change(indieblog): enable refreshing feed with a certain date
This commit is contained in:
parent
f5402ded20
commit
ba8e8f5fb8
@ -1,12 +1,26 @@
|
||||
<?php
|
||||
// php /app/data/scripts/generate_indieblog_daily_rss.php dry-run max-results=150
|
||||
// php /app/data/scripts/generate_indieblog_daily_rss.php dry-run max-results=150 date=YYYY-MM-DD
|
||||
// Configuration
|
||||
$jsonUrl = "https://indieblog.page/export";
|
||||
$rssFile = "/app/data/public/indieblogv2_feed.xml";
|
||||
$cacheFile = "/app/data/cache/indieblogv2_cache.txt";
|
||||
$cutoffDate = strtotime('-10 days'); // Seulement les 10 derniers jours
|
||||
$cutoffDate = strtotime('-10 days'); // Only last 10 days
|
||||
|
||||
// Liste des mots/expressions à exclure (insensibilité à la casse activée avec `(?i)`)
|
||||
// Allow target date override via CLI: date=YYYY-MM-DD
|
||||
$targetDate = null;
|
||||
if (PHP_SAPI === 'cli') {
|
||||
foreach ($argv as $arg) {
|
||||
if (preg_match('/^date=(\d{4}-\d{2}-\d{2})$/', $arg, $matches)) {
|
||||
$targetDate = $matches[1];
|
||||
break;
|
||||
}
|
||||
}
|
||||
}
|
||||
if (!$targetDate) {
|
||||
$targetDate = date('Y-m-d');
|
||||
}
|
||||
|
||||
// List of exclude patterns (case insensitive)
|
||||
$excludePatterns = [
|
||||
'(?i)(azure|powershell|rimworld|Olympique|Olympic|Paris 2024|sportif|windows|twitter|freebsd|mastodon|kirby|flask|wp engine|cobol)',
|
||||
'(?i)(metalcore|deathcore|death metal|black metal|death punk|deathpunk|djent|blackmetal|blackened|bitcoin)',
|
||||
@ -20,42 +34,32 @@ $excludePatterns = [
|
||||
'(?i)(Review: (?:[1-3]\.\d|[1-2]\.\d|3\.0)/5\.0|licenciement|misère|du jamais vu|il faut|Palestiniens|israél|palestine|vaccin|mpox|épidémie|violeur|rapist|pédo|criminel|poutine|le RN)',
|
||||
'(?i)(réseau social x|gendarme|policier|kurde|condamnation|perquisition|incendi|gabriel attal|attaqué|justice|tués par|tué par|snapdragon|qualcomm|porn|barnier|occultes|désinformation|immigration|racism|spacex|islam|condamné|de viol|pour viol)',
|
||||
|
||||
// Années au début, à la fin, ou précédées de "at/in/of", ou au milieu des titres spécifiques
|
||||
// Year-related filters
|
||||
'(?i)(\b\d{4}\b|(?:\b(?:at|in|of)\s\d{4}\b)|(?:\b\d{4}\b\.$|\(\d{4}\)))',
|
||||
'(?i)(^202\d\b|\b202\d$)', // Titre commençant ou terminant par 202x
|
||||
'(?i)(\bin\s202\d|\bat\s202\d)', // "in/at 202x"
|
||||
'(?i)(my\s.*\b202\d\b)', // Titres contenant "My ... 202x"
|
||||
// Exclusion des titres contenant des caractères non-latins (Unicode étendu)
|
||||
'[^\x00-\x7F]', // Tous les caractères non-ASCII (inclut Unicode comme chinois, cyrillique, etc.)
|
||||
'[\p{Cyrillic}]', // Caractères cyrilliques (russe, ukrainien, etc.)
|
||||
'[\p{Greek}]', // Alphabet grec
|
||||
'[\p{Han}\p{Hiragana}\p{Katakana}]', // Chinois, japonais, coréen
|
||||
'[äöüßÄÖÜẞ]', // Allemand (umlauts et eszett)
|
||||
|
||||
// Titres spécifiques
|
||||
'(?i)(വേമ്പനാട്ട്)', // Malayâlam
|
||||
|
||||
// Week notes et expressions relatives au nouvel an
|
||||
'(?i)(^202\d\b|\b202\d$)',
|
||||
'(?i)(\bin\s202\d|\bat\s202\d)',
|
||||
'(?i)(my\s.*\b202\d\b)',
|
||||
// Exclude non-ASCII characters, Cyrillic, Greek, Han, etc.
