friendica-addons/blockbot/blockbot.php

182 lines
8.7 KiB
PHP
Raw Normal View History

2019-04-20 14:15:45 +02:00
<?php
/**
* Name: blockbot
2019-04-20 14:15:45 +02:00
* Description: Blocking bots based on detecting bots/crawlers/spiders via the user agent and http_from header.
* Version: 0.2
2019-04-20 14:15:45 +02:00
* Author: Philipp Holzer <admin@philipp.info>
* Author: Michael Vogel <https://pirati.ca/profile/heluecht>
2019-04-20 14:15:45 +02:00
*
*/
use Friendica\Core\Hook;
2020-01-18 22:07:06 +01:00
use Friendica\DI;
2019-04-20 14:15:45 +02:00
use Jaybizzle\CrawlerDetect\CrawlerDetect;
2019-04-27 13:51:44 +02:00
use Friendica\Core\Logger;
use Friendica\Core\Renderer;
2021-11-04 21:32:16 +01:00
use Friendica\Network\HTTPException\ForbiddenException;
2019-04-20 14:15:45 +02:00
2019-04-21 12:35:33 +02:00
require_once __DIR__ . DIRECTORY_SEPARATOR . 'vendor' . DIRECTORY_SEPARATOR . 'autoload.php';
function blockbot_install()
{
2019-04-22 10:49:40 +02:00
Hook::register('init_1', __FILE__, 'blockbot_init_1');
2019-04-20 14:15:45 +02:00
}
function blockbot_addon_admin(string &$o)
{
$t = Renderer::getMarkupTemplate('admin.tpl', 'addon/blockbot/');
$o = Renderer::replaceMacros($t, [
'$submit' => DI::l10n()->t('Save Settings'),
'$good_crawlers' => ['good_crawlers', DI::l10n()->t('Allow "good" crawlers'), DI::config()->get('blockbot', 'good_crawlers'), DI::l10n()->t("Don't block fediverse crawlers, relay servers and other bots with good purposes.")],
'$socialmedia_agents' => ['socialmedia_agents', DI::l10n()->t('Allow preview agents'), DI::config()->get('blockbot', 'socialmedia_agents'), DI::l10n()->t("Don't block agents from social media systems that want to generate preview data for links that had been set by their users.")],
2024-03-05 05:54:36 +01:00
'$http_libraries' => ['http_libraries', DI::l10n()->t('Allow generic HTTP libraries'), DI::config()->get('blockbot', 'http_libraries'), DI::l10n()->t("Don't block agents from generic HTTP libraries that could be used for good or for bad and that currently can't be traced back to any known Fediverse project.")],
'$block_gab' => ['block_gab', DI::l10n()->t('Block GabSocial'), DI::config()->get('blockbot', 'block_gab'), DI::l10n()->t('Block the software GabSocial. This will block every access for that software. You can block dedicated gab instances in the blocklist settings in the admin section.')],
'$training' => ['training', DI::l10n()->t('Training mode'), DI::config()->get('blockbot', 'training'), DI::l10n()->t("Activates the training mode. This is only meant for developing purposes. Don't activate this on a production machine. This can cut communication with some systems.")],
]);
}
function blockbot_addon_admin_post()
{
DI::config()->set('blockbot', 'good_crawlers', $_POST['good_crawlers'] ?? false);
DI::config()->set('blockbot', 'socialmedia_agents', $_POST['socialmedia_agents'] ?? false);
2024-03-05 05:54:36 +01:00
DI::config()->set('blockbot', 'http_libraries', $_POST['http_libraries'] ?? false);
DI::config()->set('blockbot', 'block_gab', $_POST['block_gab'] ?? false);
DI::config()->set('blockbot', 'training', $_POST['training'] ?? false);
}
function blockbot_init_1()
{
2019-05-03 12:25:13 +02:00
if (empty($_SERVER['HTTP_USER_AGENT'])) {
return;
}
$logdata = ['agent' => $_SERVER['HTTP_USER_AGENT'], 'uri' => $_SERVER['REQUEST_URI']];
// List of known unwanted crawlers.
