1<?php 2 3/** 4 * webtrees: online genealogy 5 * Copyright (C) 2023 webtrees development team 6 * This program is free software: you can redistribute it and/or modify 7 * it under the terms of the GNU General Public License as published by 8 * the Free Software Foundation, either version 3 of the License, or 9 * (at your option) any later version. 10 * This program is distributed in the hope that it will be useful, 11 * but WITHOUT ANY WARRANTY; without even the implied warranty of 12 * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the 13 * GNU General Public License for more details. 14 * You should have received a copy of the GNU General Public License 15 * along with this program. If not, see <https://www.gnu.org/licenses/>. 16 */ 17 18declare(strict_types=1); 19 20namespace Fisharebest\Webtrees\Http\Middleware; 21 22use Fig\Http\Message\StatusCodeInterface; 23use Fisharebest\Webtrees\Registry; 24use Fisharebest\Webtrees\Validator; 25use GuzzleHttp\Client; 26use GuzzleHttp\Exception\GuzzleException; 27use Iodev\Whois\Loaders\CurlLoader; 28use Iodev\Whois\Modules\Asn\AsnRouteInfo; 29use Iodev\Whois\Whois; 30use IPLib\Address\AddressInterface; 31use IPLib\Factory as IPFactory; 32use IPLib\Range\RangeInterface; 33use Psr\Http\Message\ResponseInterface; 34use Psr\Http\Message\ServerRequestInterface; 35use Psr\Http\Server\MiddlewareInterface; 36use Psr\Http\Server\RequestHandlerInterface; 37use Throwable; 38 39use function array_filter; 40use function array_map; 41use function assert; 42use function gethostbyaddr; 43use function gethostbyname; 44use function preg_match_all; 45use function random_int; 46use function response; 47use function str_contains; 48use function str_ends_with; 49 50/** 51 * Middleware to block bad robots before they waste our valuable CPU cycles. 52 */ 53class BadBotBlocker implements MiddlewareInterface 54{ 55 private const REGEX_OCTET = '(?:25[0-5]|2[0-4][0-9]|[01]?[0-9][0-9]?)'; 56 private const REGEX_IPV4 = '/\\b' . self::REGEX_OCTET . '(?:\\.' . self::REGEX_OCTET . '){3}\\b/'; 57 58 // Cache whois requests. Try to avoid all caches expiring at the same time. 59 private const WHOIS_TTL_MIN = 28 * 86400; 60 private const WHOIS_TTL_MAX = 35 * 86400; 61 private const WHOIS_TIMEOUT = 5; 62 63 // Bad robots - SEO optimisers, advertisers, etc. This list is shared with robots.txt. 64 public const BAD_ROBOTS = [ 65 'admantx', 66 'Adsbot', 67 'AhrefsBot', 68 'Amazonbot', // Until it understands crawl-delay and noindex / nofollow 69 'AspiegelBot', 70 'Awario', // Brand management 71 'Barkrowler', 72 'BLEXBot', 73 'Bytespider', 74 'DataForSEO', 75 'DataForSeoBot', // https://dataforseo.com/dataforseo-bot 76 'DotBot', 77 'gptbot', // Collects training data for ChatCPT 78 'Grapeshot', 79 'Honolulu-bot', // Aggressive crawer, no info available 80 'ia_archiver', 81 'linabot', // Aggressive crawer, no info available 82 'Linguee', 83 'MegaIndex.ru', 84 'MJ12bot', 85 'netEstate NE', 86 'panscient', 87 'PetalBot', 88 'proximic', 89 'SeekportBot', // Pretends to be a search engine - but isn't 90 'SemrushBot', 91 'serpstatbot', 92 'SEOkicks', 93 'SiteKiosk', 94 'Turnitin', 95 'wp_is_mobile', // Nothing to do with wordpress 96 'XoviBot', 97 'YisouSpider', 98 'ZoominfoBot', 99 ]; 100 101 /** 102 * Some search engines use reverse/forward DNS to verify the IP address. 