xref: /webtrees/app/Http/Middleware/BadBotBlocker.php (revision 07661f2d8fb7221a3ac37df6ddeb71ac39477349)
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        'CCBot', // Used to train a number of LLMs
75        'ChatGPT-User', // Used by ChatGPT during operation
76        'DataForSeoBot', // https://dataforseo.com/dataforseo-bot
77        'DotBot',
78        'FacebookBot', // Collects training data for Facebook's LLM translator.
79        'Google-Extended', // Collects training data for Google Bard
80        'GPTBot', // Collects training data for ChatGPT
81        'Grapeshot',
82        'Honolulu-bot', // Aggressive crawer, no info available
83        'ia_archiver',
84        'linabot', // Aggressive crawer, no info available
85        'Linguee',
86        'MegaIndex.ru',
87        'MJ12bot',
88        'netEstate NE',
89        'Omgilibot', // Collects training data for LLMs
90        'panscient',
91        'PetalBot',
92        'proximic',
93        'SeekportBot', // Pretends to be a search engine - but isn't
94        'SemrushBot',
95        'serpstatbot',
96        'SEOkicks',
97        'SiteKiosk',
98        'TinyTestBot',
99        'Turnitin',
100        'wp_is_mobile', // Nothing to do with wordpress
101        'XoviBot',
102        'YisouSpider',
103        'ZoominfoBot',
104    ];
105
106    /**
107     * Some search engines use reverse/forward DNS to verify the IP address.
108     *
109     * @see https://developer.amazon.com/support/amazonbot
110     * @see https://support.google.com/webmasters/answer/80553?hl=en
111     * @see https://www.bing.com/webmaster/help/which-crawlers-does-bing-use-8c184ec0
112     * @see https://www.bing.com/webmaster/help/how-to-verify-bingbot-3905dc26
113     * @see https://yandex.com/support/webmaster/robot-workings/check-yandex-robots.html
114     * @see https://www.mojeek.com/bot.html
115     * @see https://support.apple.com/en-gb/HT204683
116     */
117    private const ROBOT_REV_FWD_DNS = [
118        'Amazonbot'        => ['.crawl.amazon.com'],
119        'Applebot'         => ['.applebot.apple.com'],
120        'BingPreview'      => ['.search.msn.com'],
121        'Google'           => ['.google.com', '.googlebot.com'],
122        'Mail.RU_Bot'      => ['.mail.ru'],
123        'MicrosoftPreview' => ['.search.msn.com'],
124        'MojeekBot'        => ['.mojeek.com'],
125        'Qwantify'         => ['.search.qwant.com'],
126        'Sogou'            => ['.crawl.sogou.com'],
127        'Yahoo'            => ['.crawl.yahoo.net'],
128        'Yandex'           => ['.yandex.ru', '.yandex.net', '.yandex.com'],
129        'bingbot'          => ['.search.msn.com'],
130        'msnbot'           => ['.search.msn.com'],
131    ];
132
133    /**
134     * Some search engines only use reverse DNS to verify the IP address.
135     *
136     * @see https://help.baidu.com/question?prod_id=99&class=0&id=3001
137     * @see https://napoveda.seznam.cz/en/full-text-search/seznambot-crawler
138     * @see https://www.ionos.de/terms-gtc/faq-crawler
139     */
140    private const ROBOT_REV_ONLY_DNS = [
141        'Baiduspider' => ['.baidu.com', '.baidu.jp'],
142        'FreshBot'    => ['.seznam.cz'],
143        'IonCrawl'    => ['.1und1.org'],
144        'Neevabot'    => ['.neeva.com'],
145        'SeznamBot'   => ['.seznam.cz'],
146    ];
147
148    /**
149     * Some search engines operate from designated IP addresses.
150     *
151     * @see https://www.apple.com/go/applebot
152     * @see https://help.duckduckgo.com/duckduckgo-help-pages/results/duckduckbot
153     */
154    private const ROBOT_IPS = [
155        'AppleBot'    => [
156            '17.0.0.0/8',
157        ],
158        'Ask Jeeves'  => [
159            '65.214.45.143',
160            '65.214.45.148',
161            '66.235.124.192',
162            '66.235.124.7',
163            '66.235.124.101',
164            '66.235.124.193',
165            '66.235.124.73',
166            '66.235.124.196',
167            '66.235.124.74',
168            '63.123.238.8',
169            '202.143.148.61',
170        ],
171        'DuckDuckBot' => [
172            '23.21.227.69',
173            '50.16.241.113',
174            '50.16.241.114',
175            '50.16.241.117',
176            '50.16.247.234',
177            '52.204.97.54',
178            '52.5.190.19',
179            '54.197.234.188',
180            '54.208.100.253',
181            '54.208.102.37',
182            '107.21.1.8',
183        ],
184    ];
185
186    /**
187     * Some search engines operate from designated IP addresses.
188     *
189     * @see https://bot.seekport.com/
190     */
191    private const ROBOT_IP_FILES = [
192        'SeekportBot' => 'https://bot.seekport.com/seekportbot_ips.txt',
193    ];
194
195    /**
196     * Some search engines operate from within a designated autonomous system.