2023-03-05 15:01:32 +01:00
$agents = [
'SemrushBot', 's~feedly-nikon3', 'Qwantify/Bleriot/', 'ltx71', 'Sogou web spider/',
'Diffbot/', 'YisouSpider', 'evc-batch/', 'LivelapBot/', 'TrendsmapResolver/',
2019-05-29 20:51:07 +02:00
'PaperLiBot/', 'Nuzzel', 'um-LN/', 'Google Favicon', 'Datanyze', 'BLEXBot/', '360Spider',
'adscanner/', 'HeadlessChrome', 'wpif', 'startmebot/', 'Googlebot/', 'Applebot/',
'GoogleImageProxy', 'bingbot/', 'heritrix/', 'ldspider',
'AwarioRssBot/', 'TweetmemeBot/', 'dcrawl/', 'PhantomJS/', 'Googlebot-Image/',
2019-06-01 06:51:01 +02:00
'CrowdTanglebot/', 'Mediapartners-Google', 'Baiduspider/', 'datagnionbot',
2019-06-06 22:31:16 +02:00
'MegaIndex.ru/', 'SMUrlExpander', 'Hatena-Favicon/', 'Wappalyzer', 'FlipboardProxy/',
'NetcraftSurveyAgent/', 'Dataprovider.com', 'SMTBot/', 'Nimbostratus-Bot/',
2019-06-10 16:33:42 +02:00
'DuckDuckGo-Favicons-Bot/', 'IndieWebCards/', 'proximic', 'netEstate NE Crawler',
'AhrefsBot/', 'YandexBot/', 'Exabot/', 'Mediumbot-MetaTagFetcher/',
'SurdotlyBot/', 'BingPreview/', 'SabsimBot/', 'CCBot/', 'WbSrch/',
2019-06-20 07:31:53 +02:00
'DuckDuckBot-Https/', 'HTTP Banner Detection', 'YandexImages/', 'archive.org_bot',
'ArchiveTeam ArchiveBot/', 'yacybot', 'https://developers.google.com/+/web/snippet/',
'Scrapy/', 'MJ12bot/', 'DotBot/', 'Pinterestbot/', 'Jooblebot/',
2019-07-10 04:16:57 +02:00
'Cliqzbot/', 'YaK/', 'Mediatoolkitbot', 'Snacktory', 'FunWebProducts', 'oBot/',
2019-07-29 17:48:51 +02:00
'7Siters/', 'KOCMOHABT', 'Google-SearchByImage', 'FemtosearchBot/',
2020-07-05 16:36:17 +02:00
'HubSpot Crawler', 'DomainStatsBot/', 'Re-re Studio', 'AwarioSmartBot/',
2024-03-04 16:27:44 +01:00
'DNSResearchBot/', 'PetalBot;', 'Nmap Scripting Engine;',
2020-07-09 17:08:41 +02:00
'Google-Apps-Script; beanserver;', 'woorankreview/', 'Seekport Crawler;', 'AHC/',
2024-03-05 05:54:36 +01:00
'Semanticbot/', 'XoviOnpageCrawler;', 'Pinterest/',
'GetHPinfo.com-Bot/', 'BoardReader Favicon Fetcher', 'Google-Adwords-Instant', 'newspaper/',
2024-03-05 05:54:36 +01:00
'YurichevBot/', 'Crawling at Home Project', 'InfoTigerBot/', 'AdIdxBot/',
'MicrosoftPreview/', 'masscan/', 'Timpibot/', 'everyfeed-spider/', 'AndroidDownloadManager/',
'WebZIP/', 'WDG_Validator/', 'Screaming Frog SEO Spider/', ' Bytespider;', 'ISSCyberRiskCrawler/',
2024-03-10 07:14:01 +01:00
'BitSightBot/', 'ev-crawler/', 'CensysInspect/1.1', 'Protopage/', 'Gaisbot/', 'WellKnownBot/',
'SuperBot/', 'Googlebot-Mobile/', 'GPTBot/', 'GenomeCrawlerd/', '2ip bot/', 'Ocarinabot',
'Yahoo! Slurp;', 'AdsBot-Google', 'Gregarius/', 'FAST-WebCrawler/', 'Xenu Link Sleuth/',
'Ask Jeeves'
2024-03-05 05:54:36 +01:00
];
if (DI::config()->get('blockbot', 'block_gab')) {
$agents[] = 'GabSocial/';
}
// List of "good" crawlers, mostly from the fediverse.
$good_agents = [
'fediverse.space crawler', 'fediverse.network crawler', 'Active_Pods_CheckBot_3.0',
'Social-Relay/', 'Test Certificate Info', 'Uptimebot/', 'GNUSocialBot', 'UptimeRobot/',
'PTST/', 'Zabbix', 'Poduptime/', 'FediFetcher', 'lemmy-stats-crawler',
2024-03-10 07:14:01 +01:00
'FedditLemmyverseCrawler/', 'lemmy-explorer-crawler/', 'URIports Validator',
'rss-is-dead.lol web bot;', 'fedistatsCrawler/', 'W3C_CSS_Validator_JFouffa/',
'IABot/', 'Slackbot 1',
2023-03-05 15:01:32 +01:00
];
if (!DI::config()->get('blockbot', 'good_crawlers')) {
2019-07-28 10:13:53 +02:00
$agents = array_merge($agents, $good_agents);
2024-03-05 05:54:36 +01:00
} elseif (blockbot_match($good_agents)) {
return;
}
2019-07-28 10:13:53 +02:00
2024-03-05 05:54:36 +01:00
// List of agents from social media systems that fetch preview data via opem graph or twitter cards.
