SEO: canonical, sitemap.xml, robots.txt, JSON-LD, noindex admin

public/.htaccess

@@ -16,6 +16,9 @@ RewriteRule ^feed/?$    /feed.php [L,QSA]
 RewriteRule ^rss/?$     /feed.php [L,QSA]
 RewriteRule ^rss\.xml$  /feed.php [L,QSA]
 
+# Sitemap
+RewriteRule ^sitemap\.xml$ /sitemap.php [L]
+
 # Ajoute .php si le fichier correspondant existe
 RewriteCond %{DOCUMENT_ROOT}%{REQUEST_URI}.php -f
 RewriteRule ^(.+?)/?$ /$1.php [L,QSA]

public/index.php

@@ -21,6 +21,10 @@ $action = $_GET['action'] ?? 'list';
 $uuid   = $_GET['uuid']   ?? '';
 $slug   = $_GET['slug']   ?? '';
 
+$_noindexActions = ['create', 'edit', 'admin', 'categories', 'diff', 'add_files', 'import_image', 'import_image_step2', 'sources', 'profile', 'delete_file', 'delete_external_link', 'rename_category', 'delete_category', 'toggle_private_category'];
+$metaRobots = in_array($action, $_noindexActions, true) ? 'noindex, nofollow' : null;
+unset($_noindexActions);
+
 // ─── Extraction de métadonnées depuis une URL ────────────────────────────────
 function fetchUrlMeta(string $url): array
 {

public/robots.txt

@@ -0,0 +1,16 @@
+User-agent: *
+Disallow: /?action=edit
+Disallow: /?action=create
+Disallow: /?action=admin
+Disallow: /?action=delete
+Disallow: /?action=diff
+Disallow: /?action=categories
+Disallow: /?action=add_files
+Disallow: /?action=import_image
+Disallow: /?action=sources
+Disallow: /?action=profile
+Disallow: /login
+Disallow: /logout.php
+Disallow: /oidc/
+
+Sitemap: https://varlog.a5l.fr/sitemap.xml

public/sitemap.php

@@ -0,0 +1,49 @@
+<?php
+
+declare(strict_types=1);
+
+define('BASE_PATH', realpath(__DIR__ . '/../'));
+
+require_once BASE_PATH . '/src/helpers.php';
+require_once BASE_PATH . '/config/config.php';
+require_once BASE_PATH . '/src/ArticleManager.php';
+
+$articles = new ArticleManager(BASE_PATH . '/data');
+$privateCats = $articles->getPrivateCategories();
+
+$published = array_filter($articles->getAll(true), static function (array $a) use ($privateCats): bool {
+    $cat = trim($a['category'] ?? '');
+    if ($cat !== '' && in_array($cat, $privateCats, true)) {
+        return false;
+    }
+    if (strtotime((string)($a['published_at'] ?? '')) > time()) {
+        return false;
+    }
+    return true;
+});
+
+header('Content-Type: application/xml; charset=UTF-8');
+header('X-Robots-Tag: noindex');
+
+echo '<?xml version="1.0" encoding="UTF-8"?>' . "\n";
+echo '<urlset xmlns="http://www.sitemaps.org/schemas/sitemap/0.9">' . "\n";
+
+// Homepage
+echo '  <url>' . "\n";
+echo '    <loc>' . htmlspecialchars(rtrim(APP_URL, '/') . '/') . '</loc>' . "\n";
+echo '    <changefreq>daily</changefreq>' . "\n";
+echo '    <priority>1.0</priority>' . "\n";
+echo '  </url>' . "\n";
+
+foreach ($published as $article) {
+    $loc     = htmlspecialchars(rtrim(APP_URL, '/') . '/post/' . rawurlencode($article['slug'] ?? ''));
+    $lastmod = date('Y-m-d', strtotime((string)($article['updated_at'] ?? $article['published_at'] ?? 'now')));
+    echo '  <url>' . "\n";
+    echo '    <loc>' . $loc . '</loc>' . "\n";
+    echo '    <lastmod>' . $lastmod . '</lastmod>' . "\n";
+    echo '    <changefreq>monthly</changefreq>' . "\n";
+    echo '    <priority>0.8</priority>' . "\n";
+    echo '  </url>' . "\n";
+}
+
+echo '</urlset>' . "\n";

templates/layout.php

@@ -7,7 +7,8 @@
 
     <!-- SEO -->
     <meta name="description" content="<?= htmlspecialchars(($seoDescription ?? '') ?: 'Varlog est un journal personnel en ligne de Cédrix. Informatique, hack et loisirs techniques.') ?>">
-    <meta name="robots" content="index, follow">
+    <meta name="robots" content="<?= htmlspecialchars($metaRobots ?? 'index, follow') ?>">
+    <link rel="canonical" href="<?= htmlspecialchars($canonical ?? $ogUrl ?? rtrim(APP_URL, '/') . '/') ?>">
 
