vikarti.anatra/main
1
0
Fork 0
mirror of https://github.com/manualdousuario/marreta.git synced 2026-04-30 20:40:13 +00:00
Code Issues Projects Releases Packages Wiki Activity Actions

removido logs e liberado selenium como ultimo fallback

Browse source
This commit is contained in:
Renan Bernordi 2024-12-24 13:35:20 -03:00
parent 775b6bdd3f
commit c9e59dfce2
3 changed files with 17 additions and 3 deletions
Download patch file Download diff file Expand all files Collapse all files

17
app/inc/URLAnalyzer.php
View file

@ -123,7 +123,6 @@ class URLAnalyzer
$host = preg_replace('/^www\./', '', $host);
if (in_array($host, BLOCKED_DOMAINS)) {
Logger::getInstance()->log($cleanUrl, 'BLOCKED_DOMAIN');
throw new Exception('Este domínio está bloqueado para extração.');
}
@ -153,7 +152,7 @@ class URLAnalyzer
return $processedContent;
}
} catch (Exception $e) {
Logger::getInstance()->log($cleanUrl, 'DIRECT_FETCH_ERROR', $e->getMessage());
error_log("DIRECT_FETCH_ERROR: " . $e->getMessage());
}
// 6. Tenta buscar do Wayback Machine como fallback
@ -165,7 +164,19 @@ class URLAnalyzer
return $processedContent;
}
} catch (Exception $e) {
Logger::getInstance()->log($cleanUrl, 'WAYBACK_FETCH_ERROR', $e->getMessage());
error_log("WAYBACK_FETCH_ERROR: " . $e->getMessage());
}
// 7. Tenta buscar com Selenium como fallback
try {
$content = $this->fetchFromSelenium($cleanUrl, 'firefox');
if (!empty($content)) {
$processedContent = $this->processContent($content, $host, $cleanUrl);
$this->cache->set($cleanUrl, $processedContent);
return $processedContent;
}
} catch (Exception $e) {
error_log("SELENIUM_ERROR: " . $e->getMessage());
}
Logger::getInstance()->log($cleanUrl, 'GENERAL_FETCH_ERROR');