$socialmedia_agents = ['Twitterbot', 'facebookexternalhit/', 'SkypeUriPreview Preview/',
'TelegramBot', 'WhatsApp/', 'github-camo', 'Bluesky Cardyb/', 'XING-contenttabreceiver/',
2024-03-10 07:14:01 +01:00
'LinkedInBot/', 'Instagram ', 'Synapse (bot; ', 'Discordbot/', 'SummalyBot/',
'Slackbot-LinkExpanding', 'Slack-ImgProxy',
];
2024-03-05 05:54:36 +01:00
if (!DI::config()->get('blockbot', 'socialmedia_agents')) {
$agents = array_merge($agents, $socialmedia_agents);
2024-03-05 05:54:36 +01:00
} elseif (blockbot_match($socialmedia_agents)) {
return;
}
2024-03-05 05:54:36 +01:00
// HTTP Libraries
2024-03-06 06:27:08 +01:00
$http_libraries = ['ReactorNetty/', 'GuzzleHttp/', 'Embed PHP library', 'python-urllib3/',
2024-03-10 07:14:01 +01:00
'EventMachine HttpClient', 'HTMLParser/'
];
2024-03-05 05:54:36 +01:00
if (!DI::config()->get('blockbot', 'http_libraries')) {
$agents = array_merge($agents, $http_libraries);
} elseif (blockbot_match($http_libraries)) {
return;
}
2019-05-29 20:51:07 +02:00
2024-03-05 05:54:36 +01:00
if (blockbot_match($agents)) {
throw new ForbiddenException('Bots are not allowed. If you consider this a mistake, create an issue at https://github.com/friendica/friendica');
2019-05-29 20:51:07 +02:00
}
2019-05-30 06:45:20 +02:00
// 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 (!DI::config()->get('blockbot', 'training')) {
2019-05-29 20:51:07 +02:00
return;
}
$crawlerDetect = new CrawlerDetect();
if (!$crawlerDetect->isCrawler()) {
logger::debug('Good user agent detected', $logdata);
return;
}
2024-03-05 05:54:36 +01:00
// List of known "good" agents, mostly used by Fediverse systems, feed readers, ...
2023-03-05 15:01:32 +01:00
$agents = [
'curl', 'zgrab', 'Go-http-client', 'curb', 'github.com', 'reqwest', 'Feedly/',
'Python-urllib/', 'Liferea/', 'aiohttp/', 'WordPress.com Reader', 'hackney/',
'Faraday v', 'okhttp', 'UniversalFeedParser', 'PixelFedBot', 'python-requests',
2019-05-30 12:32:01 +02:00
'WordPress/', 'http.rb/', 'Apache-HttpClient/', 'WordPress.com;', 'Pleroma',
'Dispatch/', 'Ruby', 'Java/', 'libwww-perl/', 'Mastodon/', 'FeedlyApp/',
2020-07-08 21:24:00 +02:00
'lua-resty-http/', 'Tiny Tiny RSS/', 'Wget/', 'PostmanRuntime/',
'W3C_Validator/', 'NetNewsWire', 'FeedValidator/', 'theoldreader.com', 'axios/',
'Paw/', 'PeerTube/', 'fedi.inex.dev', 'FediDB/', 'index.community crawler',
2024-03-05 05:54:36 +01:00
'Slackbot-LinkExpanding', 'Firefish/', 'Takahe/', 'Akkoma ', 'Misskey/', 'Lynx/',
2024-03-10 07:14:01 +01:00
'camo-rs asset proxy', 'gotosocial/', 'incestoma ', 'SpaceCowboys Android RSS Reader',
'NewsBlur Feed Finder', 'Lemmy/', 'enby-town/', 'rss2tg bot;', '; HTTrack ',
'MbinBot', 'kbinBot'
2023-03-05 15:01:32 +01:00
];
2019-07-28 10:13:53 +02:00
2024-03-05 05:54:36 +01:00
if (blockbot_match($agents)) {
logger::info('False positive', $logdata);
return;
2019-07-28 10:13:53 +02:00
}
2024-03-05 05:54:36 +01:00
logger::notice('Blocked bot', $logdata);
throw new ForbiddenException('Bots are not allowed. If you consider this a mistake, create an issue at https://github.com/friendica/friendica');
}
function blockbot_match(array $agents)
{
foreach ($agents as $agent) {
if (stristr($_SERVER['HTTP_USER_AGENT'], $agent)) {
2024-03-05 05:54:36 +01:00
return true;
}
}
2024-03-05 05:54:36 +01:00
return false;
}