xref: /webtrees/app/Http/Middleware/BadBotBlocker.php (revision ee551e0b6d330340f2c6e961359bd855b89571fb)
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        'AntBot', // Aggressive crawler
70        'AspiegelBot',
71        'Awario', // Brand management
72        'Barkrowler', // Crawler for babbar.tech
73        'BLEXBot',
74        'Bytespider', // Aggressive crawler from Bytedance/TikTok
75        'CCBot', // Used to train a number of LLMs
76        'CensysInspect', // Vulnerability scanner
77        'ChatGPT-User', // Used by ChatGPT during operation
78        'ClaudeBot', // Collects training data for LLMs
79        'DataForSeoBot', // https://dataforseo.com/dataforseo-bot
80        'DotBot',
81        'Expanse', // Another pointless crawler
82        'FacebookBot', // Collects training data for Facebook's LLM translator.
83        'fidget-spinner-bot', // Agressive crawler
84        'Foregenix', // Vulnerability scanner
85        'Go-http-client', // Crawler library used by many bots
86        'Google-Extended', // Collects training data for Google Bard
87        'GPTBot', // Collects training data for ChatGPT
88        'Grapeshot',
89        'Honolulu-bot', // Aggressive crawer, no info available
90        'ia_archiver',
91        'internet-measurement', // Driftnet
92        'IonCrawl',
93        'Java', // Crawler library used by many bots
94        'linabot', // Aggressive crawer, no info available
95        'Linguee',
96        'MegaIndex.ru',
97        'MJ12bot',
98        'netEstate NE',
99        'Omgilibot', // Collects training data for LLMs
100        'panscient',
101        'PetalBot',
102        'proximic',
103        'python-requests', // Crawler library used by many bots
104        'Scrapy', // Scraping tool
105        'SeekportBot', // Pretends to be a search engine - but isn't
106        'SemrushBot',
107        'serpstatbot',
108        'SEOkicks',
109        'SiteKiosk',
110        'test-bot', // Agressive crawler
111        'TinyTestBot',
112        'Turnitin',
113        'wp_is_mobile', // Nothing to do with wordpress
114        'XoviBot',
115        'YisouSpider',
116        'ZoominfoBot',
117    ];
118
119    /**
120     * Some search engines use reverse/forward DNS to verify the IP address.
121     *
122     * @see https://developer.amazon.com/support/amazonbot
123     * @see https://support.google.com/webmasters/answer/80553?hl=en
124     * @see https://www.bing.com/webmaster/help/which-crawlers-does-bing-use-8c184ec0
125     * @see https://www.bing.com/webmaster/help/how-to-verify-bingbot-3905dc26
126     * @see https://yandex.com/support/webmaster/robot-workings/check-yandex-robots.html
127     * @see https://www.mojeek.com/bot.html
128     * @see https://support.apple.com/en-gb/HT204683
129     */
130    private const ROBOT_REV_FWD_DNS = [
131        'Amazonbot'        => ['.crawl.amazon.com'],
132        'Applebot'         => ['.applebot.apple.com'],
133        'BingPreview'      => ['.search.msn.com'],
134        'Google'           => ['.google.com', '.googlebot.com'],
135        'Mail.RU_Bot'      => ['.mail.ru'],
136        'MicrosoftPreview' => ['.search.msn.com'],
137        'MojeekBot'        => ['.mojeek.com'],
138        'Qwantify'         => ['.qwant.com'],
139        'Sogou'            => ['.crawl.sogou.com'],
140        'Yahoo'            => ['.crawl.yahoo.net'],
141        'Yandex'           => ['.yandex.ru', '.yandex.net', '.yandex.com'],
142        'bingbot'          => ['.search.msn.com'],
143        'msnbot'           => ['.search.msn.com'],
144    ];
145
146    /**
147     * Some search engines only use reverse DNS to verify the IP address.
