爬虫（五）-openlaw

地址：http://openlaw.cn/login.jsp

需要登陆，Form data

 

找到_csrf和password，_csrf，在登陆页面

找加密password的js代码，ctrl+F搜索password，重新填入input

找到加密函数KeyEncrypt

找到JSEncrypt，CryptoJs

 

执行

用execjs执行js或者用js2py

node.js或Jscript报navigator is not define，可以用PhantomJS

#运行时环境设置
import execjs
import execjs.runtime_names
os.environ["EXECJS_RUNTIME"] = "PhantomJS"
PhantomJS = execjs.get(execjs.runtime_names.PhantomJS)
print(execjs.get().name)

结果如图

pytohn执行

#获取csfr
headers={
"Accept":"text/html,application/xhtml+xml,application/xml;q=0.9,image/webp,image/apng,*/*;q=0.8,application/signed-exchange;v=b3",
"Accept-Encoding":"gzip,deflate",
"Accept-Language":"zh-CN,zh;q=0.9",
"Cache-Control":"max-age=0",
"Connection":"keep-alive",
"Content-Type":"application/x-www-form-urlencoded",
"Host":"openlaw.cn",
"Upgrade-Insecure-Requests":"1",
"User-Agent":"Mozilla/5.0(WindowsNT10.0;Win64;x64)AppleWebKit/537.36(KHTML,likeGecko)Chrome/73.0.3683.86Safari/537.36",
}
r=session.get("http://openlaw.cn/login.jsp?",headers=headers1)
tree=etree.HTML(r.text)
csrf=tree.xpath('//input[@name="_csrf"]/@value')[0]

#post form data
post_url="http://openlaw.cn/login"
with open(r"C:\Users\37576\Desktop\123.js",encoding="utf-8")as f:
    js=f.read()
    ctx = execjs.compile(js)
    pwd=ctx.call("keyEncrypt","XXXXXXXX")#密码
data={
"_csrf":csrf,
"username":"#########",#账户
"password":pwd,
"_spring_security_remember_me":"true",
}
res=session.post(post_url,data=data,headers=headers)

 tree=etree.HTML(res.text)
 lala=tree.xpath('//div[@class="bbp-user-section"]//text()')

登陆成功

 

页面

url="http://openlaw.cn/guidance/16444eba67afe97881fa6521d333ef10"
headers1={
"Accept":"text/html,application/xhtml+xml,application/xml;q=0.9,image/webp,image/apng,*/*;q=0.8,application/signed-exchange;v=b3",
"Accept-Encoding":"gzip,deflate",
"Accept-Language":"zh-CN,zh;q=0.9",
"Connection":"keep-alive",
"Host":"openlaw.cn",
"Upgrade-Insecure-Requests":"1",
"User-Agent":"Mozilla/5.0(WindowsNT10.0;Win64;x64)AppleWebKit/537.36(KHTML,likeGecko)Chrome/73.0.3683.86Safari/537.36",
}
a=session.get(url,headers=headers1)

是混淆的js

抓包

有二条请求

第一条加cookie，第二条正常，需要知道怎么加的cookie

jjencodeDemo：http://utf-8.jp/public/jjencode.html
],_$_:++$,$_$$:({}+"")[$],$$_$:($[$]+"")[$],_$$:++$,$$$_:(!""+"")[$],$__:++$,$_$:++$,$$__:({}+"")[$],$$_:++$,$$$:++$,$___:++$,$__$:++$};$.$_=($.$_=$+"")[$.$_$]+($._$=$.$_[$.__$])+($.$$=($.$+"")[$.__$])+((!$)+"")[$._$$]+($.__=$.$_[$.$$_])+($.$=(!""+"")[$.__$])+($._=(!""+"")[$._$_])+$.$_[$.$_$]+$.__+$._$+$.$;$.$$=$.$+(!""+"")[

jsfunk：http://www.jsfuck.com/
+[]+!+[]+!+[]]+(!![]+[])[+!+[]]]+[])[!+[]+!+[]+!+[]]+(!![]+[][(![]+[])[+[]]+([![]]+[][[]])[+!+[]+[+[]]]+(![]+[])[!+[]+!+[]]+(!![]+[])[+[]]+(!![]+[])[!+[]+!+[]+!+[]]+(!![]+[])[+!+[]]])[+!+[]+[+[]]]+([][[]]+[])[+!+[]]+(![]+[])[!+[]+!+[]+!+[]]+(!![]+[])[+[]]+(!![]+[])[+!+[]]+([][[]]

