requests+selenium+buautifulsou爬取cnvd

一.明确爬取需求

今天要爬的是cnvd，网站叫国家信息安全漏洞共享平台，我们要爬的页面是漏洞列表页面，我们看图说话。

微信截图_20180916225155.png

大致分析一下爬取步骤：
1.爬取 漏洞列表首页，获取总页数，今天更新到5647页了。
2。循环遍历漏洞列表的每一页，获取详细的漏洞内容，点开一个看看里边都有啥？

微信截图_20180917222956.png

门前大桥下，游过一群鸭，你猜多少只？来数数啊。漏洞标题，CNVD_ID，公开日期，危害级别，影响产品，漏洞描述，参考链接，漏洞解决方案，厂商补丁。大概就是这些了，话不多说，咱开干呗！

二.干活了

用requests简单写了个初级crawler，循环获取漏洞列表中的每一页，但是只获取了8页就gg了，Are you kidding me?政府网站竟然也加了反爬，让我来研究研究 。

网页拉到最底下，发现网站服务由知道创宇公司的加速乐产品提供。百度加速乐，发现其对一些政府网站做了反爬优化，不过对cnvd没做的太过分。那就加个cookies池吧，F12查看cookie，发现有三组cookie，用selenium获取cookie，存为json文件。

微信截图_20180917225603.png

新建get_cnvd_cookie.py，直接贴代码，这个挺简单，大家自己看啊！

# _*_ coding: utf-8 _*_
from selenium import webdriver
from selenium.webdriver.common.by import By
from selenium.webdriver.support.ui import WebDriverWait
from selenium.webdriver.support import expected_conditions as EC
from selenium.webdriver.chrome.options import Options
import json
import time
if __name__ == '__main__':
    list = []
    for i in range(500):
        chrome_options = Options()
        chrome_options.add_argument('--headless')
        driver = webdriver.Chrome(chrome_options=chrome_options)
        driver.get('http://www.cnvd.org.cn/flaw/show/CNVD-2018-17745')
        ck = driver.get_cookies()
        name = []
        value = []
        for c in ck:
            name.append(c['name'])
            value.append(c['value'])
        dd = dict(zip(name, value))
        with open('cnvd_cookies.json', 'a+', encoding='utf-8') as file:
            line = json.dumps(dd, ensure_ascii=False) + '\n'
            file.write(line)
        driver.close()
        time.sleep(0.1)

有点犯懒了，手动将生成的内容改为cnvd_pool列表，也就是CNVD_POOL = []中加入刚才生成的cookies。

下边开始上主代码，时间不早了，先更一波，明天再做解读。

# -*- coding: utf-8 -*-
"""
cnvd crawler
"""

import re
import time
import random
import json
import requests
from bs4 import BeautifulSoup
from settings import CNVD_POOL


class CnvdSpider:
    def __init__(self):
        headers = {
            'Accept': 'text/html,application/xhtml+xml,application/xml;q=0.9,image/webp,image/apng,*/*;q=0.8',
            'Accept-Encoding': 'gzip, deflate',
            'Accept-Language': 'zh-CN,zh;q=0.9',
            'Host': 'www.cnvd.org.cn',
            'Proxy-Connection': 'keep-alive',
            'Referer': 'http://www.cnvd.org.cn/flaw/list.htm?flag=true',
            'Upgrade-Insecure-Requests': '1',
            'User-Agent': 'Mozilla/5.0 (Windows NT 6.1; WOW64) AppleWebKit/537.36 (KHTML, like Gecko) '
                          'Chrome/68.0.3440.106 Safari/537.36',
        }
        self.start_url = 'http://www.cnvd.org.cn/flaw/list.htm?flag=true'
        self.session = requests.session()
        self.session.headers.update(headers)
        self.update_session()
        with open('vulnerabilitys.json', 'w+', encoding='utf-8') as file:
            pass
        self.start_time = time.time()

    def update_session(self):
        req_msg = random.choice(CNVD_POOL)
        print('cookies:  ', req_msg)
        # 传入cookieJar
        requests.utils.add_dict_to_cookiejar(self.session.cookies, req_msg)


