=login',
            'loginsubmiturl': 'http://www.xxx.com/logging.php?action=login&loginsubmit=yes',
            'fidurl': 'http://www.xxx.com/forum-%s-1.html',
            'tidurl': 'http://www.xxx.com/thread-%s-1-1.html',
            'replysubmiturl': 'http://www.xxx.com/post.php?action=reply&replysubmit=yes&infloat=yes&handlekey=fastpost&fid=%s&tid=%s',
            'referer':'http://www.xxx.com/index.php'
    }
    dz = Discuz(username, password,args)
    fid = '45'
    tids = dz.GetTids('45')
    replylist = [
            u'不错，支持一下，呵呵',
            u'已阅，顶一下',
            u'看看，顶你，呵呵',
            u'多谢分享，顶一下',
            u'说的不错，支持一下',
            u'提着水桶到处转，哪里缺水哪里灌！ ',
            u'你太油菜了!'
    ]
    for tid in tids:
        content = random.choice(replylist)
        content = content.encode('gbk')
        dz.PostReply('45',tid, content)
        time.sleep(20)

formhash cookietime formhash loginfield password questionid referer username

• cookietime 浏览器自动给的是 2592000
• loginfield 默认的username
• password 密码
• questionid 这个貌似是登录时的回答问题，这个论坛没有强制回答所以用默认的0
• referer 这个则是引用地址 http://www.xxx.com/index.php
• username 用户名
• formhash 最后这个貌似这个是随机的，不固定，可也是个关键参数，所以就直接用正则查找之

args = {
            'loginurl': 'http://www.xxx.com/logging.php?action=login',
            'loginsubmiturl': 'http://www.xxx.com/logging.php?action=login&loginsubmit=yes',
            'fidurl': 'http://www.xxx.com/forum-%s-1.html',
            'tidurl': 'http://www.xxx.com/thread-%s-1-1.html',
            'replysubmiturl': 'http://www.xxx.com/post.php?action=reply&replysubmit=yes&infloat=yes&handlekey=fastpost&fid=%s&tid=%s',
            'referer':'http://www.xxx.com/index.php'
    }

• loginurl为登录面页，用于获得formhash的值
• loginsubmiturl为post登录参数的地址
• fidurl这个是版块的ID，url中%s那里即为fid，这样的url http://www.xxx.com/forum-45-1.html，fid即为45
• tidurl是帖子的id，查找方法同上
• replysubmiturl这个是回复帖子post参数的url，要定位一个帖子前提得知道fid和tid
• referer这个是引用地址，用网站的首页即可