Remove accidental crawl log file

README.md

@@ -2,7 +2,6 @@
 
 Eine schlanke Schnittstelle zwischen **Contao CMS (4.13 / 5.6 / 5.7 ready) unter PHP 8.4** und einer **selbst gehosteten Meilisearch-Instanz**.  
 Das Bundle erweitert den Contao-Suchindex um strukturierte Daten und ermöglicht eine performante, moderne Volltextsuche.
-Das Parsen von Dateien erfolgt über eine Apache-Tika-Instanz, welche extern bereitgestellt werden muss.
 
 ---
 
@@ -21,53 +20,13 @@ Das Parsen von Dateien erfolgt über eine Apache-Tika-Instanz, welche extern ber
 - Kompatibel mit:
     - Contao **4.13**, **5.6** und **5.7**
     - PHP **8.4**
+- Entwickelt als **eigenständiges Contao-Bundle**
 
 ---
 
-## ⏱️ Scheduled Indexing (Cron setup)
+## 📦 Installation
 
-Das Bundle stellt eigene Commands zur Verfügung, um Dateien zu bereinigen und den Meilisearch-Index neu aufzubauen.  
-Für den produktiven Einsatz wird empfohlen, diese Commands regelmäßig per **System-Crontab** auszuführen.
+Installation über Composer:
 
-Das Bundle nutzt **keinen eigenen Contao-Cron**, sondern System-Cronjobs.
-
-## Verfügbare Commands
-
-### Datei-Cleanup
-
-```
-/vendor/bin/contao-console meilisearch:files:cleanup
-```
-
-### Datei-Parsing
-
-```
-/vendor/bin/contao-console meilisearch:files:parse
-```
-
-### Meilisearch-Index
-
-```
-/vendor/bin/contao-console meilisearch:index
-```
-
-
-
-## Beispiel Crontab
-
-```
-0 5 * * *  /usr/bin/php8.4 /path/to/project/vendor/bin/contao-console meilisearch:files:cleanup
-1 5 * * *  /usr/bin/php8.4 /path/to/project/vendor/bin/contao-console contao:crawl
-10 5 * * * /usr/bin/php8.4 /path/to/project/vendor/bin/contao-console meilisearch:files:parse
-20 5 * * * /usr/bin/php8.4 /path/to/project/vendor/bin/contao-console meilisearch:index
-```
-
-## Logging
-
-```
->> var/logs/meilisearch_cron.log 2>&1
-```
-
-## Lizenz
-
-MIT
+```bash
+composer require mummertmedia/contao-meilisearch-bundle:^0.1

composer.json

@@ -1,5 +1,5 @@
 {
-  "name": "mummert/contao-meilisearch-bundle",
+  "name": "mummert-media/contao-meilisearch-bundle",
   "description": "Contao Meilisearch integration bundle",
   "type": "contao-bundle",
   "license": "MIT",
@@ -8,14 +8,18 @@
     "contao/core-bundle": "^4.13 || ^5.6 || ^5.7",
     "contao/calendar-bundle": "^4.13 || ^5.6 || ^5.7",
     "contao/news-bundle": "^4.13 || ^5.6 || ^5.7",
-    "meilisearch/meilisearch-php": "^1.16"
+    "meilisearch/meilisearch-php": "^1.16",
+    "smalot/pdfparser": "^2.12",
+    "phpoffice/phpword": "^1.4",
+    "phpoffice/phpspreadsheet": "^3.0",
+    "phpoffice/phppresentation": "^1.2"
   },
   "autoload": {
     "psr-4": {
-      "Mummert\\ContaoMeilisearchBundle\\": "src/"
+      "MummertMedia\\ContaoMeilisearchBundle\\": "src/"
     }
   },
   "extra": {
-    "contao-manager-plugin": "Mummert\\ContaoMeilisearchBundle\\ContaoManager\\Plugin"
+    "contao-manager-plugin": "MummertMedia\\ContaoMeilisearchBundle\\ContaoManager\\Plugin"
   }
 }