197     *
198     * @see https://developers.facebook.com/docs/sharing/webmasters/crawler
199     * @see https://www.facebook.com/peering/
200     */
201    private const ROBOT_ASNS = [
202        'facebook' => ['AS32934', 'AS63293'],
203        'twitter'  => ['AS13414'],
204    ];
205
206    /**
207     * @param ServerRequestInterface  $request
208     * @param RequestHandlerInterface $handler
209     *
210     * @return ResponseInterface
211     */
212    public function process(ServerRequestInterface $request, RequestHandlerInterface $handler): ResponseInterface
213    {
214        $ua      = Validator::serverParams($request)->string('HTTP_USER_AGENT', '');
215        $ip      = Validator::attributes($request)->string('client-ip');
216        $address = IPFactory::parseAddressString($ip);
217        assert($address instanceof AddressInterface);
218
219        foreach (self::BAD_ROBOTS as $robot) {
220            if (str_contains($ua, $robot)) {
221                return $this->response();
222            }
223        }
224
225        foreach (self::ROBOT_REV_FWD_DNS as $robot => $valid_domains) {
226            if (str_contains($ua, $robot) && !$this->checkRobotDNS($ip, $valid_domains, false)) {
227                return $this->response();
228            }
229        }
230
231        foreach (self::ROBOT_REV_ONLY_DNS as $robot => $valid_domains) {
232            if (str_contains($ua, $robot) && !$this->checkRobotDNS($ip, $valid_domains, true)) {
233                return $this->response();
234            }
235        }
236
237        foreach (self::ROBOT_IPS as $robot => $valid_ip_ranges) {
238            if (str_contains($ua, $robot)) {
239                foreach ($valid_ip_ranges as $ip_range) {
240                    $range = IPFactory::parseRangeString($ip_range);
241
242                    if ($range instanceof RangeInterface && $range->contains($address)) {
243                        continue 2;
244                    }
245                }
246
247                return $this->response();
248            }
249        }
250
251        foreach (self::ROBOT_IP_FILES as $robot => $url) {
252            if (str_contains($ua, $robot)) {
253                $valid_ip_ranges = $this->fetchIpRangesForUrl($robot, $url);
254
255                foreach ($valid_ip_ranges as $ip_range) {
256                    $range = IPFactory::parseRangeString($ip_range);
257
258                    if ($range instanceof RangeInterface && $range->contains($address)) {
259                        continue 2;
260                    }
261                }
262
263                return $this->response();
264            }
265        }
266
267        foreach (self::ROBOT_ASNS as $robot => $asns) {
268            foreach ($asns as $asn) {
269                if (str_contains($ua, $robot)) {
270                    foreach ($this->fetchIpRangesForAsn($asn) as $range) {
271                        if ($range->contains($address)) {
272                            continue 2;
273                        }
274                    }
275
276                    return $this->response();
277                }
278            }
279        }
280
281        // Allow sites to block access from entire networks.
282        $block_asn = Validator::attributes($request)->string('block_asn', '');
283        preg_match_all('/(AS\d+)/', $block_asn, $matches);
284
285        foreach ($matches[1] as $asn) {
286            foreach ($this->fetchIpRangesForAsn($asn) as $range) {
287                if ($range->contains($address)) {
288                    return $this->response();
289                }
290            }
291        }
292
293        return $handler->handle($request);
294    }
295
296    /**
297     * Check that an IP address belongs to a robot operator using a forward/reverse DNS lookup.
298     *
299     * @param string        $ip
300     * @param array<string> $valid_domains
301     * @param bool          $reverse_only
302     *
303     * @return bool
304     */
305    private function checkRobotDNS(string $ip, array $valid_domains, bool $reverse_only): bool
306    {
307        $host = gethostbyaddr($ip);
308
309        if ($host === false) {
310            return false;
311        }
312
313        foreach ($valid_domains as $domain) {
314            if (str_ends_with($host, $domain)) {
315                return $reverse_only || $ip === gethostbyname($host);
316            }
317        }
318
319        return false;
320    }
321
322    /**
323     * Perform a whois search for an ASN.
324     *
325     * @param string $asn The autonomous system number to query
326     *
327     * @return array<RangeInterface>
328     */
329    private function fetchIpRangesForAsn(string $asn): array
330    {
331        return Registry::cache()->file()->remember('whois-asn-' . $asn, static function () use ($asn): array {
332            $mapper = static fn (AsnRouteInfo $route_info): ?RangeInterface => IPFactory::parseRangeString($route_info->route ?: $route_info->route6);
333
334            try {
335                $loader = new CurlLoader(self::WHOIS_TIMEOUT);
336                $whois  = new Whois($loader);
337                $info   = $whois->loadAsnInfo($asn);
338                $routes = $info->routes;
339                $ranges = array_map($mapper, $routes);
340
341                return array_filter($ranges);
342            } catch (Throwable) {
343                return [];
344            }
345        }, random_int(self::WHOIS_TTL_MIN, self::WHOIS_TTL_MAX));
346    }
347
348    /**
349     * Fetch a list of IP addresses from a remote file.
350     *
351     * @param string $ua
352     * @param string $url
353     *
354     * @return array<string>
355     */
356    private function fetchIpRangesForUrl(string $ua, string $url): array
357    {
358        return Registry::cache()->file()->remember('url-ip-list-' . $ua, static function () use ($url): array {
359            try {
360                $client   = new Client();
361                $response = $client->get($url, ['timeout' => 5]);
362                $contents = $response->getBody()->getContents();
363
364                preg_match_all(self::REGEX_IPV4, $contents, $matches);
365
366                return $matches[0];
367            } catch (GuzzleException) {
368                return [];
369            }
370        }, random_int(self::WHOIS_TTL_MIN, self::WHOIS_TTL_MAX));
371    }
372
373    /**
374     * @return ResponseInterface
375     */
376    private function response(): ResponseInterface
377    {
378        return response('Not acceptable', StatusCodeInterface::STATUS_NOT_ACCEPTABLE);
379    }
380}
381