use Friendica\Core\Hook;
use Friendica\Core\System;
use Jaybizzle\CrawlerDetect\CrawlerDetect;
+use Friendica\Core\Logger;
require_once __DIR__ . DIRECTORY_SEPARATOR . 'vendor' . DIRECTORY_SEPARATOR . 'autoload.php';
function blockbot_init_1(App $a) {
$crawlerDetect = new CrawlerDetect();
+ // List of strings of known "good" agents
+ $agents = ['diaspora-connection-tester', 'DiasporaFederation', 'Friendica', '(compatible; zot)',
+ 'Micro.blog', 'Mastodon', 'hackney', 'GangGo', 'python/federation', 'GNU social', 'winHttp',
+ 'Go-http-client', 'Mr.4x3 Powered', 'Test Certificate Info', 'WordPress.com', 'zgrab',
+ 'curl/', 'StatusNet', 'OpenGraphReader/', 'Uptimebot/', 'python-opengraph-jaywink'];
+
if ($crawlerDetect->isCrawler()) {
+ foreach ($agents as $agent) {
+ if (stristr($_SERVER['HTTP_USER_AGENT'], $agent)) {
+ // @ToDo: Report every false positive here: https://github.com/JayBizzle/Crawler-Detect/issues/326
+ logger::notice('False positive', ['agent' => $_SERVER['HTTP_USER_AGENT']]);
+ return;
+ }
+ }
+ logger::info('Blocked bot', ['agent' => $_SERVER['HTTP_USER_AGENT']]);
System::httpExit(403, 'Bots are not allowed');
+ } else {
+ logger::debug('Good user agent detected', ['agent' => $_SERVER['HTTP_USER_AGENT']]);
}
}