src/Command/MeilisearchFilesCleanupCommand.php

@@ -1,100 +0,0 @@
-<?php
-
-namespace Mummert\ContaoMeilisearchBundle\Command;
-
-use Contao\CoreBundle\Framework\ContaoFramework;
-use Doctrine\DBAL\Connection;
-use Symfony\Component\Console\Command\Command;
-use Symfony\Component\Console\Input\InputInterface;
-use Symfony\Component\Console\Input\InputOption;
-use Symfony\Component\Console\Output\OutputInterface;
-
-class MeilisearchFilesCleanupCommand extends Command
-{
-    public function __construct(
-        private readonly ContaoFramework $framework,
-        private readonly Connection $connection,
-    ) {
-        parent::__construct();
-    }
-
-    protected function configure(): void
-    {
-        $this
-            ->setName('meilisearch:files:cleanup')
-            ->setDescription('Remove stale indexed files from tl_search_files')
-            ->addOption(
-                'grace',
-                null,
-                InputOption::VALUE_OPTIONAL,
-                'Grace period in seconds (files newer than now-grace are kept)',
-                86400
-            )
-            ->addOption(
-                'dry-run',
-                null,
-                InputOption::VALUE_NONE,
-                'Show how many entries would be removed without deleting them'
-            );
-    }
-
-    protected function execute(InputInterface $input, OutputInterface $output): int
-    {
-        $this->framework->initialize();
-
-        $this->log('Cleaner gestartet');
-
-        try {
-            $grace  = max(0, (int) $input->getOption('grace'));
-            $dryRun = (bool) $input->getOption('dry-run');
-            $cutoff = time() - $grace;
-
-            if ($dryRun) {
-                $count = $this->connection->fetchOne(
-                    'SELECT COUNT(*) FROM tl_search_files WHERE last_seen < ?',
-                    [$cutoff]
-                );
-
-                $message = sprintf(
-                    '[DRY-RUN] %d stale file(s) would be removed (last_seen < %s)',
-                    $count,
-                    date('Y-m-d H:i:s', $cutoff)
-                );
-
-                $output->writeln('<comment>' . $message . '</comment>');
-                $this->log($message);
-
-                $this->log('Cleaner stopped (dry-run)');
-                return Command::SUCCESS;
-            }
-
-            $affected = $this->connection->executeStatement(
-                'DELETE FROM tl_search_files WHERE last_seen < ?',
-                [$cutoff]
-            );
-
-            $message = sprintf(
-                'Removed %d stale file(s) (last_seen < %s)',
-                $affected,
-                date('Y-m-d H:i:s', $cutoff)
-            );
-
-            $output->writeln('<info>' . $message . '</info>');
-            $this->log($message);
-
-            $this->log('Cleaner successfully stopped');
-            return Command::SUCCESS;
-
-        } catch (\Throwable $e) {
-            $this->log('Cleaner ERROR: ' . $e->getMessage());
-            $output->writeln('<error>' . $e->getMessage() . '</error>');
-
-            return Command::FAILURE;
-        }
-    }
-
-    private function log(string $message): void
-    {
-        error_log(sprintf('[%s] %s', date('Y-m-d H:i:s'), $message));
-    }
-}

