自动评论 CSDN 的一段小脚本

csdn_comment.py

# -*- coding=utf-8 -*-
# -*- author: wellls -*-

from __future__ import division 
from bs4 import BeautifulSoup
from bs4.element import (
    CharsetMetaAttributeValue,
    Comment,
    ContentMetaAttributeValue,
    Doctype,
    SoupStrainer,
    Tag,
    NavigableString
)
import re
import requests
import os
import time
import sys
reload(sys)
sys.setdefaultencoding( "utf-8" )
import shutil
import urllib2
import urllib
import cookielib
from poster.encode import multipart_encode
from poster.streaminghttp import register_openers
import zipfile

def AutoComment(username,password):
    #login  
    url="https://passport.csdn.net/account/login"
    head={
        "User-Agent":"Mozilla/5.0 (Windows NT 10.0; WOW64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/49.0.2623.87 Safari/537.36",
    }
    s=requests.session()
    r=s.get(url,headers=head)
    lt_execution_id=re.findall('name="lt" value="(.*?)".*\sname="execution" value="(.*?)"', r.text, re.S)
    #print lt_execution_id
    payload={
        "username":username,
        "password":password,
        "lt":lt_execution_id[0][0],
        "execution":lt_execution_id[0][1],
        "_eventId":"submit"
    }
    r2=s.post(url,headers=head,data=payload)
    #print r2.text    #login success and response a js
    print "*"*100     #split
    
    NotCommented = []    #store the url to be commented
    for page in range(1,54):                # get url
        r3=s.get("http://download.csdn.net/my/downloads/" + str(page),headers=head)
        #print page
        result_file = open("r3.html",'w')
        result_file.write(r3.text)
        #print r3.text.encode("GBK")
        
        #Get the list to be commented
        soup = BeautifulSoup(open("r3.html"),"html.parser")
        soup_body = soup.body
        i = 0
        for child in soup_body.children:
            if type(child) == Tag and child.name == "div":
                i = i + 1
                if i == 2:
                    div_1 = child
                    break
        #print div_1.encode("GBK")
        div_2 = div_1.link.link.link.link.link.div
        #print div_2.encode("GBK")
        j = 0
        for child in div_2.children:
            if type(child) == Tag:
                j = j + 1
                if j == 2:
                    div_3 = child
                    break
        #print div_3.encode("GBK")
        k = 0
        for child in div_3.children:
            if type(child) == Tag and child.name == "div":
                k = k + 1
                if k == 2:
                    dl_1 = child.dl
        #print dl_1.encode("GBK")
        for child in dl_1.children:
            if type(child) == Tag and child.name == "dt":
                if child.a.attrs.has_key("target") == False:    #to be commented
                    NotCommented.append(child.a["href"])
    print len(NotCommented)
    #comment
    #'''
    for url in NotCommented:
        searchObj = re.search(r'\/(.*)\/(.*?)\/(.*?)#comment', url, re.M|re.I)
        id = searchObj.group(3)
        print "comment id: " + id
        #time.sleep(60)
        r4=s.get("http://download.csdn.net/index.php/comment/post_comment?jsonpcallback=&sourceid=" + id + "&content=%E8%B5%84%E6%BA%90%E5%BE%88%E4%B8%8D%E9%94%99%EF%BC%8C%E6%84%9F%E8%B0%A2%E5%8D%9A%E4%B8%BB%E7%9A%84%E5%88%86%E4%BA%AB%EF%BC%81&txt_validcode=undefined&rating=5&t=&_=",headers=head)
        if r4.text.encode("GBK") == "({\"succ\":1})":
            print id + " done"
        else:
            print id + " something wrong"
    #'''
try:    
    AutoComment(sys.argv[1],sys.argv[2])
except Exception,e:
    print "Usage: CsdnAutoComment.py username password"
    print str(e)