xref: /webtrees/app/Http/Middleware/BadBotBlocker.php (revision 1dc9522fdcf3e2d9a2344f57cec6274a46efc3d5)
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        'DataForSeoBot', // https://dataforseo.com/dataforseo-bot
78        'DotBot',
79        'FacebookBot', // Collects training data for Facebook's LLM translator.
80        'fidget-spinner-bot', // Agressive crawler
81        'Google-Extended', // Collects training data for Google Bard
82        'GPTBot', // Collects training data for ChatGPT
83        'Grapeshot',
84        'Honolulu-bot', // Aggressive crawer, no info available
85        'ia_archiver',
86        'linabot', // Aggressive crawer, no info available
87        'Linguee',
88        'MegaIndex.ru',
89        'MJ12bot',
90        'netEstate NE',
91        'Omgilibot', // Collects training data for LLMs
92        'panscient',
93        'PetalBot',
94        'proximic',
95        'SeekportBot', // Pretends to be a search engine - but isn't
96        'SemrushBot',
97        'serpstatbot',
98        'SEOkicks',
99        'SiteKiosk',
100        'test-bot', // Agressive crawler
101        'TinyTestBot',
102        'Turnitin',
103        'wp_is_mobile', // Nothing to do with wordpress
104        'XoviBot',
105        'YisouSpider',
106        'ZoominfoBot',
107    ];
108
109    /**
110     * Some search engines use reverse/forward DNS to verify the IP address.
111     *
112     * @see https://developer.amazon.com/support/amazonbot
113     * @see https://support.google.com/webmasters/answer/80553?hl=en
114     * @see https://www.bing.com/webmaster/help/which-crawlers-does-bing-use-8c184ec0
115     * @see https://www.bing.com/webmaster/help/how-to-verify-bingbot-3905dc26
116     * @see https://yandex.com/support/webmaster/robot-workings/check-yandex-robots.html
117     * @see https://www.mojeek.com/bot.html
118     * @see https://support.apple.com/en-gb/HT204683
119     */
120    private const ROBOT_REV_FWD_DNS = [
121        'Amazonbot'        => ['.crawl.amazon.com'],
122        'Applebot'         => ['.applebot.apple.com'],
123        'BingPreview'      => ['.search.msn.com'],
124        'Google'           => ['.google.com', '.googlebot.com'],
125        'Mail.RU_Bot'      => ['.mail.ru'],
126        'MicrosoftPreview' => ['.search.msn.com'],
127        'MojeekBot'        => ['.mojeek.com'],
128        'Qwantify'         => ['.search.qwant.com'],
129        'Sogou'            => ['.crawl.sogou.com'],
130        'Yahoo'            => ['.crawl.yahoo.net'],
131        'Yandex'           => ['.yandex.ru', '.yandex.net', '.yandex.com'],
132        'bingbot'          => ['.search.msn.com'],
133        'msnbot'           => ['.search.msn.com'],
134    ];
135
136    /**
137     * Some search engines only use reverse DNS to verify the IP address.
138     *
139     * @see https://help.baidu.com/question?prod_id=99&class=0&id=3001
140     * @see https://napoveda.seznam.cz/en/full-text-search/seznambot-crawler
141     * @see https://www.ionos.de/terms-gtc/faq-crawler
142     */
143    private const ROBOT_REV_ONLY_DNS = [
144        'Baiduspider' => ['.baidu.com', '.baidu.jp'],
145        'FreshBot'    => ['.seznam.cz'],
146        'IonCrawl'    => ['.1und1.org'],
147        'Neevabot'    => ['.neeva.com'],
148        'SeznamBot'   => ['.seznam.cz'],
149    ];
150
151    /**
152     * Some search engines operate from designated IP addresses.
153     *
154     * @see https://www.apple.com/go/applebot
155     * @see https://help.duckduckgo.com/duckduckgo-help-pages/results/duckduckbot
156     */
157    private const ROBOT_IPS = [
158        'AppleBot'    => [
159            '17.0.0.0/8',
160        ],
161        'Ask Jeeves'  => [
162            '65.214.45.143',
163            '65.214.45.148',
164            '66.235.124.192',
165            '66.235.124.7',
166            '66.235.124.101',
167            '66.235.124.193',
168            '66.235.124.73',
169            '66.235.124.196',
170            '66.235.124.74',
171            '63.123.238.8',
172            '202.143.148.61',
173        ],
174        'DuckDuckBot' => [
175            '23.21.227.69',
176            '50.16.241.113',
177            '50.16.241.114',
178            '50.16.241.117',
179            '50.16.247.234',
180            '52.204.97.54',
181            '52.5.190.19',
182            '54.197.234.188',
183            '54.208.100.253',
184            '54.208.102.37',
185            '107.21.1.8',
186        ],
187    ];
188
189    /**
190     * Some search engines operate from designated IP addresses.
191     *
192     * @see https://bot.seekport.com/
193     */
194    private const ROBOT_IP_FILES = [
195        'SeekportBot' => 'https://bot.seekport.com/seekportbot_ips.txt',
196    ];
197
198    /**
199     * Some search engines operate from within a designated autonomous system.