148     *
149     * @see https://help.baidu.com/question?prod_id=99&class=0&id=3001
150     * @see https://napoveda.seznam.cz/en/full-text-search/seznambot-crawler
151     * @see https://www.ionos.de/terms-gtc/faq-crawler
152     */
153    private const ROBOT_REV_ONLY_DNS = [
154        'Baiduspider' => ['.baidu.com', '.baidu.jp'],
155        'FreshBot'    => ['.seznam.cz'],
156        'IonCrawl'    => ['.1und1.org'],
157        'Neevabot'    => ['.neeva.com'],
158        'SeznamBot'   => ['.seznam.cz'],
159    ];
160
161    /**
162     * Some search engines operate from designated IP addresses.
163     *
164     * @see https://www.apple.com/go/applebot
165     * @see https://help.duckduckgo.com/duckduckgo-help-pages/results/duckduckbot
166     */
167    private const ROBOT_IPS = [
168        'AppleBot'    => [
169            '17.0.0.0/8',
170        ],
171        'Ask Jeeves'  => [
172            '65.214.45.143',
173            '65.214.45.148',
174            '66.235.124.192',
175            '66.235.124.7',
176            '66.235.124.101',
177            '66.235.124.193',
178            '66.235.124.73',
179            '66.235.124.196',
180            '66.235.124.74',
181            '63.123.238.8',
182            '202.143.148.61',
183        ],
184        'DuckDuckBot' => [
185            '23.21.227.69',
186            '50.16.241.113',
187            '50.16.241.114',
188            '50.16.241.117',
189            '50.16.247.234',
190            '52.204.97.54',
191            '52.5.190.19',
192            '54.197.234.188',
193            '54.208.100.253',
194            '54.208.102.37',
195            '107.21.1.8',
196        ],
197    ];
198
199    /**
200     * Some search engines operate from designated IP addresses.
201     *
202     * @see https://bot.seekport.com/
203     */
204    private const ROBOT_IP_FILES = [
205        'SeekportBot' => 'https://bot.seekport.com/seekportbot_ips.txt',
206    ];
207
208    /**
209     * Some search engines operate from within a designated autonomous system.
210     *
211     * @see https://developers.facebook.com/docs/sharing/webmasters/crawler
212     * @see https://www.facebook.com/peering/
213     */
214    private const ROBOT_ASNS = [
215        'facebook' => ['AS32934', 'AS63293'],
216        'twitter'  => ['AS13414'],
217    ];
218
219    /**
220     * @param ServerRequestInterface  $request
221     * @param RequestHandlerInterface $handler
222     *
223     * @return ResponseInterface
224     */
225    public function process(ServerRequestInterface $request, RequestHandlerInterface $handler): ResponseInterface
226    {
227        $ua      = Validator::serverParams($request)->string('HTTP_USER_AGENT', '');
228        $ip      = Validator::attributes($request)->string('client-ip');
229        $address = IPFactory::parseAddressString($ip);
230        assert($address instanceof AddressInterface);
231
232        foreach (self::BAD_ROBOTS as $robot) {
233            if (str_contains($ua, $robot)) {
234                return $this->response();
235            }
236        }
237
238        foreach (self::ROBOT_REV_FWD_DNS as $robot => $valid_domains) {
239            if (str_contains($ua, $robot) && !$this->checkRobotDNS($ip, $valid_domains, false)) {
240                return $this->response();
241            }
242        }
243
244        foreach (self::ROBOT_REV_ONLY_DNS as $robot => $valid_domains) {
245            if (str_contains($ua, $robot) && !$this->checkRobotDNS($ip, $valid_domains, true)) {
246                return $this->response();
247            }
248        }
249
250        foreach (self::ROBOT_IPS as $robot => $valid_ip_ranges) {
251            if (str_contains($ua, $robot)) {
252                foreach ($valid_ip_ranges as $ip_range) {
253                    $range = IPFactory::parseRangeString($ip_range);
254
255                    if ($range instanceof RangeInterface && $range->contains($address)) {
256                        continue 2;
257                    }
258                }
259
260                return $this->response();
261            }
262        }
263
264        foreach (self::ROBOT_IP_FILES as $robot => $url) {
265            if (str_contains($ua, $robot)) {
266                $valid_ip_ranges = $this->fetchIpRangesForUrl($robot, $url);
267
268                foreach ($valid_ip_ranges as $ip_range) {
269                    $range = IPFactory::parseRangeString($ip_range);
270
271                    if ($range instanceof RangeInterface && $range->contains($address)) {
272                        continue 2;
273                    }
274                }
275
276                return $this->response();
277            }
278        }
279
280        foreach (self::ROBOT_ASNS as $robot => $asns) {
281            foreach ($asns as $asn) {
282                if (str_contains($ua, $robot)) {
283                    foreach ($this->fetchIpRangesForAsn($asn) as $range) {
284                        if ($range->contains($address)) {
285                            continue 2;
286                        }
287                    }
288
289                    return $this->response();
290                }
291            }
292        }
293
294        // Allow sites to block access from entire networks.