src/Command/MeilisearchFilesParseCommand.php

@@ -1,277 +0,0 @@
-<?php
-
-namespace Mummert\ContaoMeilisearchBundle\Command;
-
-use Contao\CoreBundle\Framework\ContaoFramework;
-use Contao\Database;
-use Contao\System;
-use Symfony\Component\Console\Command\Command;
-use Symfony\Component\Console\Input\InputInterface;
-use Symfony\Component\Console\Input\InputOption;
-use Symfony\Component\Console\Output\OutputInterface;
-use Symfony\Component\HttpClient\HttpClient;
-
-class MeilisearchFilesParseCommand extends Command
-{
-    public function __construct(
-        private readonly ContaoFramework $framework,
-    ) {
-        parent::__construct();
-    }
-
-    protected function configure(): void
-    {
-        $this
-            ->setName('meilisearch:files:parse')
-            ->setDescription('Parse indexed files via Apache Tika and enrich tl_search_files')
-            ->addOption(
-                'limit',
-                null,
-                InputOption::VALUE_OPTIONAL,
-                'Maximum number of files to check per run'
-            )
-            ->addOption(
-                'dry-run',
-                null,
-                InputOption::VALUE_NONE,
-                'Do not send files to Tika'
-            );
-    }
-
-    protected function execute(InputInterface $input, OutputInterface $output): int
-    {
-        $this->framework->initialize();
-        $this->log('Parser gestartet');
-
-        $dryRun = (bool) $input->getOption('dry-run');
-
-        $limitOption = $input->getOption('limit');
-        $limit = $limitOption !== null ? max(1, (int) $limitOption) : null;
-
-        $tikaUrl = rtrim((string) ($GLOBALS['TL_CONFIG']['meilisearch_tika_url'] ?? ''), '/');
-        if ($tikaUrl === '') {
-            $output->writeln('<error>Tika URL not configured</error>');
-            return Command::FAILURE;
-        }
-
-        $db = Database::getInstance();
-
-        $sql = "SELECT * FROM tl_search_files ORDER BY tstamp ASC";
-        if ($limit !== null) {
-            $sql .= " LIMIT " . (int) $limit;
-        }
-
-        $files = $db->query($sql)->fetchAllAssoc();
-
-        if (!$files) {
-            $this->log('No files to parse');
-            return Command::SUCCESS;
-        }
-
-        $client = HttpClient::create([
-            'timeout' => 180,
-        ]);
-
-        foreach ($files as $file) {
-
-            $originalUrl   = (string) $file['url'];
-            $existingTitle = trim((string) ($file['title'] ?? ''));
-            $normalized    = $originalUrl;
-
-            // -------------------------------------------------
-            // Normalize URL
-            // -------------------------------------------------
-            if (str_contains($normalized, '?')) {
-                $parts = parse_url($normalized);
-                if (!empty($parts['query'])) {
-                    parse_str($parts['query'], $query);
-                    if (!empty($query['file'])) {
-                        $normalized = (string) $query['file'];
-                    } else {
-                        $this->log('Not a direct file url, skip', ['url' => $originalUrl]);
-                        continue;
-                    }
-                }
-            }
-
-            $normalized = strtok($normalized, '#');
-            $normalized = rawurldecode($normalized);
-            $normalized = ltrim($normalized, '/');
-
-            if (!str_starts_with($normalized, 'files/')) {
-                $this->log('Not in files/, skip', ['url' => $originalUrl]);
-                continue;
-            }
-
-            $root = defined('TL_ROOT')
-                ? TL_ROOT
-                : System::getContainer()->getParameter('kernel.project_dir') . '/public';
-
-            $absolutePath = $root . '/' . $normalized;
-
-            if (!is_file($absolutePath)) {
-                $this->log('File missing, skip', [
-                    'url'  => $originalUrl,
-                    'path' => $absolutePath,
-                ]);
-                continue;
-            }
-
-            $mtime    = filemtime($absolutePath) ?: 0;
-            $checksum = md5($normalized . '|' . $mtime);
-
-            // -------------------------------------------------
-            // Skip unchanged
-            // -------------------------------------------------
-            if ($file['checksum'] === $checksum && !empty($file['text'])) {
-                continue;
-            }
-
-            if ($dryRun) {
-                $output->writeln('[DRY-RUN] Would parse: ' . $normalized);
-                continue;
-            }
-
-            // -------------------------------------------------
-            // MIME-Type
-            // -------------------------------------------------
-            $ext = strtolower(pathinfo($normalized, PATHINFO_EXTENSION));
-
-            $mimeType = match ($ext) {
-                'pdf'  => 'application/pdf',
-                'docx' => 'application/vnd.openxmlformats-officedocument.wordprocessingml.document',
-                'xlsx' => 'application/vnd.openxmlformats-officedocument.spreadsheetml.sheet',
