friendica-addons/blockbot/blockbot.php

72 lines
2.4 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: 0.1
* Author: Philipp Holzer <admin@philipp.info>
*
*/
use Friendica\App;
use Friendica\Core\Hook;
use Friendica\Core\System;
use Jaybizzle\CrawlerDetect\CrawlerDetect;
2019-04-27 13:51:44 +02:00
use Friendica\Core\Logger;
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_uninstall() {
2019-04-22 10:49:40 +02:00
Hook::unregister('init_1', __FILE__, 'blockbot_init_1');
2019-04-20 14:15:45 +02:00
}
function blockbot_init_1(App $a) {
2019-04-20 14:15:45 +02:00
$crawlerDetect = new CrawlerDetect();
2019-05-03 12:25:13 +02:00
if (empty($_SERVER['HTTP_USER_AGENT'])) {
return;
}
$logdata = ['agent' => $_SERVER['HTTP_USER_AGENT'], 'uri' => $_SERVER['REQUEST_URI']];
if (!$crawlerDetect->isCrawler()) {
logger::debug('Good user agent detected', $logdata);
return;
}
// List of false positives' strings of known "good" agents.
$agents = ['fediverse.network crawler', 'Active_Pods_CheckBot_3.0', 'Social-Relay/',
2019-04-28 08:44:56 +02:00
'curl', 'zgrab', 'Go-http-client', 'curb', 'github.com', 'reqwest', 'Feedly/',
'Python-urllib/', 'Liferea/', 'aiohttp/', 'WordPress.com Reader', 'hackney/',
'Faraday v', 'okhttp', 'UniversalFeedParser', 'PixelFedBot', 'python-requests',
2019-05-03 12:25:13 +02:00
'WordPress/', 'http.rb/', 'Apache-HttpClient/', 'WordPress.com;'];
foreach ($agents as $agent) {
if (stristr($_SERVER['HTTP_USER_AGENT'], $agent)) {
logger::notice('False positive', $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.
$agents = ['SEMrushBot', 's~feedly-nikon3', 'Qwantify/Bleriot/', 'ltx71', 'Sogou web spider/',
'Diffbot/', 'Twitterbot/', 'YisouSpider/', 'evc-batch/', 'LivelapBot/', 'TrendsmapResolver/',
2019-04-28 08:44:56 +02:00
'PaperLiBot/', 'Nuzzel', 'um-LN/', 'Google Favicon', 'Datanyze', 'BLEXBot/', '360Spider',
2019-04-28 23:01:24 +02:00
'adscanner/', 'HeadlessChrome', 'wpif', 'startmebot/', 'Googlebot/', 'Applebot/',
2019-04-29 17:46:04 +02:00
'facebookexternalhit/', 'GoogleImageProxy', 'bingbot/', 'heritrix/', 'ldspider'];
foreach ($agents as $agent) {
if (stristr($_SERVER['HTTP_USER_AGENT'], $agent)) {
2019-05-03 12:25:13 +02:00
System::httpExit(403, ['title' => 'Bots are not allowed']);
2019-04-27 13:51:44 +02:00
}
2019-04-20 14:15:45 +02:00
}
logger::info('Blocked bot', $logdata);
2019-05-03 12:25:13 +02:00
System::httpExit(403, ['title' => 'Bots are not allowed']);
2019-04-20 14:15:45 +02:00
}