295        $block_asn = Validator::attributes($request)->string('block_asn', '');
296        preg_match_all('/(AS\d+)/', $block_asn, $matches);
297
298        foreach ($matches[1] as $asn) {
299            foreach ($this->fetchIpRangesForAsn($asn) as $range) {
300                if ($range->contains($address)) {
301                    return $this->response();
302                }
303            }
304        }
305
306        return $handler->handle($request);
307    }
308
309    /**
310     * Check that an IP address belongs to a robot operator using a forward/reverse DNS lookup.
311     *
312     * @param string        $ip
313     * @param array<string> $valid_domains
314     * @param bool          $reverse_only
315     *
316     * @return bool
317     */
318    private function checkRobotDNS(string $ip, array $valid_domains, bool $reverse_only): bool
319    {
320        $host = gethostbyaddr($ip);
321
322        if ($host === false) {
323            return false;
324        }
325
326        foreach ($valid_domains as $domain) {
327            if (str_ends_with($host, $domain)) {
328                return $reverse_only || $ip === gethostbyname($host);
329            }
330        }
331
332        return false;
333    }
334
335    /**
336     * Perform a whois search for an ASN.
337     *
338     * @param string $asn The autonomous system number to query
339     *
340     * @return array<RangeInterface>
341     */
342    private function fetchIpRangesForAsn(string $asn): array
343    {
344        return Registry::cache()->file()->remember('whois-asn-' . $asn, static function () use ($asn): array {
345            $mapper = static fn (AsnRouteInfo $route_info): ?RangeInterface => IPFactory::parseRangeString($route_info->route ?: $route_info->route6);
346
347            try {
348                $loader = new CurlLoader(self::WHOIS_TIMEOUT);
349                $whois  = new Whois($loader);
350                $info   = $whois->loadAsnInfo($asn);
351                $routes = $info->routes;
352                $ranges = array_map($mapper, $routes);
353
354                return array_filter($ranges);
355            } catch (Throwable) {
356                return [];
357            }
358        }, random_int(self::WHOIS_TTL_MIN, self::WHOIS_TTL_MAX));
359    }
360
361    /**
362     * Fetch a list of IP addresses from a remote file.
363     *
364     * @param string $ua
365     * @param string $url
366     *
367     * @return array<string>
368     */
369    private function fetchIpRangesForUrl(string $ua, string $url): array
370    {
371        return Registry::cache()->file()->remember('url-ip-list-' . $ua, static function () use ($url): array {
372            try {
373                $client   = new Client();
374                $response = $client->get($url, ['timeout' => 5]);
375                $contents = $response->getBody()->getContents();
376
377                preg_match_all(self::REGEX_IPV4, $contents, $matches);
378
379                return $matches[0];
380            } catch (GuzzleException) {
381                return [];
382            }
383        }, random_int(self::WHOIS_TTL_MIN, self::WHOIS_TTL_MAX));
384    }
385
386    /**
387     * @return ResponseInterface
388     */
389    private function response(): ResponseInterface
390    {
391        return response('Not acceptable', StatusCodeInterface::STATUS_NOT_ACCEPTABLE);
392    }
393}
394