103 * 104 * @see https://developer.amazon.com/support/amazonbot 105 * @see https://support.google.com/webmasters/answer/80553?hl=en 106 * @see https://www.bing.com/webmaster/help/which-crawlers-does-bing-use-8c184ec0 107 * @see https://www.bing.com/webmaster/help/how-to-verify-bingbot-3905dc26 108 * @see https://yandex.com/support/webmaster/robot-workings/check-yandex-robots.html 109 * @see https://www.mojeek.com/bot.html 110 * @see https://support.apple.com/en-gb/HT204683 111 */ 112 private const ROBOT_REV_FWD_DNS = [ 113 'Amazonbot' => ['.crawl.amazon.com'], 114 'Applebot' => ['.applebot.apple.com'], 115 'BingPreview' => ['.search.msn.com'], 116 'Google' => ['.google.com', '.googlebot.com'], 117 'Mail.RU_Bot' => ['.mail.ru'], 118 'MicrosoftPreview' => ['.search.msn.com'], 119 'MojeekBot' => ['.mojeek.com'], 120 'Qwantify' => ['.search.qwant.com'], 121 'Sogou' => ['.crawl.sogou.com'], 122 'Yahoo' => ['.crawl.yahoo.net'], 123 'Yandex' => ['.yandex.ru', '.yandex.net', '.yandex.com'], 124 'bingbot' => ['.search.msn.com'], 125 'msnbot' => ['.search.msn.com'], 126 ]; 127 128 /** 129 * Some search engines only use reverse DNS to verify the IP address. 130 * 131 * @see https://help.baidu.com/question?prod_id=99&class=0&id=3001 132 * @see https://napoveda.seznam.cz/en/full-text-search/seznambot-crawler 133 * @see https://www.ionos.de/terms-gtc/faq-crawler 134 */ 135 private const ROBOT_REV_ONLY_DNS = [ 136 'Baiduspider' => ['.baidu.com', '.baidu.jp'], 137 'FreshBot' => ['.seznam.cz'], 138 'IonCrawl' => ['.1und1.org'], 139 'Neevabot' => ['.neeva.com'], 140 'SeznamBot' => ['.seznam.cz'], 141 ]; 142 143 /** 144 * Some search engines operate from designated IP addresses. 145 * 146 * @see https://www.apple.com/go/applebot 147 * @see https://help.duckduckgo.com/duckduckgo-help-pages/results/duckduckbot 148 */ 149 private const ROBOT_IPS = [ 150 'AppleBot' => [ 151 '17.0.0.0/8', 152 ], 153 'Ask Jeeves' => [ 154 '65.214.45.143', 155 '65.214.45.148', 156 '66.235.124.192', 157 '66.235.124.7', 158 '66.235.124.101', 159 '66.235.124.193', 160 '66.235.124.73', 161 '66.235.124.196', 162 '66.235.124.74', 163 '63.123.238.8', 164 '202.143.148.61', 165 ], 166 'DuckDuckBot' => [ 167 '23.21.227.69', 168 '50.16.241.113', 169 '50.16.241.114', 170 '50.16.241.117', 171 '50.16.247.234', 172 '52.204.97.54', 173 '52.5.190.19', 174 '54.197.234.188', 175 '54.208.100.253', 176 '54.208.102.37', 177 '107.21.1.8', 178 ], 179 ]; 180 181 /** 182 * Some search engines operate from designated IP addresses. 183 * 184 * @see https://bot.seekport.com/ 185 */ 186 private const ROBOT_IP_FILES = [ 187 'SeekportBot' => 'https://bot.seekport.com/seekportbot_ips.txt', 188 ]; 189 190 /** 191 * Some search engines operate from within a designated autonomous system. 192 * 193 * @see https://developers.facebook.com/docs/sharing/webmasters/crawler 194 * @see https://www.facebook.com/peering/ 195 */ 196 private const ROBOT_ASNS = [ 197 'facebook' => ['AS32934', 'AS63293'], 198 'twitter' => ['AS13414'], 199 ]; 200 201 /** 202 * @param ServerRequestInterface $request 203 * @param RequestHandlerInterface $handler 204 * 205 * @return ResponseInterface 206 */ 207 public function process(ServerRequestInterface $request, RequestHandlerInterface $handler): ResponseInterface 208 { 209 $ua = Validator::serverParams($request)->string('HTTP_USER_AGENT', ''); 210 $ip = Validator::attributes($request)->string('client-ip'); 211 $address = IPFactory::parseAddressString($ip); 212 assert($address instanceof AddressInterface); 213 214 foreach (self::BAD_ROBOTS as $robot) { 215 if (str_contains($ua, $robot)) { 216 return $this->response(); 217 } 218 } 219 220 foreach (self::ROBOT_REV_FWD_DNS as $robot => $valid_domains) { 221 if (str_contains($ua, $robot) && !$this->checkRobotDNS($ip, $valid_domains, false)) { 222 return $this->response(); 223 } 224 } 225 226 foreach (self::ROBOT_REV_ONLY_DNS as $robot => $valid_domains) { 227 if (str_contains($ua, $robot) && !