*/
use Friendica\App;
+use Friendica\Core\Config;
use Friendica\Core\Hook;
use Friendica\Core\System;
use Jaybizzle\CrawlerDetect\CrawlerDetect;
}
function blockbot_init_1(App $a) {
- $crawlerDetect = new CrawlerDetect();
+ if (empty($_SERVER['HTTP_USER_AGENT'])) {
+ return;
+ }
$logdata = ['agent' => $_SERVER['HTTP_USER_AGENT'], 'uri' => $_SERVER['REQUEST_URI']];
- if (!$crawlerDetect->isCrawler()) {
- logger::debug('Good user agent detected', $logdata);
- return;
- }
+ // List of known crawlers.
+ $agents = ['SEMrushBot', 's~feedly-nikon3', 'Qwantify/Bleriot/', 'ltx71', 'Sogou web spider/',
+ 'Diffbot/', 'Twitterbot/', 'YisouSpider/', 'evc-batch/', 'LivelapBot/', 'TrendsmapResolver/',
+ 'PaperLiBot/', 'Nuzzel', 'um-LN/', 'Google Favicon', 'Datanyze', 'BLEXBot/', '360Spider',
+ 'adscanner/', 'HeadlessChrome', 'wpif', 'startmebot/', 'Googlebot/', 'Applebot/',
+ 'facebookexternalhit/', 'GoogleImageProxy', 'bingbot/', 'heritrix/', 'ldspider',
+ 'AwarioRssBot/', 'Zabbix', 'TweetmemeBot/', 'dcrawl/', 'PhantomJS/', 'Googlebot-Image/',
+ 'CrowdTanglebot/', 'Mediapartners-Google', 'Baiduspider/', 'datagnionbot'];
- // List of strings of reported false positives
- $agents = ['Mastodon', 'hackney', 'Faraday', 'okhttp', 'UniversalFeedParser', 'PixelFedBot', 'python-requests',
- 'WordPress', 'http.rb'];
foreach ($agents as $agent) {
if (stristr($_SERVER['HTTP_USER_AGENT'], $agent)) {
- // The agents had been reported to https://github.com/JayBizzle/Crawler-Detect/issues/
- logger::notice('Already reported wrong detection', $logdata);
- return;
+ System::httpExit(403, 'Bots are not allowed');
}
}
- // List of strings of known "good" agents
- $agents = ['diaspora-connection-tester', 'DiasporaFederation', 'Friendica', '(compatible; zot)',
- 'Micro.blog', 'GangGo', 'python/federation', 'GNU social', 'winHttp',
- 'Go-http-client', 'Mr.4x3 Powered', 'Test Certificate Info', 'WordPress.com', 'zgrab',
- 'curl/', 'StatusNet', 'OpenGraphReader/', 'Uptimebot/', 'python-opengraph-jaywink',
- 'fediverse.network crawler', 'Active_Pods_CheckBot_3.0', 'Social-Relay'];
+ // This switch here is only meant for developers who want to add more bots to the list above, it is not safe for production.
+ if (!Config::get('blockbot', 'training')) {
+ return;
+ }
- foreach ($agents as $agent) {
- if (stristr($_SERVER['HTTP_USER_AGENT'], $agent)) {
- // Report every false positive here: https://github.com/JayBizzle/Crawler-Detect/issues/
- // After report move it into the array above
- logger::notice('False positive', $logdata);
- return;
- }
+ $crawlerDetect = new CrawlerDetect();
+
+ if (!$crawlerDetect->isCrawler()) {
+ logger::debug('Good user agent detected', $logdata);
+ return;
}
- // List of known crawlers. They are added here to avoid having them logged at the end of the function.
- // This helps to detect false positives
- $agents = ['Mozilla/5.0 (compatible; SemrushBot/3~bl; +http://www.semrush.com/bot.html)', 'SEMrushBot',
- 'Mozilla/5.0 (Macintosh; Intel Mac OS X 10_13_4) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/65.0.3325.181 Safari/537.36 AppEngine-Google; (+http://code.google.com/appengine; appid: s~feedly-nikon3)'];
+ // List of false positives' strings of known "good" agents.
+ $agents = ['fediverse.network crawler', 'Active_Pods_CheckBot_3.0', 'Social-Relay/',
+ 'curl', 'zgrab', 'Go-http-client', 'curb', 'github.com', 'reqwest', 'Feedly/',
+ 'Python-urllib/', 'Liferea/', 'aiohttp/', 'WordPress.com Reader', 'hackney/',
+ 'Faraday v', 'okhttp', 'UniversalFeedParser', 'PixelFedBot', 'python-requests',
+ 'WordPress/', 'http.rb/', 'Apache-HttpClient/', 'WordPress.com;', 'Pleroma',
+ 'Dispatch/'];
foreach ($agents as $agent) {
- if ($_SERVER['HTTP_USER_AGENT'] == $agent) {
- System::httpExit(403, 'Bots are not allowed');
+ if (stristr($_SERVER['HTTP_USER_AGENT'], $agent)) {
+ logger::notice('False positive', $logdata);
+ return;
}
}