-                'pptx' => 'application/vnd.openxmlformats-officedocument.presentationml.presentation',
-                default => null,
-            };
-
-            if ($mimeType === null) {
-                $this->log('Unsupported file type, skip', ['url' => $normalized]);
-                continue;
-            }
-
-            // -------------------------------------------------
-            // Tika BODY (roher Plaintext)
-            // -------------------------------------------------
-            try {
-                $this->log('Parsing file', ['url' => $normalized]);
-
-                $bodyResponse = $client->request(
-                    'PUT',
-                    $tikaUrl . '/tika/main',
-                    [
-                        'headers' => [
-                            'Accept'       => 'text/plain',
-                            'Content-Type' => $mimeType,
-                        ],
-                        'body' => fopen($absolutePath, 'rb'),
-                    ]
-                );
-
-                $text = trim((string) $bodyResponse->getContent(false));
-
-            } catch (\Throwable $e) {
-                $this->log('Body parse failed', [
-                    'url'   => $normalized,
-                    'error' => $e->getMessage(),
-                ]);
-                continue;
-            }
-
-            // -------------------------------------------------
-            // TITLE: keep existing editor-defined title
-            // -------------------------------------------------
-            $title = $existingTitle !== '' ? $existingTitle : null;
-
-            // -------------------------------------------------
-            // Tika METADATA (Title) – only if no existing title
-            // -------------------------------------------------
-            if ($title === null) {
-                try {
-                    $metaResponse = $client->request(
-                        'PUT',
-                        $tikaUrl . '/meta',
-                        [
-                            'headers' => [
-                                'Accept'       => 'application/json',
-                                'Content-Type' => $mimeType,
-                            ],
-                            'body' => fopen($absolutePath, 'rb'),
-                        ]
-                    );
-
-                    $meta = json_decode($metaResponse->getContent(false), true);
-
-                    $rawTitle =
-                        $meta['dc:title'][0]
-                        ?? $meta['pdf:docinfo:title'][0]
-                        ?? null;
-
-                    if ($rawTitle) {
-                        $title = html_entity_decode(
-                            $rawTitle,
-                            ENT_QUOTES | ENT_HTML5,
-                            'UTF-8'
-                        );
-                    }
-
-                } catch (\Throwable) {
-                    // Metadata optional
-                }
-            }
-
-            // -------------------------------------------------
-            // TITLE → ASCII SAFE (only if newly generated)
-            // -------------------------------------------------
-            if ($existingTitle === '' && $title) {
-                $title = iconv('UTF-8', 'ASCII//TRANSLIT//IGNORE', $title);
-                $title = preg_replace('/\s+/', ' ', $title);
-                $title = trim($title);
-            }
-
-            // -------------------------------------------------
-            // FALLBACK: Dateiname (only if still empty)
-            // -------------------------------------------------
-            if (!$title || strlen($title) < 5) {
-                $title = pathinfo($normalized, PATHINFO_FILENAME);
-                $title = str_replace(['_', '-'], ' ', $title);
-                $title = preg_replace('/\s+/', ' ', $title);
-                $title = trim($title);
-            }
-
-            // -------------------------------------------------
-            // Store result
-            // -------------------------------------------------
-            $db->prepare(
-                "UPDATE tl_search_files
-                 SET text = ?, title = ?, checksum = ?, file_mtime = ?, tstamp = ?
-                 WHERE id = ?"
-            )->execute(
-                $text,
-                $title,
-                $checksum,
-                $mtime,
-                time(),
-                $file['id']
-            );
-
-            $this->log('File parsed', [
-                'url'   => $normalized,
-                'chars' => mb_strlen($text),
-                'title' => $title,
-            ]);
-        }
-
-        $this->log('Parser finished');
-        return Command::SUCCESS;
-    }
-
-    private function log(string $message, array $context = []): void
-    {
-        $ctx = $context
-            ? ' | ' . json_encode($context, JSON_UNESCAPED_SLASHES | JSON_UNESCAPED_UNICODE)
-            : '';
-
-        error_log('[MeilisearchFilesParse] ' . $message . $ctx);
-    }
-}