     <!-- Open Graph -->
     <meta property="og:title" content="<?= htmlspecialchars(($seoTitle ?? '') ?: ($title ?? 'varlog')) ?>">
@@ -25,6 +26,10 @@
     <meta property="article:published_time" content="<?= htmlspecialchars(date('c', strtotime((string)$articlePublishedAt))) ?>">
     <?php endif; ?>
 
+    <?php if (!empty($jsonLd ?? '')): ?>
+    <script type="application/ld+json"><?= $jsonLd ?></script>
+    <?php endif; ?>
+
     <!-- RSS autodiscovery -->
     <link rel="alternate" type="application/rss+xml" title="varlog" href="/feed">
 

templates/post_list.php

@@ -90,4 +90,28 @@ ob_start();
 <?php
 $content = ob_get_clean();
 $title   = 'varlog';
+
+// Pages avec curseur = contenu non-canonique → noindex
+if (!empty($cursor)) {
+    $metaRobots = 'noindex, follow';
+    $canonical  = rtrim(APP_URL, '/') . '/';
+} elseif ($filterCat !== '') {
+    $canonical = rtrim(APP_URL, '/') . '/?' . http_build_query(['cat' => $filterCat]);
+} else {
+    $canonical = rtrim(APP_URL, '/') . '/';
+}
+
+// JSON-LD WebSite sur la page d'accueil sans filtre
+if (empty($cursor) && $filterCat === '') {
+    $jsonLd = json_encode([
+        '@context' => 'https://schema.org',
+        '@type'    => 'WebSite',
+        'name'     => 'varlog',
+        'url'      => rtrim(APP_URL, '/') . '/',
+        'description' => 'Journal personnel de Cédrix. Informatique, hack et loisirs techniques.',
+        'inLanguage' => 'fr-FR',
+        'author'   => ['@type' => 'Person', 'name' => 'Cédrix'],
+    ], JSON_UNESCAPED_SLASHES | JSON_UNESCAPED_UNICODE);
+}
+
 include __DIR__ . '/layout.php';

templates/post_view.php

@@ -255,10 +255,45 @@ $hasSources     = (!empty($externalLinks) || !empty($files))
 $content            = ob_get_clean();
 $title              = htmlspecialchars($article['title']);
 $seoTitle           = ($article['seo_title'] ?? '') ?: $article['title'];
-$seoDescription     = $article['seo_description'] ?? '';
-$ogImage            = $article['og_image'] ?? '';
 $ogType             = 'article';
 $ogUrl              = url('post/' . rawurlencode($article['slug'] ?? ''));
+$canonical          = $ogUrl;
 $articlePublishedAt = $article['published_at'] ?? '';
 $mainClass          = 'container-fluid';
+
+// Auto-description depuis le contenu si le champ SEO est vide
+$seoDescription = $article['seo_description'] ?? '';
+if ($seoDescription === '') {
+    $plain = strip_tags((new Parsedown())->text($article['content']));
+    $plain = preg_replace('/\s+/', ' ', $plain);
+    $seoDescription = mb_strimwidth(trim((string)$plain), 0, 155, '…');
+}
+
+// og:image : cover puis fallback og_image du meta
+if ($ogImage === null || $ogImage === '') {
+    $ogImage = $article['og_image'] ?? '';
+}
+
+// JSON-LD Article
+$jsonLdData = [
+    '@context' => 'https://schema.org',
+    '@type'    => 'BlogPosting',
+    'headline' => $seoTitle,
+    'description' => $seoDescription,
+    'url'      => $canonical,
+    'datePublished' => date('c', strtotime((string)$articlePublishedAt)),
+    'dateModified'  => date('c', strtotime((string)($article['updated_at'] ?? $articlePublishedAt))),
+    'author'   => ['@type' => 'Person', 'name' => 'Cédrix'],
+    'publisher' => [
+        '@type' => 'Person',
+        'name'  => 'Cédrix',
+        'url'   => rtrim(APP_URL, '/'),
+    ],
+    'inLanguage' => 'fr-FR',
+];
+if (!empty($ogImage)) {
+    $jsonLdData['image'] = $ogImage;
+}
+$jsonLd = json_encode($jsonLdData, JSON_UNESCAPED_SLASHES | JSON_UNESCAPED_UNICODE);
+
 include __DIR__ . '/layout.php';