xref: /webtrees/app/Http/Middleware/BadBotBlocker.php (revision 58a588c63196cb1f4cc1926f6fe552c754b0d27a)
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',
73        'BLEXBot',
74        'Bytespider',
75        'CCBot', // Used to train a number of LLMs
76        'ChatGPT-User', // Used by ChatGPT during operation
77        'ClaudeBot', // Collects training data for LLMs
78        'DataForSeoBot', // https://dataforseo.com/dataforseo-bot
79        'DotBot',
80        'FacebookBot', // Collects training data for Facebook's LLM translator.
81        'fidget-spinner-bot', // Agressive crawler
82        'Google-Extended', // Collects training data for Google Bard
83        'GPTBot', // Collects training data for ChatGPT
84        'Grapeshot',
85        'Honolulu-bot', // Aggressive crawer, no info available
86        'ia_archiver',
87        'linabot', // Aggressive crawer, no info available
88        'Linguee',
89        'MegaIndex.ru',
90        'MJ12bot',
91        'netEstate NE',
92        'Omgilibot', // Collects training data for LLMs
93        'panscient',
94        'PetalBot',
95        'proximic',
96        'SeekportBot', // Pretends to be a search engine - but isn't
97        'SemrushBot',
98        'serpstatbot',
99        'SEOkicks',
100        'SiteKiosk',
101        'test-bot', // Agressive crawler
102        'TinyTestBot',
103        'Turnitin',
104        'wp_is_mobile', // Nothing to do with wordpress
105        'XoviBot',
106        'YisouSpider',
107        'ZoominfoBot',
108    ];
109
110    /**
111     * Some search engines use reverse/forward DNS to verify the IP address.
112     *
113     * @see https://developer.amazon.com/support/amazonbot
114     * @see https://support.google.com/webmasters/answer/80553?hl=en
115     * @see https://www.bing.com/webmaster/help/which-crawlers-does-bing-use-8c184ec0
116     * @see https://www.bing.com/webmaster/help/how-to-verify-bingbot-3905dc26
117     * @see https://yandex.com/support/webmaster/robot-workings/check-yandex-robots.html
118     * @see https://www.mojeek.com/bot.html
119     * @see https://support.apple.com/en-gb/HT204683
120     */
121    private const ROBOT_REV_FWD_DNS = [
122        'Amazonbot'        => ['.crawl.amazon.com'],
123        'Applebot'         => ['.applebot.apple.com'],
124        'BingPreview'      => ['.search.msn.com'],
125        'Google'           => ['.google.com', '.googlebot.com'],
126        'Mail.RU_Bot'      => ['.mail.ru'],
127        'MicrosoftPreview' => ['.search.msn.com'],
128        'MojeekBot'        => ['.mojeek.com'],
129        'Qwantify'         => ['.qwant.com'],
130        'Sogou'            => ['.crawl.sogou.com'],
131        'Yahoo'            => ['.crawl.yahoo.net'],
132        'Yandex'           => ['.yandex.ru', '.yandex.net', '.yandex.com'],
133        'bingbot'          => ['.search.msn.com'],
134        'msnbot'           => ['.search.msn.com'],
135    ];
136
137    /**
138     * Some search engines only use reverse DNS to verify the IP address.
139     *
140     * @see https://help.baidu.com/question?prod_id=99&class=0&id=3001
141     * @see https://napoveda.seznam.cz/en/full-text-search/seznambot-crawler
142     * @see https://www.ionos.de/terms-gtc/faq-crawler
143     */
144    private const ROBOT_REV_ONLY_DNS = [
145        'Baiduspider' => ['.baidu.com', '.baidu.jp'],
146        'FreshBot'    => ['.seznam.cz'],
147        'IonCrawl'    => ['.1und1.org'],
148        'Neevabot'    => ['.neeva.com'],
149        'SeznamBot'   => ['.seznam.cz'],
150    ];
151
152    /**
153     * Some search engines operate from designated IP addresses.
154     *
155     * @see https://www.apple.com/go/applebot
156     * @see https://help.duckduckgo.com/duckduckgo-help-pages/results/duckduckbot
157     */
158    private const ROBOT_IPS = [
159        'AppleBot'    => [
160            '17.0.0.0/8',
161        ],
162        'Ask Jeeves'  => [
163            '65.214.45.143',
164            '65.214.45.148',
165            '66.235.124.192',
166            '66.235.124.7',
167            '66.235.124.101',
168            '66.235.124.193',
169            '66.235.124.73',
170            '66.235.124.196',
171            '66.235.124.74',
172            '63.123.238.8',
173            '202.143.148.61',
174        ],
175        'DuckDuckBot' => [
176            '23.21.227.69',
177            '50.16.241.113',
178            '50.16.241.114',
179            '50.16.241.117',
180            '50.16.247.234',
181            '52.204.97.54',
182            '52.5.190.19',
183            '54.197.234.188',
184            '54.208.100.253',
185            '54.208.102.37',
186            '107.21.1.8',
187        ],
188    ];
189
190    /**
191     * Some search engines operate from designated IP addresses.
192     *
193     * @see https://bot.seekport.com/
194     */
195    private const ROBOT_IP_FILES = [
196        'SeekportBot' => 'https://bot.seekport.com/seekportbot_ips.txt',
197    ];
198
199    /**
200     * Some search engines operate from within a designated autonomous system.
