xref: /webtrees/app/Http/Middleware/BadBotBlocker.php (revision 96126da4647bfcf678ecdfc78fb14877280efec0)
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        'phxbot', // Badly written crawler
103        'proximic',
104        'python-requests', // Crawler library used by many bots
105        'Scrapy', // Scraping tool
106        'SeekportBot', // Pretends to be a search engine - but isn't
107        'SemrushBot',
108        'serpstatbot',
109        'SEOkicks',
110        'SiteKiosk',
111        'test-bot', // Agressive crawler
112        'TinyTestBot',
113        'Turnitin',
114        'wp_is_mobile', // Nothing to do with wordpress
115        'XoviBot',
116        'YisouSpider',
117        'ZoominfoBot',
118    ];
119
120    /**
121     * Some search engines use reverse/forward DNS to verify the IP address.
122     *
123     * @see https://developer.amazon.com/support/amazonbot
124     * @see https://support.google.com/webmasters/answer/80553?hl=en
125     * @see https://www.bing.com/webmaster/help/which-crawlers-does-bing-use-8c184ec0
126     * @see https://www.bing.com/webmaster/help/how-to-verify-bingbot-3905dc26
127     * @see https://yandex.com/support/webmaster/robot-workings/check-yandex-robots.html
128     * @see https://www.mojeek.com/bot.html
129     * @see https://support.apple.com/en-gb/HT204683
130     */
131    private const ROBOT_REV_FWD_DNS = [
132        'Amazonbot'        => ['.crawl.amazon.com'],
133        'Applebot'         => ['.applebot.apple.com'],
134        'BingPreview'      => ['.search.msn.com'],
135        'Google'           => ['.google.com', '.googlebot.com'],
136        'Mail.RU_Bot'      => ['.mail.ru'],
137        'MicrosoftPreview' => ['.search.msn.com'],
138        'MojeekBot'        => ['.mojeek.com'],
139        'Qwantify'         => ['.qwant.com'],
140        'Sogou'            => ['.crawl.sogou.com'],
141        'Yahoo'            => ['.crawl.yahoo.net'],
142        'Yandex'           => ['.yandex.ru', '.yandex.net', '.yandex.com'],
143        'bingbot'          => ['.search.msn.com'],
144        'msnbot'           => ['.search.msn.com'],
145    ];
146
147    /**
148     * Some search engines only use reverse DNS to verify the IP address.
149     *
150     * @see https://help.baidu.com/question?prod_id=99&class=0&id=3001
151     * @see https://napoveda.seznam.cz/en/full-text-search/seznambot-crawler
152     * @see https://www.ionos.de/terms-gtc/faq-crawler
153     */
154    private const ROBOT_REV_ONLY_DNS = [
155        'Baiduspider' => ['.baidu.com', '.baidu.jp'],
156        'FreshBot'    => ['.seznam.cz'],
157        'IonCrawl'    => ['.1und1.org'],
158        'Neevabot'    => ['.neeva.com'],
159        'SeznamBot'   => ['.seznam.cz'],
160    ];
161
162    /**
163     * Some search engines operate from designated IP addresses.
164     *
165     * @see https://www.apple.com/go/applebot
166     * @see https://help.duckduckgo.com/duckduckgo-help-pages/results/duckduckbot
167     */
168    private const ROBOT_IPS = [
169        'AppleBot'    => [
170            '17.0.0.0/8',
171        ],
172        'Ask Jeeves'  => [
173            '65.214.45.143',
174            '65.214.45.148',
175            '66.235.124.192',
176            '66.235.124.7',
177            '66.235.124.101',
178            '66.235.124.193',
179            '66.235.124.73',
180            '66.235.124.196',
181            '66.235.124.74',
182            '63.123.238.8',
183            '202.143.148.61',
184        ],
185        'DuckDuckBot' => [
186            '23.21.227.69',
187            '50.16.241.113',
188            '50.16.241.114',
189            '50.16.241.117',
190            '50.16.247.234',
191            '52.204.97.54',
192            '52.5.190.19',
193            '54.197.234.188',
194            '54.208.100.253',
195            '54.208.102.37',
196            '107.21.1.8',
197        ],
198    ];
199
200    /**
201     * Some search engines operate from designated IP addresses.
202     *
203     * @see https://bot.seekport.com/
204     */
205    private const ROBOT_IP_FILES = [
206        'SeekportBot' => 'https://bot.seekport.com/seekportbot_ips.txt',
207    ];
208
209    /**
210     * Some search engines operate from within a designated autonomous system.