src/Command/MeilisearchIndexCommand.php

@@ -1,8 +1,8 @@
 <?php
 
-namespace Mummert\ContaoMeilisearchBundle\Command;
+namespace MummertMedia\ContaoMeilisearchBundle\Command;
 
-use Mummert\ContaoMeilisearchBundle\Service\MeilisearchIndexService;
+use MummertMedia\ContaoMeilisearchBundle\Service\MeilisearchIndexService;
 use Symfony\Component\Console\Command\Command;
 use Symfony\Component\Console\Input\InputInterface;
 use Symfony\Component\Console\Output\OutputInterface;
@@ -24,34 +24,12 @@ class MeilisearchIndexCommand extends Command
 
     protected function execute(InputInterface $input, OutputInterface $output): int
     {
-        $this->log('Meilisearch index gestartet');
         $output->writeln('<info>Meilisearch index started</info>');
 
-        try {
         $this->indexService->run();
 
-            $this->log('Meilisearch index successfully stopped');
         $output->writeln('<info>Meilisearch index finished</info>');
 
         return Command::SUCCESS;
-
-        } catch (\Throwable $e) {
-            $this->log('Meilisearch index ERROR: ' . $e->getMessage());
-            $output->writeln('<error>' . $e->getMessage() . '</error>');
-
-            return Command::FAILURE;
-        }
-    }
-
-    /**
-     * Einheitliches Logging mit Zeitstempel
-     */
-    private function log(string $message): void
-    {
-        error_log(sprintf(
-            '[%s] %s',
-            date('Y-m-d H:i:s'),
-            $message
-        ));
     }
 }

src/ContaoManager/Plugin.php

@@ -1,6 +1,6 @@
 <?php
 
-namespace Mummert\ContaoMeilisearchBundle\ContaoManager;
+namespace MummertMedia\ContaoMeilisearchBundle\ContaoManager;
 
 use Contao\CalendarBundle\ContaoCalendarBundle;
 use Contao\CoreBundle\ContaoCoreBundle;
@@ -8,7 +8,7 @@ use Contao\ManagerPlugin\Bundle\BundlePluginInterface;
 use Contao\ManagerPlugin\Bundle\Config\BundleConfig;
 use Contao\ManagerPlugin\Bundle\Parser\ParserInterface;
 use Contao\NewsBundle\ContaoNewsBundle;
-use Mummert\ContaoMeilisearchBundle\ContaoMeilisearchBundle;
+use MummertMedia\ContaoMeilisearchBundle\ContaoMeilisearchBundle;
 
 class Plugin implements BundlePluginInterface
 {

src/ContaoMeilisearchBundle.php

@@ -1,6 +1,6 @@
 <?php
 
-namespace Mummert\ContaoMeilisearchBundle;
+namespace MummertMedia\ContaoMeilisearchBundle;
 
 use Symfony\Component\HttpKernel\Bundle\Bundle;
 

src/Controller/FrontendModule/MeilisearchSearchController.php

@@ -1,6 +1,6 @@
 <?php
 
-namespace Mummert\ContaoMeilisearchBundle\Controller\FrontendModule;
+namespace MummertMedia\ContaoMeilisearchBundle\Controller\FrontendModule;
 
 use Contao\Config;
 use Contao\CoreBundle\Controller\FrontendModule\AbstractFrontendModuleController;

src/Cron/MeilisearchIndexCron.php

@@ -0,0 +1,22 @@
+<?php
+namespace MummertMedia\ContaoMeilisearchBundle\Cron;
+
+use Contao\CoreBundle\Framework\ContaoFramework;
+use MummertMedia\ContaoMeilisearchBundle\Service\MeilisearchIndexService;
+
+class MeilisearchIndexCron
+{
+    public function __construct(
+        private readonly MeilisearchIndexService $indexService,
+        private readonly ContaoFramework $framework,
+    ) {}
+
+    public function __invoke(): void
+    {
+        // Contao initialisieren (wichtig!)
+        $this->framework->initialize();
+
+        // einmal täglich indexieren
+        $this->indexService->run();
+    }
+}