aaencode：http://utf-8.jp/public/aaencode.html

ﾟωﾟﾉ= /｀ｍ´）ﾉ ~┻━┻ //*´∇｀*/ ['_']; o=(ﾟｰﾟ) =_=3; c=(ﾟΘﾟ) =(ﾟｰﾟ)-(ﾟｰﾟ); (ﾟДﾟ) =(ﾟΘﾟ)= (o^_^o)/ (o^_^o);(ﾟДﾟ)={ﾟΘﾟ: '_' ,ﾟωﾟﾉ : ((ﾟωﾟﾉ==3) +'_') [ﾟΘﾟ] ,ﾟｰﾟﾉ :(ﾟωﾟﾉ+ '_')[o^_^o -(ﾟΘﾟ)] ,ﾟДﾟﾉ:((ﾟｰﾟ==3) +'_')[ﾟｰﾟ] }; (ﾟДﾟ) [ﾟΘﾟ] =((ﾟωﾟﾉ==3) +'_') [c^_^o];(ﾟДﾟ) ['c'] = ((ﾟДﾟ)+'_') [ (ﾟｰﾟ)+(ﾟｰﾟ)-(ﾟΘﾟ) ];

代码执行一下或者找工具解密

(function anonymous() {
    if (window.wp1) {
        document.cookie = 'OPEN_E4C0C08491=' + wp3;
        document.cookie = 'OPEN_ID=B32pJOwfO/AJZJ3ZGmnunQNTaUie84jnyHoNCq/onZQ='
    }
}
)

加cookie的部分

js：

var wp1 = 'xKIOt5jUZQcS8wurcX4lghwlET9K0IWBUsS3ZqmSamCBTAPg7Fl5sVr443dyK42p';
var wp2 = "ABCDEFGHIJKLMNOPQRSTUVWXYZabcdefghijklmnopqrstuvwxyz0123456789+/=";
var wp3 = "548545fe3ced4f3494934e537e304705";
var wp4 = "464acf618bf62df3e7e37e80c2c1a04d";
var wp5 = "0aa9566c4aa30883e34d0e1a530bf508";
var wp6 = "a8f49232d343e4b94ff7958088fa6734";
var wp7 = "82f778b45a9fce8589105c852174947b";
var wp8 = "3221d819c9b96d2631d0b5084c736f69";
if (window.wp1) {
        document.cookie = 'OPEN_E4C0C08491=' + wp3;
        document.cookie = 'OPEN_ID=B32pJOwfO/AJZJ3ZGmnunQNTaUie84jnyHoNCq/onZQ='
    }

var getGlobal = function() {
        if (typeof self !== '') {
            return self
        }
        ;if (typeof window !== '') {
            return window
        }
        ;if (typeof global !== '') {
            return global
        }
        ;throw new Error('unable to locate global object')
    };
function _process(s) {
        var result = s.substring(5, 7).concat('s').concat(s.substring(14, 15)).concat('v').concat(s.substring(9, 13)).concat('g').concat(s.substring(5)).concat('l').concat(s.substring(11, 19));
        return result.substr(0, 32);
    }
    ;var _switch = false;
    if ((getGlobal() === window)) {
        _switch = true
    } else {
        _switch = false
    };
document.cookie = 'SIGNIN_ID=3e2291010f36c9244c633aba4cdcdd42';
document.cookie = 'SIGNIN_UC=9e65d8fc646e49fb7a0f9f8237915bfb';
document.cookie = 'UNDEFINED=432e9111ecf89746f3046ca970814f41';
if (_switch) {
        document.cookie = 'OPEN_589D5869B5=' + _process(wp3);
        setTimeout("location.reload();", 100);
}

。。。很多参数不知道哪里来的搞不定了，用selenium

#配置chorme
from selenium import webdriver
from selenium.webdriver.chrome.options import Options
chrome_options = Options()
chrome_options.add_argument('--headless')
driver = webdriver.Chrome(chrome_options=chrome_options,executable_path = 'D:\python\chromedriver.exe')
from time import sleep

#登陆
driver.get("http://openlaw.cn/login.jsp?logout") 
driver.find_element_by_id("username").send_keys("XXXXXXXXX")
driver.find_element_by_id("password").send_keys("XXXXXXXXX")
sleep(2)
driver.find_element_by_id("submit").click()

#页面数据
driver.get("http://openlaw.cn/guidance/167ce490ba94c3ed26baab5689fdd620") 
sleep(2)
title=driver.find_element_by_xpath('//h1[@class="entry-title"]').text
guid=driver.find_element_by_xpath('//div[@class="annotator-wrapper"]/p').text
content=driver.find_element_by_xpath('//div[@class="annotator-wrapper"]/p[2]').text
page_detail={"title":title,"content":guid+"\n"+content.replace("\u3000"," ")}

结果