211     *
212     * @see https://developers.facebook.com/docs/sharing/webmasters/crawler
213     * @see https://www.facebook.com/peering/
214     */
215    private const ROBOT_ASNS = [
216        'facebook' => ['AS32934', 'AS63293'],
217        'twitter'  => ['AS13414'],
218    ];
219
220    /**
221     * @param ServerRequestInterface  $request
222     * @param RequestHandlerInterface $handler
223     *
224     * @return ResponseInterface
225     */
226    public function process(ServerRequestInterface $request, RequestHandlerInterface $handler): ResponseInterface
227    {
228        $ua      = Validator::serverParams($request)->string('HTTP_USER_AGENT', '');
229        $ip      = Validator::attributes($request)->string('client-ip');
230        $address = IPFactory::parseAddressString($ip);
231        assert($address instanceof AddressInterface);
232
233        foreach (self::BAD_ROBOTS as $robot) {
234            if (str_contains($ua, $robot)) {
235                return $this->response();
236            }
237        }
238
239        foreach (self::ROBOT_REV_FWD_DNS as $robot => $valid_domains) {
240            if (str_contains($ua, $robot) && !$this->checkRobotDNS($ip, $valid_domains, false)) {
241                return $this->response();
242            }
243        }
244
245        foreach (self::ROBOT_REV_ONLY_DNS as $robot => $valid_domains) {
246            if (str_contains($ua, $robot) && !$this->checkRobotDNS($ip, $valid_domains, true)) {
247                return $this->response();
248            }
249        }
250
251        foreach (self::ROBOT_IPS as $robot => $valid_ip_ranges) {
252            if (str_contains($ua, $robot)) {
253                foreach ($valid_ip_ranges as $ip_range) {
254                    $range = IPFactory::parseRangeString($ip_range);
255
256                    if ($range instanceof RangeInterface && $range->contains($address)) {
257                        continue 2;
258                    }
259                }
260
261                return $this->response();
262            }
263        }
264
265        foreach (self::ROBOT_IP_FILES as $robot => $url) {
266            if (str_contains($ua, $robot)) {
267                $valid_ip_ranges = $this->fetchIpRangesForUrl($robot, $url);
268
269                foreach ($valid_ip_ranges as $ip_range) {
270                    $range = IPFactory::parseRangeString($ip_range);
271
272                    if ($range instanceof RangeInterface && $range->contains($address)) {
273                        continue 2;
274                    }
275                }
276
277                return $this->response();
278            }
279        }
280
281        foreach (self::ROBOT_ASNS as $robot => $asns) {
282            foreach ($asns as $asn) {
283                if (str_contains($ua, $robot)) {
284                    foreach ($this->fetchIpRangesForAsn($asn) as $range) {
285                        if ($range->contains($address)) {
286                            continue 2;
287                        }
288                    }
289
290                    return $this->response();
291                }
292            }
293        }
294
295        // Allow sites to block access from entire networks.
296        $block_asn = Validator::attributes($request)->string('block_asn', '');
297        preg_match_all('/(AS\d+)/', $block_asn, $matches);
298
299        foreach ($matches[1] as $asn) {
300            foreach ($this->fetchIpRangesForAsn($asn) as $range) {
301                if ($range->contains($address)) {
302                    return $this->response();
303                }
304            }
305        }
306
307        return $handler->handle($request);
308    }
309
310    /**
311     * Check that an IP address belongs to a robot operator using a forward/reverse DNS lookup.
312     *
313     * @param string        $ip
314     * @param array<string> $valid_domains
315     * @param bool          $reverse_only
316     *
317     * @return bool
318     */
319    private function checkRobotDNS(string $ip, array $valid_domains, bool $reverse_only): bool
320    {
321        $host = gethostbyaddr($ip);
322
323        if ($host === false) {
324            return false;
325        }
326
327        foreach ($valid_domains as $domain) {
328            if (str_ends_with($host, $domain)) {
329                return $reverse_only || $ip === gethostbyname($host);
330            }
331        }
332
333        return false;
334    }
335
336    /**
337     * Perform a whois search for an ASN.
338     *
339     * @param string $asn The autonomous system number to query
340     *
341     * @return array<RangeInterface>
342     */
343    private function fetchIpRangesForAsn(string $asn): array
344    {
345        return Registry::cache()->file()->remember('whois-asn-' . $asn, static function () use ($asn): array {
346            $mapper = static fn (AsnRouteInfo $route_info): ?RangeInterface => IPFactory::parseRangeString($route_info->route ?: $route_info->route6);
347
348            try {
349                $loader = new CurlLoader(self::WHOIS_TIMEOUT);
350                $whois  = new Whois($loader);
351                $info   = $whois->loadAsnInfo($asn);
352                $routes = $info->routes;
353                $ranges = array_map($mapper, $routes);
354
355                return array_filter($ranges);
356            } catch (Throwable) {
357                return [];
358            }
359        }, random_int(self::WHOIS_TTL_MIN, self::WHOIS_TTL_MAX));
360    }
361
362    /**
363     * Fetch a list of IP addresses from a remote file.
364     *
365     * @param string $ua
366     * @param string $url
367     *
368     * @return array<string>
369     */
370    private function fetchIpRangesForUrl(string $ua, string $url): array
371    {
372        return Registry::cache()->file()->remember('url-ip-list-' . $ua, static function () use ($url): array {
373            try {
374                $client   = new Client();
375                $response = $client->get($url, ['timeout' => 5]);
376                $contents = $response->getBody()->getContents();
377
378                preg_match_all(self::REGEX_IPV4, $contents, $matches);
379
380                return $matches[0];
381            } catch (GuzzleException) {
382                return [];
383            }
384        }, random_int(self::WHOIS_TTL_MIN, self::WHOIS_TTL_MAX));
385    }
386
387    /**
388     * @return ResponseInterface
389     */
390    private function response(): ResponseInterface
391    {
392        return response('Not acceptable', StatusCodeInterface::STATUS_NOT_ACCEPTABLE);
393    }
394}
395