src/DependencyInjection/ContaoMeilisearchExtension.php

@@ -1,6 +1,6 @@
 <?php
 
-namespace Mummert\ContaoMeilisearchBundle\DependencyInjection;
+namespace MummertMedia\ContaoMeilisearchBundle\DependencyInjection;
 
 use Symfony\Component\Config\FileLocator;
 use Symfony\Component\DependencyInjection\ContainerBuilder;

src/EventListener/IndexPageListener.php

@@ -1,20 +1,22 @@
 <?php
 
-namespace Mummert\ContaoMeilisearchBundle\EventListener;
+namespace MummertMedia\ContaoMeilisearchBundle\EventListener;
 
 use Contao\Config;
-use Contao\System;
-use Mummert\ContaoMeilisearchBundle\Service\MeilisearchFileHelper;
+use MummertMedia\ContaoMeilisearchBundle\Service\PdfIndexService;
+use MummertMedia\ContaoMeilisearchBundle\Service\OfficeIndexService;
 
 class IndexPageListener
 {
     public function __construct(
-        private readonly MeilisearchFileHelper $fileHelper,
-    ) {
-    }
+        private readonly PdfIndexService $pdfIndexService,
+        private readonly OfficeIndexService $officeIndexService,
+    ) {}
 
     private function debug(string $message, array $context = []): void
     {
+        // Debug bewusst immer aktiv (bis du es wieder entfernst)
+        // Kontext kurz halten, damit Logs nicht explodieren
         $ctx = $context ? ' | ' . json_encode($context, JSON_UNESCAPED_SLASHES | JSON_UNESCAPED_UNICODE) : '';
         error_log('[ContaoMeilisearch][IndexPageListener] ' . $message . $ctx);
     }
@@ -28,6 +30,22 @@ class IndexPageListener
             'set_keys'  => array_keys($set),
         ]);
 
+        /*
+         * =====================
+         * PDF: Reset genau 1× pro Crawl
+         * =====================
+         */
+        try {
+            $this->debug('PDF resetTableOnce(): call');
+            $this->pdfIndexService->resetTableOnce();
+            $this->debug('PDF resetTableOnce(): ok');
+        } catch (\Throwable $e) {
+            $this->debug('PDF resetTableOnce(): failed', [
+                'error' => $e->getMessage(),
+                'class' => $e::class,
+            ]);
+        }
+
         /*
          * =====================
          * SEITEN-METADATEN
@@ -76,6 +94,8 @@ class IndexPageListener
                     $parsed['page']['keywords']  ?? null,
                 ];
 
+                $this->debug('Meta: keyword sources', ['sources' => $keywordSources]);
+
                 $keywords = [];
                 foreach ($keywordSources as $src) {
                     if (!is_string($src) || trim($src) === '') {
@@ -90,17 +110,33 @@ class IndexPageListener
                     $set['keywords'] = implode(' ', array_unique($keywords));
                 }
 
-                // IMAGEPATH
-                if (!empty($parsed['page']['searchimage'] ?? null)) {
-                    $set['imagepath'] = trim((string) $parsed['page']['searchimage']);
+                $this->debug('Meta: keywords result', [
+                    'keywords' => $set['keywords'] ?? null,
+                ]);
+
+                // IMAGEPATH (UUID)
+                $searchImage = $parsed['page']['searchimage'] ?? null;
+                $this->debug('Meta: searchimage candidate', ['searchimage' => $searchImage]);
+
+                if (!empty($searchImage)) {
+                    // >>> HINWEIS: falls dein tl_search-Feld "image" heißt, hier auf $set['image'] ändern!
+                    $set['imagepath'] = trim((string) $searchImage);
                 }
 
                 // STARTDATE
-                if (is_numeric($parsed['event']['startDate'] ?? null)) {
-                    $set['startDate'] = (int) $parsed['event']['startDate'];
+                $startDate =
+                    $parsed['event']['startDate']
+                    ?? $parsed['news']['startDate']
+                    ?? null;
+
+                $this->debug('Meta: startDate candidate', ['startDate' => $startDate]);
+
+                if (is_numeric($startDate) && (int) $startDate > 0) {
+                    $set['startDate'] = (int) $startDate;
                 }
 
