get_vulinfo.py

# coding: utf-8  
import csv
import sys
import datetime
import requests
from bs4 import BeautifulSoup

BASE_URL = 'https://jvndb.jvn.jp'
SOFTWARE_NAME = sys.argv[1]

def get(url, params = None):
    r = requests.get(url, params=params)
    r.encoding = r.apparent_encoding
    return r

def get_list(num, is_init = None):
    url = 'https://jvndb.jvn.jp/search/index.php'
    data = {
            'mode': '_vulnerability_search_IA_VulnSearch',
            'lang': 'ja', 
            'keyword': SOFTWARE_NAME, 
            'useSynonym': '1', 
            'vendor': '', 
            'product': '', 
            'datePublicFromYear': '', 
            'datePublicFromMonth': '', 
            'datePublicToYear': '', 
            'datePublicToMonth': '', 
            'dateLastPublishedFromYear': '', 
            'dateLastPublishedFromMonth': '', 
            'dateLastPublishedToYear': '', 
            'dateLastPublishedToMonth': '', 
            'cwe': '', 
            'searchProductId': '',
            'pageNo' : str(num)
        }

    r = requests.get(url, params=data)
    r.encoding = r.apparent_encoding
    
    bs = BeautifulSoup(r.text, 'html.parser')

    if is_init is True:
        number_of_vulns = bs.select('td[class="pager_count_class"]')[0].string
        number_of_vulns = int(number_of_vulns.split('件')[0])
        if number_of_vulns % 100 == 0:
            return number_of_vulns // 100
        else:
            return number_of_vulns // 100 + 1

    vuln_list = []
    for tr in bs.select('table[class="result_class"]')[0].find_all('tr'):
        for a in tr.select('a[href^="/ja/contents/"]'):
            vuln_list.append([
                a.get('href'),
                tr.select('td:nth-of-type(2)')[0].string
            ])
    return vuln_list

def get_info(lists):
    res = []
    for item in lists:
        r = get(BASE_URL + item[0])
        bs = BeautifulSoup(r.text, 'html.parser')
        affect_version = ''
        for li in bs.select('tr:nth-of-type(9) li'):
            affect_version += li.string.replace('\xa0', ' ') + '\x0d\x0a'
        cve = ''
        for li in bs.select('tr:nth-of-type(20) li'):
            cve = li.string

        date_string = bs.select('div[class="modifytxt"]')[0].string[-10:]
        date = datetime.datetime.strptime(date_string, '%Y/%m/%d')
        incomming_date = date.strftime('%Y-%m-%d')

        res.append([
                SOFTWARE_NAME,
                item[1],
                affect_version[:-1],
                cve,
                BASE_URL + item[0],
                incomming_date,
            ])
    return res

if __name__ == '__main__':
    max_page = get_list(1, True)
    with open('vuln_info.csv', 'w', encoding='shift_jis') as f:
        writer = csv.writer(f, lineterminator='\n', quoting=csv.QUOTE_NONNUMERIC)
        writer.writerow(['ソフトウェア名', '概要', '影響のあるバージョン', 'CVE', '情報元URL', '情報着信日'])
        for i in range(1,max_page+1):
            vuln_list = get_list(i)
            res = get_info(vuln_list)
            writer.writerows(res)