$this->checkRobotDNS($ip, $valid_domains, true)) { 228 return $this->response(); 229 } 230 } 231 232 foreach (self::ROBOT_IPS as $robot => $valid_ip_ranges) { 233 if (str_contains($ua, $robot)) { 234 foreach ($valid_ip_ranges as $ip_range) { 235 $range = IPFactory::parseRangeString($ip_range); 236 237 if ($range instanceof RangeInterface && $range->contains($address)) { 238 continue 2; 239 } 240 } 241 242 return $this->response(); 243 } 244 } 245 246 foreach (self::ROBOT_IP_FILES as $robot => $url) { 247 if (str_contains($ua, $robot)) { 248 $valid_ip_ranges = $this->fetchIpRangesForUrl($robot, $url); 249 250 foreach ($valid_ip_ranges as $ip_range) { 251 $range = IPFactory::parseRangeString($ip_range); 252 253 if ($range instanceof RangeInterface && $range->contains($address)) { 254 continue 2; 255 } 256 } 257 258 return $this->response(); 259 } 260 } 261 262 foreach (self::ROBOT_ASNS as $robot => $asns) { 263 foreach ($asns as $asn) { 264 if (str_contains($ua, $robot)) { 265 foreach ($this->fetchIpRangesForAsn($asn) as $range) { 266 if ($range->contains($address)) { 267 continue 2; 268 } 269 } 270 271 return $this->response(); 272 } 273 } 274 } 275 276 // Allow sites to block access from entire networks. 277 $block_asn = Validator::attributes($request)->string('block_asn', ''); 278 preg_match_all('/(AS\d+)/', $block_asn, $matches); 279 280 foreach ($matches[1] as $asn) { 281 foreach ($this->fetchIpRangesForAsn($asn) as $range) { 282 if ($range->contains($address)) { 283 return $this->response(); 284 } 285 } 286 } 287 288 return $handler->handle($request); 289 } 290 291 /** 292 * Check that an IP address belongs to a robot operator using a forward/reverse DNS lookup. 293 * 294 * @param string $ip 295 * @param array<string> $valid_domains 296 * @param bool $reverse_only 297 * 298 * @return bool 299 */ 300 private function checkRobotDNS(string $ip, array $valid_domains, bool $reverse_only): bool 301 { 302 $host = gethostbyaddr($ip); 303 304 if ($host === false) { 305 return false; 306 } 307 308 foreach ($valid_domains as $domain) { 309 if (str_ends_with($host, $domain)) { 310 return $reverse_only || $ip === gethostbyname($host); 311 } 312 } 313 314 return false; 315 } 316 317 /** 318 * Perform a whois search for an ASN. 319 * 320 * @param string $asn The autonomous system number to query 321 * 322 * @return array<RangeInterface> 323 */ 324 private function fetchIpRangesForAsn(string $asn): array 325 { 326 return Registry::cache()->file()->remember('whois-asn-' . $asn, static function () use ($asn): array { 327 $mapper = static fn (AsnRouteInfo $route_info): ?RangeInterface => IPFactory::parseRangeString($route_info->route ?: $route_info->route6); 328 329 try { 330 $loader = new CurlLoader(self::WHOIS_TIMEOUT); 331 $whois = new Whois($loader); 332 $info = $whois->loadAsnInfo($asn); 333 $routes = $info->routes; 334 $ranges = array_map($mapper, $routes); 335 336 return array_filter($ranges); 337 } catch (Throwable) { 338 return []; 339 } 340 }, random_int(self::WHOIS_TTL_MIN, self::WHOIS_TTL_MAX)); 341 } 342 343 /** 344 * Fetch a list of IP addresses from a remote file. 345 * 346 * @param string $ua 347 * @param string $url 348 * 349 * @return array<string> 350 */ 351 private function fetchIpRangesForUrl(string $ua, string $url): array 352 { 353 return Registry::cache()->file()->remember('url-ip-list-' . $ua, static function () use ($url): array { 354 try { 355 $client = new Client(); 356 $response = $client->get($url, ['timeout' => 5]); 357 $contents = $response->getBody()->getContents(); 358 359 preg_match_all(self::REGEX_IPV4, $contents, $matches); 360 361 return $matches[0]; 362 } catch (GuzzleException) { 363 return []; 364 } 365 }, random_int(self::WHOIS_TTL_MIN, self::WHOIS_TTL_MAX)); 366 } 367 368 /** 369 * @return ResponseInterface 370 */ 371 private function response(): ResponseInterface 372 { 373 return response('Not acceptable', StatusCodeInterface::STATUS_NOT_ACCEPTABLE); 374 } 375} 376