                 // CHECKSUM
+                try {
                     $checksumSeed  = (string) ($data['checksum'] ?? '');
                     $checksumSeed .= '|' . ($set['keywords']  ?? '');
                     $checksumSeed .= '|' . ($set['priority']  ?? '');
@@ -108,53 +144,109 @@ class IndexPageListener
                     $checksumSeed .= '|' . ($set['startDate'] ?? '');
 
                     $set['checksum'] = md5($checksumSeed);
+
+                    $this->debug('Checksum generated', [
+                        'seed_preview' => substr($checksumSeed, 0, 120) . (strlen($checksumSeed) > 120 ? '…' : ''),
+                        'checksum'     => $set['checksum'],
+                    ]);
+                } catch (\Throwable $e) {
+                    $this->debug('Failed to generate checksum', [
+                        'error' => $e->getMessage(),
+                        'class' => $e::class,
+                    ]);
+                }
+
+                $this->debug('Meta: final set snapshot', [
+                    'priority'  => $set['priority'] ?? null,
+                    'keywords'  => $set['keywords'] ?? null,
+                    'imagepath' => $set['imagepath'] ?? null,
+                    'startDate' => $set['startDate'] ?? null,
+                    'checksum'  => $set['checksum'] ?? null,
+                ]);
             }
         }
 
         /*
          * =====================
-         * DATEI-ERKENNUNG (NUR ERKENNUNG!)
+         * DATEI-INDEXIERUNG (PDF / OFFICE)
          * =====================
          */
         if ((int) ($data['protected'] ?? 0) !== 0) {
+            $this->debug('Abort: protected page', ['protected' => $data['protected'] ?? null]);
             return;
         }
 
-        if (!Config::get('meilisearch_index_files')) {
+        $indexPdfs   = (bool) Config::get('meilisearch_index_pdfs');
+        $indexOffice = (bool) Config::get('meilisearch_index_office');
+
+        $this->debug('File indexing settings', [
+            'meilisearch_index_pdfs'   => $indexPdfs,
+            'meilisearch_index_office' => $indexOffice,
+        ]);
+
+        if (!$indexPdfs && !$indexOffice) {
+            $this->debug('Abort: file indexing disabled');
             return;
         }
 
         $links = $this->findAllLinks($content);
-        $fileLinks = [];
+        $this->debug('Links found', ['count' => count($links)]);
+
+        $pdfLinks    = [];
+        $officeLinks = [];
 
         foreach ($links as $link) {
             $type = $this->detectIndexableFileType($link['url']);
-            if ($type !== null) {
-                $fileLinks[] = $link + ['type' => $type];
+
+            if ($type === 'pdf' && $indexPdfs) {
+                $pdfLinks[] = $link;
+                continue;
+            }
+
+            if (in_array($type, ['docx', 'xlsx', 'pptx'], true) && $indexOffice) {
+                $officeLinks[] = $link;
             }
         }
 
-        $this->debug('Indexable file links found', [
-            'count' => count($fileLinks),
+        $this->debug('Indexable file links', [
+            'pdf'    => count($pdfLinks),
+            'office' => count($officeLinks),
         ]);
 
-        if ($fileLinks) {
-            foreach ($fileLinks as $file) {
-                $this->fileHelper->collect(
-                    $file['url'],
-                    $file['type'],
-                    (int) ($data['pid'] ?? 0)
-                );
+        try {
+            if ($pdfLinks !== []) {
+                $this->debug('PDF handlePdfLinks(): call', ['count' => count($pdfLinks)]);
+                $this->pdfIndexService->handlePdfLinks($pdfLinks);
+                $this->debug('PDF handlePdfLinks(): ok');
             }
+
+            if ($officeLinks !== []) {
+                $this->debug('Office handleOfficeLinks(): call', ['count' => count($officeLinks)]);
+                $this->officeIndexService->handleOfficeLinks($officeLinks);
+                $this->debug('Office handleOfficeLinks(): ok');
+            }
+        } catch (\Throwable $e) {
+            $this->debug('File indexing failed', [
+                'error' => $e->getMessage(),
+                'class' => $e::class,
+            ]);
         }
 
