CMS 识别（Web 指纹识别）扫描器开发

CMS 识别原理展开目录

CMS 英文全称是：Content Management System，中文名称是：网站内容管理系统。CMS 识别原理就是得到一些 CMS 的一些固有特征，通过得到这个特征来判断 CMS 的类别。

这里我们采用 MD5 识别和正则表达式识别的方式，具体来说就是用特定的文件路径访问网站，或者这个文件的 MD5 或者用正则表达式匹配某个关键词，如果匹配成功就说明是这个 CMS。所以识别的成功率与字典有关

指纹格式展开目录

这里截取一些 Web 指纹作为参考：

{
    "url": "/install/",
    "re": "aspcms",
    "name": "AspCMS",
    "md5": ""
},
{
    "url": "/about/_notes/dwsync.xml",
    "re": "aspcms",
    "name": "AspCMS",
    "md5": ""
},
{
    "url": "/admin/_Style/_notes/dwsync.xml",
    "re": "aspcms",
    "name": "AspCMS",
    "md5": ""
},
{
    "url": "/apply/_notes/dwsync.xml",
    "re": "aspcms",
    "name": "AspCMS",
    "md5": ""
},
{
    "url": "/tpl/green/common/images/notebg.jpg",
    "re": "",
    "name": "自动发卡平台",
    "md5": "690f337298c331f217c0407cc11620e9"
},
{
    "url": "/images/download.png",
    "re": "",
    "name": "全程oa",
    "md5": "9921660baaf9e0b3b747266eb5af880f"
},
{
    "url": "/kindeditor/license.txt",
    "re": "",
    "name": "T-Site建站系统",
    "md5": "b0d181292c99cf9bb2ae9166dd3a0239"
},
{
    "url": "/public/ico/favicon.png",
    "re": "",
    "name": "悟空CRM",
    "md5": "834089ffa1cd3a27b920a335d7c067d7"
},
{
    "url": "/public/js/php/file_manager_json.php",
    "re": "",
    "name": "悟空CRM",
    "md5": "c64fd0278d72826eb9041773efa1f587"
},
{
    "url": "/plugins/weathermap/images/exclamation.png",
    "re": "",
    "name": "CactiEZ插件",
    "md5": "2e25cb083312b0eabfa378a89b07cd03"
}

指纹文件展开目录

在 data 目录下存放 data.json 文件格式的 Web 指纹，总共有 1400 + 的国内常见指纹，[下载地址]()

代码编写展开目录

思路虽然简单，但实现起来还有很多问题，比如效率，1000 + 指纹说明需要访问 1000 + 的网页，单步的话速度太慢，所以需要使用多线程，等用多了也会发现多线程也太慢了，所以可以使用协程，以后再慢慢优化，这里就使用多线程就行了

新建文件 lib/core/webcms.py，代码如下

# __author__ = 'mathor'
import json, os, sys, hashlib, threading, queue
from lib.core import Download

class webcms(object):
    workQueue = queue.Queue()
    URL = ""
    threadNum = 0
    NotFound = True
    Downloader = Download.Downloader()
    result = ""
    
    def __init__(self, url, threadNum = 10):
        self.URL = url
        self.threadNum = threadNum
        filename = os.path.join(sys.path[0], 'data', 'data.json')
        fp = open(filename, encoding = 'utf-8')
        webdata = json.load(fp, encoding = 'UTF-8')
        for i in webdata:
            self.workQueue.put(i)
        fp.close
        
    def getmd5(self, body):
        m2 = hashlib.md5()
        m2.update(body)
        return m2.hexdigest()
        
    def th_whatweb(self):
        if (self.workQueue.empty()):
            self.NotFound = False
            return False
    
        if (self.NotFound is False):
            return False
        cms = self.workQueue.get()
        _url = self.URL + cms['url']
        html = self.Downloader.get(_url)
        print("[whatweb log]:checking %s" % _url)
        if (html is None):
            return False
        if cms['re']:
            if (html.find(cms['re']) != -1):
                self.result = cms['name']
                self.NotFound = False
                return True
        else:
            md5 = self.getmd5(html)
            if (md5 == cms['md5']):
                self.result = cms['name']
                self.NotFound = False
                return True
    
    def run(self):
        while(self.NotFound):
            th = []
            for i in range(self.threadNum):
                t = threading.Thread(target = self.th_whatweb)
                t.start()
                th.append(t)
            for t in th:
                t.join()
        if (self.result):
            print("[webcms]:%s cms is %s" % (self.URL, self.result))
        else:
            print("[webcms]:%s cms NOTFound!" % self.URL)

调用展开目录

重写主文件 w8ay.py

#-*- coding:utf-8 -*-
'''
Name: w8ayScan
Author: mathor
Copyright (c) 2019
'''
import sys
from lib.core.Spider import SpiderMain
from lib.core import webcms
def main():
    root = "https://wmathor.com"
    threadNum = 1000
    
    # webcms
    ww = webcms.webcms(root, threadNum)
    ww.run()
    
    # spider
    w8 = SpiderMain(root, threadNum)
    w8.craw()
 
if __name__ == "__main__":
    main()