python爬取google搜索结果，配合sqlmap做sql注入检测

　　前段时间试了半自动sql注入检测的小程序：https://www.cnblogs.com/theseventhson/p/13755588.html  原理很简单：先在百度用关键词爬取目标url后保存在txt文件；再开启sqlmap的api调用服务，用python脚本把爬取的url推动到sqlmap的api实现批量检测sql注入的目的；从实际的效果来看，sql注入的检测效果还不错，存在注入的url还挺多的；但从百度爬取目标url的效果就很差了，很多站点都不带关键词的也被展示在结果页面，这些页面让sqlmap去检测，有sql注入的非常少，导致整体的效率并不高；今天参考别人的框架代码重新写了一个爬取google搜索结果的爬虫，源码如下：

from gevent import monkey;monkey.patch_all()
from bs4 import BeautifulSoup
import chardet
import os
import urllib.parse
import re
import random
from demo import fenpei
import configparser
import requests
import gevent
import re

class Google_Disk(object):
    def __init__(self):
        self.conf={}
        self.iplist=[]
        self.calc=0
        self.html=b''
        self.cookies={}
        self.cookie='HSID=AnOingRydX5d2psm6; SSID=ADt9T-YUVJhcGL4qL; APISID=wJEaAiaIyzvEaudB/AcoN5lpzTLnX5Reo_; SAPISID=f7PURACCKCHWwSNN/AzvNr8jk9DaahBOjn; CONSENT=YES+CN.zh-CN+20170611-09-0; SID=BQd-7E64xr8N2KPkSozUAhhUGA1yC2pOm44rxZeltI5oyZczMhTQXcaLdnFMy6KuYM7CVQ.; _ga=GA1.1.1066659943.1561908462; _gcl_au=1.1.1103150496.1563265661; ANID=AHWqTUkF83QBPYbfQq0kmzf1KcFRM9zsr6E6DzhE_HothF5Y28xI_VdxHrB1fMar; SEARCH_SAMESITE=CgQIzY0B; GOOGLE_ABUSE_EXEMPTION=ID=becbf893a4904d44:TM=1566184449:C=r:IP=47.75.69.236-:S=APGng0se1h0QgE8PglXBZJi1H6W3jRYdzw; NID=188=I04uuKTsGOjSp5c3G9QzFnfHqsL7ZQE3t9FdHLq25aPPiAHLfdWBsh3j3v14esoRRMVNXV6Pg8WXsqliJ8c7G46efNs-16lEr8ZZn6Fvz0GzYcw6wzcJ78OWUOuiz0K8W63M0zuBNTUDDmzVBxiud788TjTvbI5CZurTIcD6z2TTwQ_TuoGvjP2cuutFWcs5C8_11nk35jERGC2_A2UPda-AtI2mnVspSF5NNpawFUwW8PgQpxM; DV=oylrE6tRiwhOECBuCtWvdH13M-J_yhYIrTZO_A7m2wIAAABsoyqeic4gCwEAAFj9N_RUZyHkUQAAAA; 1P_JAR=2019-8-19-3; SIDCC=AN0-TYtz7HmrYpB6Cyw9ogysPbuDr2AY0pBl89HytGxEBiBr2lsZ4ceFMNWkG4Efolz2ihLVoMth'
        for v in self.cookie.split(';'):
            key,value=v.split('=',1)
            self.cookies[key]=value

        self.headers={'user-agent':'Mozilla/5.0 (Windows NT 6.1; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/76.0.3809.100 Safari/537.36'}
        self.host='https://www.google.tw/search?'
        if os.path.exists('Config.ini'):
            print('[~] 读取配置文件')
            config=configparser.ConfigParser()
            config.read("Config.ini",encoding='utf-8')
            self.conf['proxy']=config['config']['proxy']
            self.conf['save']=config['config']['save_name']
            self.conf['search']=config['config']['search_grammar']
            self.conf['page']=config['config']['page']
            self.conf['sleep']=config['config']['sleep']
            print('[+] 读取完成')
        else:
            print('[-] 找不到配置文件')
            exit()

        if os.path.exists('iplist.txt'):
            print('[~] 检测到iplist.txt,采用每次请求随机抽取一个IP')
            dk=open('iplist.txt','r')
            for r in dk.readlines():
                data="".join(r.split('
'))
                self.iplist.append(data)
            proxy=self.iplist
        else:
            proxy=self.conf['proxy']

        print('[config] 代理设置:{}'.format(proxy))
        print('[config] 搜索语法:{}'.format(self.conf['search']))
        print('[config] 抓取的页数:{}'.format(self.conf['page']))
        print('[config] 保存文件名:{}'.format(self.conf['save']))

