str_replace('/', '\/', $pattern), $excludePatterns);
// Crée une expression régulière unique pour matcher toutes les exclusions
$excludeRegex = '/' . implode('|', $excludePatterns) . '/i';
// Vérifier si le script est en mode dry run
$isDryRun = in_array('dry-run', $argv);
$maxResults = 100; // Limite par défaut pour la sortie
foreach ($argv as $arg) {
if (preg_match('/^max-results=(\d+)$/', $arg, $matches)) {
$maxResults = (int)$matches[1];
}
}
@mkdir(dirname($rssFile), 0777, true);
@mkdir(dirname($cacheFile), 0777, true);
$jsonData = @file_get_contents($jsonUrl);
if (!$jsonData) die("Error: Unable to fetch JSON from $jsonUrl\n");
$data = json_decode($jsonData, true);
if (!$data) die("Error: Invalid JSON data from $jsonUrl\n");
$includedIds = file_exists($cacheFile) ? file($cacheFile, FILE_IGNORE_NEW_LINES) : [];
// Filtrer les articles récents, non déjà inclus, et ne contenant pas de termes exclus
$newItems = [];
$excludedItems = [];
foreach ($data as $item) {
if (!isset($item['published'], $item['itemid'], $item['itemtitle'])) {
continue;
}
$published = $item['published'];
$itemId = $item['itemid'];
$itemTitle = $item['itemtitle'];
if (in_array($itemId, $includedIds) || $published < $cutoffDate || preg_match($excludeRegex, $itemTitle)) {
$excludedItems[] = $item;
} else {
$newItems[] = $item;
}
}
usort($newItems, fn($a, $b) => $b['published'] <=> $a['published']);
usort($excludedItems, fn($a, $b) => $b['published'] <=> $a['published']);
if ($isDryRun) {
// Mode dry-run : afficher les entrées incluses et exclues
echo "=== Dry Run Mode ===\n";
$showIncluded = array_slice($newItems, 0, $maxResults);
$showExcluded = array_slice($excludedItems, 0, $maxResults);
echo "Filtered IN (" . count($newItems) . " total, showing up to $maxResults):\n";
foreach ($showIncluded as $item) {
echo "- " . ($item['itemtitle'] ?? 'No title') . "\n";
}
echo "\nFiltered OUT (" . count($excludedItems) . " total, showing up to $maxResults):\n";
foreach ($showExcluded as $item) {
echo "- " . ($item['itemtitle'] ?? 'No title') . "\n";
}
exit(0);
}
// Groupement par jour pour les nouveaux éléments
$groupedByDay = [];
foreach ($newItems as $item) {
$day = isset($item['published']) && is_numeric($item['published'])
? date('Y-m-d', $item['published'])
: date('Y-m-d');
$groupedByDay[$day][] = $item;
}
$rss = new SimpleXMLElement('');
$channel = $rss->addChild('channel');
$channel->addChild('title', 'IndieBlog Feed (Filtered, Last 10 Days)');
$channel->addChild('link', 'https://indieblog.page/');
$channel->addChild('description', 'RSS feed of articles from the last 10 days, excluding unwanted topics');
$channel->addChild('language', 'en');
$newEntries = [];
foreach ($groupedByDay as $day => $items) {
$rssItem = $channel->addChild('item');
$rssItem->addChild('title', "Links for $day");
$rssItem->addChild('link', "https://indieblog.page/$day#" . md5(json_encode($items)));
$rssItem->addChild('guid', "https://indieblog.page/$day#" . md5(json_encode($items)));
$rssItem->addChild('pubDate', date(DATE_RSS, strtotime($day)));
$description = "";
foreach ($items as $item) {
$postTitle = htmlspecialchars($item['itemtitle'] ?? 'No title', ENT_XML1);
$postUrl = htmlspecialchars($item['itemurl'] ?? '#', ENT_XML1);
$blogTitle = htmlspecialchars($item['feedtitle'] ?? 'Unknown Blog', ENT_XML1);
$blogUrl = htmlspecialchars($item['feedurl'] ?? '#', ENT_XML1);
$description .= "- $postTitle ($blogTitle)
";
$newEntries[] = $item['itemid'];
}
$description .= "
";
addCData($rssItem->addChild('description'), $description);
}
if (!empty($newEntries)) {
file_put_contents($cacheFile, implode("\n", array_merge($includedIds, $newEntries)));
}
$rss->asXML($rssFile);
function addCData(SimpleXMLElement $node, $content)
{
$domNode = dom_import_simplexml($node);
$domOwner = $domNode->ownerDocument;
$domNode->appendChild($domOwner->createCDATASection($content));
}