200     *
201     * @see https://developers.facebook.com/docs/sharing/webmasters/crawler
202     * @see https://www.facebook.com/peering/
203     */
204    private const ROBOT_ASNS = [
205        'facebook' => ['AS32934', 'AS63293'],
206        'twitter'  => ['AS13414'],
207    ];
208
209    /**
210     * @param ServerRequestInterface  $request
211     * @param RequestHandlerInterface $handler
212     *
213     * @return ResponseInterface
214     */
215    public function process(ServerRequestInterface $request, RequestHandlerInterface $handler): ResponseInterface
216    {
217        $ua      = Validator::serverParams($request)->string('HTTP_USER_AGENT', '');
218        $ip      = Validator::attributes($request)->string('client-ip');
219        $address = IPFactory::parseAddressString($ip);
220        assert($address instanceof AddressInterface);
221
222        foreach (self::BAD_ROBOTS as $robot) {
223            if (str_contains($ua, $robot)) {
224                return $this->response();
225            }
226        }
227
228        foreach (self::ROBOT_REV_FWD_DNS as $robot => $valid_domains) {
229            if (str_contains($ua, $robot) && !$this->checkRobotDNS($ip, $valid_domains, false)) {
230                return $this->response();
231            }
232        }
233
234        foreach (self::ROBOT_REV_ONLY_DNS as $robot => $valid_domains) {
235            if (str_contains($ua, $robot) && !$this->checkRobotDNS($ip, $valid_domains, true)) {
236                return $this->response();
237            }
238        }
239
240        foreach (self::ROBOT_IPS as $robot => $valid_ip_ranges) {
241            if (str_contains($ua, $robot)) {
242                foreach ($valid_ip_ranges as $ip_range) {
243                    $range = IPFactory::parseRangeString($ip_range);
244
245                    if ($range instanceof RangeInterface && $range->contains($address)) {
246                        continue 2;
247                    }
248                }
249
250                return $this->response();
251            }
252        }
253
254        foreach (self::ROBOT_IP_FILES as $robot => $url) {
255            if (str_contains($ua, $robot)) {
256                $valid_ip_ranges = $this->fetchIpRangesForUrl($robot, $url);
257
258                foreach ($valid_ip_ranges as $ip_range) {
259                    $range = IPFactory::parseRangeString($ip_range);
260
261                    if ($range instanceof RangeInterface && $range->contains($address)) {
262                        continue 2;
263                    }
264                }
265
266                return $this->response();
267            }
268        }
269
270        foreach (self::ROBOT_ASNS as $robot => $asns) {
271            foreach ($asns as $asn) {
272                if (str_contains($ua, $robot)) {
273                    foreach ($this->fetchIpRangesForAsn($asn) as $range) {
274                        if ($range->contains($address)) {
275                            continue 2;
276                        }
277                    }
278
279                    return $this->response();
280                }
281            }
282        }
283
284        // Allow sites to block access from entire networks.
285        $block_asn = Validator::attributes($request)->string('block_asn', '');
286        preg_match_all('/(AS\d+)/', $block_asn, $matches);
287
288        foreach ($matches[1] as $asn) {
289            foreach ($this->fetchIpRangesForAsn($asn) as $range) {
290                if ($range->contains($address)) {
291                    return $this->response();
292                }
293            }
294        }
295
296        return $handler->handle($request);
297    }
298
299    /**
300     * Check that an IP address belongs to a robot operator using a forward/reverse DNS lookup.
301     *
302     * @param string        $ip
303     * @param array<string> $valid_domains
304     * @param bool          $reverse_only
305     *
306     * @return bool
307     */
308    private function checkRobotDNS(string $ip, array $valid_domains, bool $reverse_only): bool
309    {
310        $host = gethostbyaddr($ip);
311
312        if ($host === false) {
313            return false;
314        }
315
316        foreach ($valid_domains as $domain) {
317            if (str_ends_with($host, $domain)) {
318                return $reverse_only || $ip === gethostbyname($host);
319            }
320        }
321
322        return false;
323    }
324
325    /**
326     * Perform a whois search for an ASN.
327     *
328     * @param string $asn The autonomous system number to query
329     *
330     * @return array<RangeInterface>
331     */
332    private function fetchIpRangesForAsn(string $asn): array
333    {
334        return Registry::cache()->file()->remember('whois-asn-' . $asn, static function () use ($asn): array {
335            $mapper = static fn (AsnRouteInfo $route_info): ?RangeInterface => IPFactory::parseRangeString($route_info->route ?: $route_info->route6);
336
337            try {
338                $loader = new CurlLoader(self::WHOIS_TIMEOUT);
339                $whois  = new Whois($loader);
340                $info   = $whois->loadAsnInfo($asn);
341                $routes = $info->routes;
342                $ranges = array_map($mapper, $routes);
343
344                return array_filter($ranges);
345            } catch (Throwable) {
346                return [];
347            }
348        }, random_int(self::WHOIS_TTL_MIN, self::WHOIS_TTL_MAX));
349    }
350
351    /**
352     * Fetch a list of IP addresses from a remote file.
353     *
354     * @param string $ua
355     * @param string $url
356     *
357     * @return array<string>
358     */
359    private function fetchIpRangesForUrl(string $ua, string $url): array
360    {
361        return Registry::cache()->file()->remember('url-ip-list-' . $ua, static function () use ($url): array {
362            try {
363                $client   = new Client();
364                $response = $client->get($url, ['timeout' => 5]);
365                $contents = $response->getBody()->getContents();
366
367                preg_match_all(self::REGEX_IPV4, $contents, $matches);
368
369                return $matches[0];
370            } catch (GuzzleException) {
371                return [];
372            }
373        }, random_int(self::WHOIS_TTL_MIN, self::WHOIS_TTL_MAX));
374    }
375
376    /**
377     * @return ResponseInterface
378     */
379    private function response(): ResponseInterface
380    {
381        return response('Not acceptable', StatusCodeInterface::STATUS_NOT_ACCEPTABLE);
382    }
383}
384