v1.6.31 : analyse complète des logs + détection bots

- AccessLogParser : tous chemins/statuts pour IPs publiques (ipAllPaths, ipAllDays, ipAgents)
- Détection bots par patterns (data/bots.json, ~50 patterns initiaux)
- Section « Agents détectés » en bas de page admin/stats avec badge 🤖
- Panneau d'édition des patterns bots (formulaire avec CSRF)
- Drill-down IP : section « Autres chemins » (hors articles/livres)

Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>

public/assets/js/admin-stats.js

@@ -1,4 +1,4 @@
-/* Admin stats : graphiques, sparklines, accordéon pays/AS/IP */
+/* Admin stats : graphiques, sparklines, accordéon pays/AS/IP, agents */
 
 function esc(s) {
     return String(s).replace(/&/g, '&amp;').replace(/</g, '&lt;').replace(/>/g, '&gt;').replace(/"/g, '&quot;');
@@ -8,6 +8,20 @@ function trunc(s, n) {
     return s.length > n ? s.slice(0, n) + '…' : s;
 }
 
+// Détection de bot par correspondance partielle insensible à la casse
+var _botPatterns = (typeof FOLIO_BOT_PATTERNS !== 'undefined') ? FOLIO_BOT_PATTERNS : [];
+function isBot(ua) {
+    if (!ua) { return false; }
+    var lo = ua.toLowerCase();
+    for (var i = 0; i < _botPatterns.length; i++) {
+        if (lo.indexOf(_botPatterns[i].toLowerCase()) !== -1) { return true; }
+    }
+    return false;
+}
+function botBadge(ua) {
+    return isBot(ua) ? '<span title="Bot connu" style="font-size:.85rem">🤖</span> ' : '';
+}
+
 // ── Visiteurs par pays ────────────────────────────────────────────────────────
 (function () {
     var el     = document.getElementById('stats-country-container');
@@ -27,7 +41,7 @@ function trunc(s, n) {
         return String.fromCodePoint(cp[0], cp[1]) + ' ';
     }
 
-    // Index IPs par ASN pour le drill-down
+    // Index IPs par ASN
     var ipsByAsn = {};
     Object.keys(ipData).forEach(function (ip) {
         var d   = ipData[ip];
@@ -39,7 +53,6 @@ function trunc(s, n) {
         ipsByAsn[k].sort(function (a, b) { return b.hits - a.hits; });
     });
 
-    // Mini sparkline (80x20px polyline) pour chaque IP
     function ipSparkline(daily) {
         if (!daily || !daily.length) { return ''; }
         var W = 80, H = 20, padX = 1, padY = 2;
@@ -89,36 +102,44 @@ function trunc(s, n) {
             var asnKey = n.asn || '__unknown__';
             var ips    = ipsByAsn[asnKey] || [];
 
-            // Lignes IP : adresse + agents à gauche, sparkline, chemins, hits
             var ipRows = ips.slice(0, 20).map(function (ipInfo) {
-                // Agents sous l'IP
+                // Agents sous l'IP avec badge bot
                 var agentsHtml = '';
                 (ipInfo.agents || []).forEach(function (ua) {
                     agentsHtml += '<div style="font-size:.65rem;color:#adb5bd;line-height:1.4;word-break:break-all">'
-                        + esc(trunc(ua, 55)) + '</div>';
+                        + botBadge(ua) + esc(trunc(ua, 55)) + '</div>';
                 });
 
-                // Chemins triés par date desc
-                var articles = [], books = [];
+                // Chemins triés : /post/ et /book/ avec ts, reste sans ts
+                var postBook = [], other = [];
                 Object.keys(ipInfo.paths || {}).forEach(function (path) {
                     var p   = ipInfo.paths[path];
                     var cnt = (p && typeof p === 'object') ? p.n : p;
                     var ts  = (p && typeof p === 'object') ? p.ts : 0;
-                    if (path.indexOf('/post/') === 0)      { articles.push({ path: path, cnt: cnt, ts: ts }); }
-                    else if (path.indexOf('/book/') === 0) { books.push({ path: path, cnt: cnt, ts: ts }); }
+                    if (ts > 0) { postBook.push({ path: path, cnt: cnt, ts: ts }); }
+                    else        { other.push({ path: path, cnt: cnt }); }
                 });
-                articles.sort(function (a, b) { return b.ts - a.ts; });
-                books.sort(function (a, b) { return b.ts - a.ts; });
+                postBook.sort(function (a, b) { return b.ts - a.ts; });
+                other.sort(function (a, b) { return b.cnt - a.cnt; });
 
                 function pathLine(p, prefix) {
-                    var slug = decodeURIComponent(p.path.replace(prefix, ''));
+                    var raw  = p.path.replace(prefix, '');
+                    var slug = '';
+                    try { slug = decodeURIComponent(raw); } catch (e) { slug = raw; }
                     return '<div style="font-size:.75rem;line-height:1.5">'
                         + '<a href="' + esc(p.path) + '" target="_blank" style="color:#495057">'
-                        + esc(trunc(slug, 40)) + '</a>'
+                        + esc(trunc(slug || p.path, 40)) + '</a>'
+                        + ' <span style="color:#adb5bd">(' + p.cnt + ')</span></div>';
+                }
+                function otherLine(p) {
+                    return '<div style="font-size:.72rem;color:#868e96;line-height:1.4">'
+                        + '<code style="font-size:.72rem;color:#868e96">' + esc(trunc(p.path, 44)) + '</code>'
                         + ' <span style="color:#adb5bd">(' + p.cnt + ')</span></div>';
                 }
 