    def crawl_list_page(self):
        data = {'max': 100, 'offset': 0}
        response = self.session.post(self.start_url, data=data)
        response.encoding = 'utf-8'
        # print(response.text)
        soup = BeautifulSoup(response.text, 'lxml')
        # get_text() 必须加括号
        last_page = soup.select_one('.pages > a:nth-of-type(10)').get_text()
        for page in range(int(last_page)):
            data['offset'] = page * 100
            while True:
                try:
                    resp = self.session.post(self.start_url, data=data)
                except:
                    self.update_session()
                    continue
                if resp.text != '':
                    break
                else:
                    # time.sleep(1)
                    self.update_session()
            with open('logs.txt', 'a', encoding='utf-8') as file:
                file.writelines('页码:-------------------->max={}&offset={}\n'.format(data['max'], data['offset']))
            self.crawl_detailed_page(resp)
            break
    def crawl_detailed_page(self, response):
        print('crawl_detailed_page: {}'.format(response.url))
        response.encoding = 'utf-8'
        soup = BeautifulSoup(response.text, 'lxml')
        detailed_pages = soup.select('.tlist tbody > tr > td:nth-of-type(1) > a')
        left_url = 'http://www.cnvd.org.cn'
        for page in detailed_pages:
            while True:
                try:
                    resp = self.session.get(left_url + page.attrs['href'])
                except:
                    self.update_session()
                    continue
                if resp.text != '':
                    break
                else:
                    # time.sleep(1)
                    self.update_session()
            self.get_detailed_info(resp, page.attrs['title'])

    def get_detailed_info(self, response, title):
        take_time = int((time.time() - self.start_time) / 60)
        print('time:{}min.     get_detailed_info: {}'.format(take_time, response.url))
        left_url = 'http://www.cnvd.org.cn'
        detailed_dict = {
            "number": None, "title": title, "openTime": None, "serverity": None, "products": None,
            "cve": None, "description": None, "referenceLink": None, "formalWay": None,
            "patch": None,}
        response.encoding = 'utf-8'
        soup = BeautifulSoup(response.text, 'lxml')
        trs = soup.select('.gg_detail tbody tr')
        for tr in trs:
            name = tr.find('td').get_text()
            if name == 'CNVD-ID':
                value = tr.select_one('td:nth-of-type(2)').get_text().strip()
                detailed_dict['number'] = value
            elif name == '公开日期':
                value = tr.select_one('td:nth-of-type(2)').get_text().strip()
                detailed_dict['openTime'] = value
            elif name == '危害级别':
                value = tr.select_one('td:nth-of-type(2)').get_text().strip()
                detailed_dict['serverity'] = value[:1]
            elif name == '影响产品':
                value = tr.select_one('td:nth-of-type(2)').get_text()
                pattern=re.compile('(\S.*?)\s\s', re.S)
                value_dict = re.findall(pattern, value+'  ')
                detailed_dict['products'] = value_dict

            elif name == 'CVE ID':
                value = tr.select_one('td:nth-of-type(2)').get_text().strip()
                cve_href = tr.select_one('td:nth-of-type(2) a').attrs['href']
                detailed_dict['cve'] = value
                cve_dict = {'id': value, 'href': cve_href}
                detailed_dict['cve'] = cve_dict
            elif name == '漏洞描述':
                value = tr.select_one('td:nth-of-type(2)').get_text()\
                    .replace('\r\n', '').replace('\t', '').strip()
                detailed_dict['description'] = value
            elif name == '参考链接':
                try:
                    value = tr.select_one('td:nth-of-type(2) a').get_text().strip()
                    detailed_dict['referenceLink'] = value
                except AttributeError:
                    pass
            elif name == '漏洞解决方案':
                value = tr.select_one('td:nth-of-type(2)').get_text().strip()
                pattern = re.compile('\t(http.*)', re.S)
                value = re.findall(pattern, value)
                detailed_dict['formalWay'] = value
            elif name == '厂商补丁':
                try:
                    href = tr.select_one('td:nth-of-type(2) a').attrs['href']
                    value = tr.select_one('td:nth-of-type(2)').get_text().strip()
                    patch_dict = {'patchName': value, 'href': left_url+href}
                    detailed_dict['patch'] = patch_dict
                except AttributeError:
                    pass
        # 写入json文件
        with open('vulnerabilitys.json', 'a', encoding='utf-8') as file:
            line = json.dumps(detailed_dict, ensure_ascii=False) + '\n'
            file.write(line)
        with open('logs.txt', 'a', encoding='utf-8') as file:
            file.writelines('漏洞详细信息url----->{}\n'.format(response.url))

if __name__ == '__main__':
    a = CnvdSpider()
    a.crawl_list_page()