Merge pull request #849 from annando/false-positive

Blockbot: Avoid false positives
This commit is contained in:
Hypolite Petovan 2019-06-23 23:28:28 -04:00 committed by GitHub
commit 796ef5d507
No known key found for this signature in database
GPG key ID: 4AEE18F83AFDEB23
2 changed files with 61 additions and 3 deletions

View file

@ -8,9 +8,11 @@
*/
use Friendica\App;
use Friendica\Core\Config;
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';
@ -24,9 +26,64 @@ function blockbot_uninstall() {
}
function blockbot_init_1(App $a) {
if (empty($_SERVER['HTTP_USER_AGENT'])) {
return;
}
$logdata = ['agent' => $_SERVER['HTTP_USER_AGENT'], 'uri' => $_SERVER['REQUEST_URI']];
// List of known crawlers.
$agents = ['SemrushBot', 's~feedly-nikon3', 'Qwantify/Bleriot/', 'ltx71', 'Sogou web spider/',
'Diffbot/', 'Twitterbot/', 'YisouSpider', 'evc-batch/', 'LivelapBot/', 'TrendsmapResolver/',
'PaperLiBot/', 'Nuzzel', 'um-LN/', 'Google Favicon', 'Datanyze', 'BLEXBot/', '360Spider',
'adscanner/', 'HeadlessChrome', 'wpif', 'startmebot/', 'Googlebot/', 'Applebot/',
'facebookexternalhit/', 'GoogleImageProxy', 'bingbot/', 'heritrix/', 'ldspider',
'AwarioRssBot/', 'Zabbix', 'TweetmemeBot/', 'dcrawl/', 'PhantomJS/', 'Googlebot-Image/',
'CrowdTanglebot/', 'Mediapartners-Google', 'Baiduspider/', 'datagnionbot',
'MegaIndex.ru/', 'SMUrlExpander', 'Hatena-Favicon/', 'Wappalyzer', 'FlipboardProxy/',
'NetcraftSurveyAgent/', 'Dataprovider.com', 'SMTBot/', 'Nimbostratus-Bot/',
'DuckDuckGo-Favicons-Bot/', 'IndieWebCards/', 'proximic', 'netEstate NE Crawler',
'AhrefsBot/', 'YandexBot/', 'Exabot/', 'Mediumbot-MetaTagFetcher/', 'WhatsApp/',
'TelegramBot', 'SurdotlyBot/', 'BingPreview/', 'SabsimBot/', 'CCBot/', 'WbSrch/',
'DuckDuckBot-Https/', 'HTTP Banner Detection', 'YandexImages/', 'archive.org_bot',
'ArchiveTeam ArchiveBot/', 'yacybot', 'https://developers.google.com/+/web/snippet/',
'Scrapy/', 'github-camo', 'MJ12bot/', 'DotBot/', 'Pinterestbot/', 'Jooblebot/',
'Cliqzbot/', 'YaK/', 'Mediatoolkitbot'];
foreach ($agents as $agent) {
if (stristr($_SERVER['HTTP_USER_AGENT'], $agent)) {
System::httpExit(403, 'Bots are not allowed');
}
}
// 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;
}
$crawlerDetect = new CrawlerDetect();
if ($crawlerDetect->isCrawler()) {
System::httpExit(403, 'Bots are not allowed');
if (!$crawlerDetect->isCrawler()) {
logger::debug('Good user agent detected', $logdata);
return;
}
// List of false positives' strings of known "good" agents.
$agents = ['fediverse.network crawler', 'Active_Pods_CheckBot_3.0', 'Social-Relay/',
'curl', 'zgrab', 'Go-http-client', 'curb', 'github.com', 'reqwest', 'Feedly/',
'Python-urllib/', 'Liferea/', 'aiohttp/', 'WordPress.com Reader', 'hackney/',
'Faraday v', 'okhttp', 'UniversalFeedParser', 'PixelFedBot', 'python-requests',
'WordPress/', 'http.rb/', 'Apache-HttpClient/', 'WordPress.com;', 'Pleroma',
'Dispatch/', 'Ruby', 'Uptimebot/', 'Java/', 'libwww-perl/', 'Mastodon/',
'lua-resty-http/', 'Test Certificate Info'];
foreach ($agents as $agent) {
if (stristr($_SERVER['HTTP_USER_AGENT'], $agent)) {
logger::notice('False positive', $logdata);
return;
}
}
logger::info('Blocked bot', $logdata);
System::httpExit(403, 'Bots are not allowed');
}

View file

@ -575,7 +575,7 @@ function twitter_post_hook(App $a, array &$b)
return;
}
Logger::log('twitter post invoked');
Logger::notice('twitter post invoked', ['id' => $b['id'], 'guid' => $b['guid']]);
PConfig::load($b['uid'], 'twitter');
@ -610,6 +610,7 @@ function twitter_post_hook(App $a, array &$b)
$b['body'] = twitter_update_mentions($b['body']);
$msgarr = ItemContent::getPlaintextPost($b, $max_char, true, 8);
Logger::info('Got plaintext', $msgarr);
$msg = $msgarr["text"];
if (($msg == "") && isset($msgarr["title"])) {