201     *
202     * @see https://developers.facebook.com/docs/sharing/webmasters/crawler
203     * @see https://www.facebook.com/peering/
204     */
205    private const ROBOT_ASNS = [
206        'facebook' => ['AS32934', 'AS63293'],
207        'twitter'  => ['AS13414'],
208    ];
209
210    /**
211     * @param ServerRequestInterface  $request
212     * @param RequestHandlerInterface $handler
213     *
214     * @return ResponseInterface
215     */
216    public function process(ServerRequestInterface $request, RequestHandlerInterface $handler): ResponseInterface
217    {
218        $ua      = Validator::serverParams($request)->string('HTTP_USER_AGENT', '');
219        $ip      = Validator::attributes($request)->string('client-ip');
220        $address = IPFactory::parseAddressString($ip);
221        assert($address instanceof AddressInterface);
222
223        foreach (self::BAD_ROBOTS as $robot) {
224            if (str_contains($ua, $robot)) {
225                return $this->response();
226            }
227        }
228
229        foreach (self::ROBOT_REV_FWD_DNS as $robot => $valid_domains) {
230            if (str_contains($ua, $robot) && !$this->checkRobotDNS($ip, $valid_domains, false)) {
231                return $this->response();
232            }
233        }
234
235        foreach (self::ROBOT_REV_ONLY_DNS as $robot => $valid_domains) {
236            if (str_contains($ua, $robot) && !$this->checkRobotDNS($ip, $valid_domains, true)) {
237                return $this->response();
238            }
239        }
240
241        foreach (self::ROBOT_IPS as $robot => $valid_ip_ranges) {
242            if (str_contains($ua, $robot)) {
243                foreach ($valid_ip_ranges as $ip_range) {
244                    $range = IPFactory::parseRangeString($ip_range);
245
246                    if ($range instanceof RangeInterface && $range->contains($address)) {
247                        continue 2;
248                    }
249                }
250
251                return $this->response();
252            }
253        }
254
255        foreach (self::ROBOT_IP_FILES as $robot => $url) {
256            if (str_contains($ua, $robot)) {
257                $valid_ip_ranges = $this->fetchIpRangesForUrl($robot, $url);
258
259                foreach ($valid_ip_ranges as $ip_range) {
260                    $range = IPFactory::parseRangeString($ip_range);
261
262                    if ($range instanceof RangeInterface && $range->contains($address)) {
263                        continue 2;
264                    }
265                }
266
267                return $this->response();
268            }
269        }
270
271        foreach (self::ROBOT_ASNS as $robot => $asns) {
272            foreach ($asns as $asn) {
273                if (str_contains($ua, $robot)) {
274                    foreach ($this->fetchIpRangesForAsn($asn) as $range) {
275                        if ($range->contains($address)) {
276                            continue 2;
277                        }
278                    }
279
280                    return $this->response();
281                }
282            }
283        }
284
285        // Allow sites to block access from entire networks.
286        $block_asn = Validator::attributes($request)->string('block_asn', '');
287        preg_match_all('/(AS\d+)/', $block_asn, $matches);
288
289        foreach ($matches[1] as $asn) {
290            foreach ($this->fetchIpRangesForAsn($asn) as $range) {
291                if ($range->contains($address)) {
292                    return $this->response();
293                }
294            }
295        }
296
297        return $handler->handle($request);
298    }
299
300    /**
301     * Check that an IP address belongs to a robot operator using a forward/reverse DNS lookup.
302     *
303     * @param string        $ip
304     * @param array<string> $valid_domains
305     * @param bool          $reverse_only
306     *
307     * @return bool
308     */
309    private function checkRobotDNS(string $ip, array $valid_domains, bool $reverse_only): bool
310    {
311        $host = gethostbyaddr($ip);
312
313        if ($host === false) {
314            return false;
315        }
316
317        foreach ($valid_domains as $domain) {
318            if (str_ends_with($host, $domain)) {
319                return $reverse_only || $ip === gethostbyname($host);
320            }
321        }
322
323        return false;
324    }
325
326    /**
327     * Perform a whois search for an ASN.
328     *
329     * @param string $asn The autonomous system number to query
330     *
331     * @return array<RangeInterface>
332     */
333    private function fetchIpRangesForAsn(string $asn): array
334    {
335        return Registry::cache()->file()->remember('whois-asn-' . $asn, static function () use ($asn): array {
336            $mapper = static fn (AsnRouteInfo $route_info): ?RangeInterface => IPFactory::parseRangeString($route_info->route ?: $route_info->route6);
337
338            try {
339                $loader = new CurlLoader(self::WHOIS_TIMEOUT);
340                $whois  = new Whois($loader);
341                $info   = $whois->loadAsnInfo($asn);
342                $routes = $info->routes;
343                $ranges = array_map($mapper, $routes);
344
345                return array_filter($ranges);
346            } catch (Throwable) {
347                return [];
348            }
349        }, random_int(self::WHOIS_TTL_MIN, self::WHOIS_TTL_MAX));
350    }
351
352    /**
353     * Fetch a list of IP addresses from a remote file.
354     *
355     * @param string $ua
356     * @param string $url
357     *
358     * @return array<string>
359     */
360    private function fetchIpRangesForUrl(string $ua, string $url): array
361    {
362        return Registry::cache()->file()->remember('url-ip-list-' . $ua, static function () use ($url): array {
363            try {
364                $client   = new Client();
365                $response = $client->get($url, ['timeout' => 5]);
366                $contents = $response->getBody()->getContents();
367
368                preg_match_all(self::REGEX_IPV4, $contents, $matches);
369
370                return $matches[0];
371            } catch (GuzzleException) {
372                return [];
373            }
374        }, random_int(self::WHOIS_TTL_MIN, self::WHOIS_TTL_MAX));
375    }
376
377    /**
378     * @return ResponseInterface
379     */
380    private function response(): ResponseInterface
381    {
382        return response('Not acceptable', StatusCodeInterface::STATUS_NOT_ACCEPTABLE);
383    }
384}
385