                 var pathsHtml = '';
+                var articles = postBook.filter(function (p) { return p.path.indexOf('/post/') === 0; });
+                var books    = postBook.filter(function (p) { return p.path.indexOf('/book/') === 0; });
                 if (articles.length) {
                     pathsHtml += '<div style="font-size:.7rem;color:#adb5bd;margin-top:2px">Articles</div>'
                         + articles.map(function (p) { return pathLine(p, '/post/'); }).join('');
@@ -127,6 +148,10 @@ function trunc(s, n) {
                     pathsHtml += '<div style="font-size:.7rem;color:#adb5bd;margin-top:2px">Livres</div>'
                         + books.map(function (p) { return pathLine(p, '/book/'); }).join('');
                 }
+                if (other.length) {
+                    pathsHtml += '<div style="font-size:.7rem;color:#adb5bd;margin-top:2px">Autres chemins</div>'
+                        + other.map(otherLine).join('');
+                }
                 if (!pathsHtml) { pathsHtml = '<span style="font-size:.75rem;color:#adb5bd">—</span>'; }
 
                 return '<div class="d-flex gap-2 py-2 border-bottom align-items-start">'
@@ -141,9 +166,9 @@ function trunc(s, n) {
                     + '</div>';
             }).join('');
 
-            var hasIps       = ips.length > 0;
-            var toggleAttrs  = hasIps ? ' data-bs-toggle="collapse" data-bs-target="#' + asId + '" role="button"' : '';
-            var chevron      = hasIps ? '<span class="text-muted ms-1" style="font-size:.65rem">▾</span>' : '';
+            var hasIps      = ips.length > 0;
+            var toggleAttrs = hasIps ? ' data-bs-toggle="collapse" data-bs-target="#' + asId + '" role="button"' : '';
+            var chevron     = hasIps ? '<span class="text-muted ms-1" style="font-size:.65rem">▾</span>' : '';
 
             return '<div>'
                 + '<div class="d-flex align-items-center gap-2 py-1"' + toggleAttrs + '>'
@@ -184,6 +209,76 @@ function trunc(s, n) {
     el.innerHTML = html;
 }());
 
+// ── Liste consolidée de tous les agents ──────────────────────────────────────
+(function () {
+    var el     = document.getElementById('stats-agents-container');
+    var badge  = document.getElementById('agents-count');
+    var ipData = (typeof FOLIO_IP_DATA !== 'undefined') ? FOLIO_IP_DATA : {};
+    if (!el) { return; }
+
+    // Agréger toutes les UAs depuis FOLIO_IP_DATA
+    var uaCounts = {};
+    Object.keys(ipData).forEach(function (ip) {
+        (ipData[ip].agents || []).forEach(function (ua) {
+            uaCounts[ua] = (uaCounts[ua] || 0) + (ipData[ip].hits || 0);
+        });
+    });
+
+    var agents = Object.keys(uaCounts).map(function (ua) {
+        return { ua: ua, hits: uaCounts[ua], bot: isBot(ua) };
+    }).sort(function (a, b) {
+        // Bots d'abord, puis par hits desc
+        if (a.bot !== b.bot) { return a.bot ? -1 : 1; }
+        return b.hits - a.hits;
+    });
+
+    if (!agents.length) {
+        el.innerHTML = '<p class="text-muted p-3 mb-0">Aucun agent détecté.</p>';
+        return;
+    }
+
+    var bots    = agents.filter(function (a) { return a.bot; });
+    var unknown = agents.filter(function (a) { return !a.bot; });
+    if (badge) { badge.textContent = '— ' + bots.length + ' bot(s) détecté(s) sur ' + agents.length; }
+
+    function agentRow(a) {
+        return '<tr>'
+            + '<td class="ps-3" style="width:1.5rem;vertical-align:top;padding-top:6px">'
+            + (a.bot ? '<span title="Bot">🤖</span>' : '<span class="text-muted" title="Inconnu">?</span>') + '</td>'
+            + '<td style="word-break:break-all;vertical-align:top">'
+            + '<code style="font-size:.72rem">' + esc(a.ua) + '</code></td>'
+            + '<td class="text-end text-muted small pe-3" style="width:5rem;vertical-align:top;white-space:nowrap">'
+            + a.hits.toLocaleString('fr-FR') + '</td>'
+            + '</tr>';
+    }
+
+    var botsHtml    = bots.map(agentRow).join('');
+    var unknownHtml = unknown.map(agentRow).join('');
+
+    var html = '<div class="table-responsive">'
+        + '<table class="table table-sm table-hover mb-0 small">'
+        + '<thead class="table-light"><tr>'
+        + '<th class="ps-3" style="width:1.5rem"></th>'
+        + '<th>User-Agent</th>'
+        + '<th class="text-end pe-3" style="width:5rem">Req.</th>'
+        + '</tr></thead>'
+        + '<tbody>';
+
+    if (botsHtml) {
+        html += '<tr class="table-light"><td colspan="3" class="ps-3 py-1">'
+            + '<small class="fw-semibold text-muted">Bots connus (' + bots.length + ')</small></td></tr>'
+            + botsHtml;
+    }
+    if (unknownHtml) {
+        html += '<tr class="table-light"><td colspan="3" class="ps-3 py-1">'
+            + '<small class="fw-semibold text-muted">Agents non classés (' + unknown.length + ')</small></td></tr>'
+            + unknownHtml;
+    }
+
+    html += '</tbody></table></div>';
+    el.innerHTML = html;
+}());
+
 // ── Pages les plus visitées (RSS XML + sparklines) ───────────────────────────
 (function () {
     var container  = document.getElementById('stats-pages-container');
@@ -303,9 +398,7 @@ function trunc(s, n) {
             + '<path d="' + areaPath + '" fill="url(#area-grad)"/>'
             + '<path d="' + linePath + '" fill="none" stroke="var(--bs-primary,#0d6efd)"'
             + ' stroke-width="2" stroke-linejoin="round" stroke-linecap="round"/>'
-            + dots
-            + yLabels
-            + xLabels
+            + dots + yLabels + xLabels
             + '</svg>';
     }
 