    def search(self):
        for p in range(0,int(self.conf['page'])):
            page=p*10
            if len(self.iplist)>0:
                proxy=random.choice(self.iplist)
            else:
                proxy=self.conf['proxy']
            try:
                html=fenpei(proxy=proxy,search=self.conf['search'], page=page,sleep=self.conf['sleep'])
                if b'302 Moved' not in html:
                    #print(html)
                    self.html+=html
                else:
                    print('[-] Google又要你输验证码啦...')
            except Exception as r:
                print("in search exception---------------------------------------------------------------------------------------------
")
                print(r)

    def chuli(self):
        try:
            link_list = re.findall(r"<div class="yuRUbf"><a href="(.+?)" ping="", str(self.html))
            for url in link_list:  
                print (url.replace('&amp;sa=U&amp;','').replace('&amp;','')) 
                print(url.replace('&amp;sa=U&amp;','').replace('&amp;',''),file=open(self.conf['save'],'a'))
        except:
            pass

if __name__ == '__main__':
    obj=Google_Disk()
    obj.search()
    obj.chuli()

　　demo.py: 常用的功能函数封装；这里有个关键点：一般情况下，我只检测国内的站点，所以在url需要添加lr=lang_zh-CN&tbs=lr%3Alang_1zh-CN参数，完整的url如下：

url='https://{}/search?lr=lang_zh-CN&tbs=lr%3Alang_1zh-CN&q={}&btnG=Search&gbv=10&start={}'.format(domains,search,page)

import random
import requests
import time
from bs4 import BeautifulSoup
from http.cookiejar import LWPCookieJar
from urllib.request import Request, urlopen
from urllib.parse import quote_plus, urlparse, parse_qs

def read():
    dk=open('user_agents.txt','r',encoding='utf-8')
    for r in dk.readlines():
        data="".join(r.split('
'))
        yield data

def reads():
    dk=open('domain.txt','r',encoding='utf-8')
    for r in dk.readlines():
        data="".join(r.split('
'))
        yield data


def fenpei(proxy,search,page,sleep):
    user_agents=[]
    google_searchs=[]
    for ua in read():
        user_agents.append(ua)


    for domain in reads():
        google_searchs.append(domain)

    time.sleep(int(sleep))
    proxy={'http':'http://{}'.format(proxy),'https':'https://{}'.format(proxy)}
    domains=random.choice(google_searchs)
    u_s={'user-agent':random.choice(user_agents),'Content-type':"text/html;charset=utf-8"}
    url='https://{}/search?lr=lang_zh-CN&tbs=lr%3Alang_1zh-CN&q={}&btnG=Search&gbv=10&start={}'.format(domains,search,page)
    requests.packages.urllib3.disable_warnings(requests.packages.urllib3.exceptions.InsecureRequestWarning)
    rqt=requests.get(url=url,headers=u_s,allow_redirects=False,verify=False,proxies=proxy,timeout=30)
    return rqt.content

　　所有文件如下：py文件就是上面的python脚本

　　　　

•  Config.ini：爬虫的配置文件，分别是代理服务器的ip和端口、爬取结果保存的文件、搜索的关键词、爬取google搜索结果的页面数、每次爬取间隔时间（防止被google要求验证）

[config]
proxy=127.0.0.1:12639
save_name=save.txt
search_grammar=inurl:php?id=
page=10
sleep=5

•  domain.txt: google在全球各地的分站，可以多尝试不同的分站，避免被要求验证

　　

•  user-agent.txt：不同的操作系统、浏览器有不同的user-agent，这里可以来回切换，避免被要求验证；经过尝试，不同的user-agent会导致google返回不同的结果，给后续的url抽取带来困难，我这里暂时固化用这个：Mozilla/5.0 (Windows NT 6.2; WOW64) AppleWebKit/537.36 (KHTML like Gecko) Chrome/44.0.2403.155 Safari/537.36
•  iplist.txt：代理服务器的列表，格式为ip:portal
• save.txt：爬取结果保存的文档，都包含了我们自己设置的关键词，比百度精准太多了；

　　

 　　从sqlmap跑的结果来看，存在sql注入的url蛮多的（爬取的50个url中，检测出sql注入的有11个，比例已经超过了20%，保存在同目录的injection文件中），怪不得sql注入的漏洞在OWASP排名长期靠前（另一个是xss，近期用xray、awvs扫出来的高危漏洞超过一半都是xss）