|
||||
'[^\x00-\x7F]',
|
||||
'[\p{Cyrillic}]',
|
||||
'[\p{Greek}]',
|
||||
'[\p{Han}\p{Hiragana}\p{Katakana}]',
|
||||
'[äöüßÄÖÜẞ]',
|
||||
'(?i)(വേമ്പനാട്ട്)',
|
||||
'(?i)(the week|week notes|weekly|weeknote|digest)',
|
||||
'(?i)(new year|yearly)',
|
||||
|
||||
// Mentions de jours ou mois (en anglais), ou relatifs au temps
|
||||
'(?i)(monday|tuesday|wednesday|thursday|friday|saturday|sunday)',
|
||||
'(?i)(january|february|march|april|may|june|july|august|september|october|november|december)',
|
||||
'(?i)(today|yesterday|tomorrow|this year|this week)',
|
||||
|
||||
// "Merry Christmas"
|
||||
'(?i)(merry christmas)'
|
||||
];
|
||||
// Corrige les barres obliques pour qu'elles ne cassent pas `preg_match`
|
||||
// Fix slashes for regex
|
||||
$excludePatterns = array_map(fn($pattern) => 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
|
||||
// Check for dry-run mode
|
||||
$isDryRun = in_array('dry-run', $argv);
|
||||
$maxResults = 100; // Limite par défaut pour la sortie
|
||||
$maxResults = 100; // Default limit
|
||||
foreach ($argv as $arg) {
|
||||
if (preg_match('/^max-results=(\d+)$/', $arg, $matches)) {
|
||||
$maxResults = (int)$matches[1];
|
||||
@ -73,7 +77,7 @@ 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
|
||||
// Filter recent articles, not already included, and not matching exclude regex
|
||||
$newItems = [];
|
||||
$excludedItems = [];
|
||||
|
||||
@ -81,11 +85,9 @@ foreach ($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[] = $item;
|
||||
} else {
|
||||
@ -96,18 +98,15 @@ foreach ($data as $item) {
|
||||
usort($newItems, fn($a, $b) => $b['published'] <=> $a['published']);
|
||||
usort($excludedItems, fn($a, $b) => $b['published'] <=> $a['published']);
|
||||
|
||||
// Dry run: display filtered items
|
||||
if ($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";
|
||||
@ -115,12 +114,15 @@ if ($isDryRun) {
|
||||
exit(0);
|
||||
}
|
||||
|
||||
// Groupement par jour pour les nouveaux éléments
|
||||
// Group items only for the target date
|
||||
$groupedByDay = [];
|
||||
foreach ($newItems as $item) {
|
||||
$day = isset($item['published']) && is_numeric($item['published'])
|
||||
? date('Y-m-d', $item['published'])
|
||||
: date('Y-m-d');
|
||||
if ($day !== $targetDate) {
|
||||
continue;
|
||||
}
|
||||
$groupedByDay[$day][] = $item;
|
||||
}
|
||||
|
||||
@ -141,19 +143,16 @@ foreach ($groupedByDay as $day => $items) {
|
||||
$rssItem->addChild('link', "https://indieblog.page/$day#" . md5(json_encode($items)));
|
||||
$rssItem->addChild('guid', "https://indieblog.page/$day#" . md5(json_encode($items)));
|
||||
$rssItem->addChild('pubDate', date(DATE_RSS, strtotime($day)));
|
||||
|
||||
$description = "<ul>";
|
||||
foreach ($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);
|
||||
|
||||
$description .= "<li><a href=\"$postUrl\">$postTitle</a> (<a href=\"$blogUrl\">$blogTitle</a>)</li>";
|
||||
$newEntries[] = $item['itemid'];
|
||||
}
|
||||
$description .= "</ul>";
|
||||
|
||||
addCData($rssItem->addChild('description'), $description);
|
||||
}
|
||||
|
||||
@ -169,3 +168,4 @@ function addCData(SimpleXMLElement $node, $content)
|
||||
$domOwner = $domNode->ownerDocument;
|
||||
$domNode->appendChild($domOwner->createCDATASection($content));
|
||||
}
|
||||
?>
|
||||
|
||||
Loading…
x
Reference in New Issue
Block a user