public/index.php

@@ -2768,6 +2768,37 @@ switch ($action) {
             $adminData['as_groups']           = asGroups();
             $adminData['stats_pages_by_day']  = $statsRaw['pages_by_day'] ?? [];
             $adminData['stats_ip_data']       = $statsRaw['ip_data'] ?? [];
+
+            // Patterns de bots — initialisation si absent
+            $botsFile = DATA_PATH . '/bots.json';
+            if (!file_exists($botsFile)) {
+                $defaultBots = [
+                    'Googlebot','Googlebot-Image','Google-InspectionTool','Google-Extended',
+                    'bingbot','BingPreview','msnbot',
+                    'DuckDuckBot','DuckDuckGo-Favicons-Bot',
+                    'Baiduspider','YandexBot','YandexImages','YandexMetrika',
+                    'Applebot',
+                    'facebookexternalhit','facebot',
+                    'Twitterbot','LinkedInBot','Slackbot','TelegramBot','WhatsApp','Discordbot',
+                    'PetalBot','Bytespider','SogouSpider','SeznamBot','Exabot',
+                    'AhrefsBot','SemrushBot','MJ12bot','DotBot','rogerbot','BLEXBot','DataForSeoBot',
+                    'Screaming Frog SEO Spider',
+                    'ClaudeBot','GPTBot','Google-Extended','PerplexityBot','cohere-ai','anthropic-ai',
+                    'meta-externalagent','OAI-SearchBot','Amazonbot',
+                    'CCBot','ia_archiver','archive.org_bot',
+                    'NetcraftSurveyAgent',
+                    'python-requests','python-urllib','Python/',
+                    'curl/','wget/','Wget/',
+                    'Go-http-client/1','Java/','Apache-HttpClient','okhttp/',
+                    'Scrapy','HeadlessChrome','PhantomJS','Puppeteer','Playwright','Selenium',
+                    'UptimeRobot','Pingdom','StatusCake','Site24x7','GTmetrix',
+                    'Chrome-Lighthouse','PageSpeed','Zabbix','check_http',
+                    'libwww-perl','GuzzleHttp','masscan','zgrab','nuclei',
+                ];
+                @mkdir(dirname($botsFile), 0755, true);
+                @file_put_contents($botsFile, json_encode($defaultBots, JSON_UNESCAPED_UNICODE | JSON_PRETTY_PRINT));
+            }
+            $adminData['bot_patterns'] = json_decode((string) file_get_contents($botsFile), true) ?: [];
         }
 
         if ($tab === 'categories') {
@@ -3206,6 +3237,20 @@ switch ($action) {
         header('Location: /admin/stats?' . ($ok ? 'saved=1' : 'error=write'));
         exit;
 
+    case 'admin_save_bots':
+        requireAuth();
+        if (!isAdmin() || $_SERVER['REQUEST_METHOD'] !== 'POST') {
+            http_response_code(403);
+            exit;
+        }
+        $botsFile = DATA_PATH . '/bots.json';
+        $patterns = array_values(array_unique(array_filter(
+            array_map('trim', explode("\n", (string) ($_POST['bot_patterns'] ?? '')))
+        )));
+        $ok = @file_put_contents($botsFile, json_encode($patterns, JSON_UNESCAPED_UNICODE | JSON_PRETTY_PRINT)) !== false;
+        header('Location: /admin/stats?' . ($ok ? 'saved=1' : 'error=write'));
+        exit;
+
     case 'admin_create_role':
         requireAuth();
         if (!isAdmin() || $_SERVER['REQUEST_METHOD'] !== 'POST') {

public/version.txt

@@ -1 +1 @@
-1.6.30
+1.6.31