         $this->debug('Hook end', [
             'final_set_keys' => array_keys($set),
+            'final_set'      => [
+                'priority'  => $set['priority'] ?? null,
+                'keywords'  => $set['keywords'] ?? null,
+                'imagepath' => $set['imagepath'] ?? null,
+                'startDate' => $set['startDate'] ?? null,
+                'checksum'  => $set['checksum'] ?? null,
+            ],
         ]);
     }
 
-    /* === Hilfsmethoden unverändert === */
-
+    /**
+     * Extrahiert MEILISEARCH_JSON aus HTML-Kommentar
+     */
     private function extractMeilisearchJson(string $content): ?array
     {
         if (!preg_match('/<!--\s*MEILISEARCH_JSON\s*(\{.*?\})\s*-->/s', $content, $m)) {
@@ -169,6 +261,9 @@ class IndexPageListener
             : null;
     }
 
+    /**
+     * Sammle alle <a href="…"> Links
+     */
     private function findAllLinks(string $content): array
     {
         if (!preg_match_all(
@@ -191,11 +286,20 @@ class IndexPageListener
         return $result;
     }
 
+    /**
+     * Ermittelt indexierbaren Dateityp (pdf|docx|xlsx|pptx) oder null
+     */
     private function detectIndexableFileType(string $url): ?string
     {
+        // Hash entfernen
         $url = strtok($url, '#');
-        $parts = parse_url($url);
 
+        $parts = parse_url($url);
+        if (!$parts) {
+            return null;
+        }
+
+        // direkter Pfad (/files/…)
         if (!empty($parts['path'])) {
             $ext = strtolower(pathinfo($parts['path'], PATHINFO_EXTENSION));
             if (in_array($ext, ['pdf', 'docx', 'xlsx', 'pptx'], true)) {
@@ -203,12 +307,18 @@ class IndexPageListener
             }
         }
 
+        // Query-Parameter (Contao 4 + 5)
         if (!empty($parts['query'])) {
             parse_str($parts['query'], $query);
 
             foreach (['file', 'p', 'f'] as $param) {
                 if (!empty($query[$param])) {
-                    $candidate = rawurldecode(html_entity_decode((string) $query[$param], ENT_QUOTES));
+                    $candidate = (string) $query[$param];
+
+                    // sicher decodieren (Contao 4 + 5)
+                    $candidate = html_entity_decode($candidate, ENT_QUOTES);
+                    $candidate = rawurldecode($candidate);
+
                     $ext = strtolower(pathinfo($candidate, PATHINFO_EXTENSION));
 
                     if (in_array($ext, ['pdf', 'docx', 'xlsx', 'pptx'], true)) {

src/EventListener/MeilisearchPageMarkerListener.php

@@ -1,6 +1,6 @@
 <?php
 
-namespace Mummert\ContaoMeilisearchBundle\EventListener;
+namespace MummertMedia\ContaoMeilisearchBundle\EventListener;
 
 use Contao\CalendarEventsModel;
 use Contao\Config;
@@ -16,13 +16,6 @@ class MeilisearchPageMarkerListener
             return $buffer;
         }
 
-        // ⛔ Marker bereits vorhanden → nichts mehr tun
-        if (str_contains($buffer, '⟦MEILISEARCH_META⟧')
-            || str_contains($buffer, 'MEILISEARCH_JSON')
-        ) {
-            return $buffer;
-        }
-
         $data = [];
 
         /*

src/Resources/config/services.yaml

@@ -2,24 +2,24 @@ services:
   # Alias MUSS vorhanden sein (richtig platziert)
   Psr\Container\ContainerInterface: '@service_container'
 
-  Mummert\ContaoMeilisearchBundle\:
-    resource: '../../{Command,EventListener,Service}'
+  MummertMedia\ContaoMeilisearchBundle\:
+    resource: '../../{Command,Cron,EventListener,Service}'
     autowire: true
     autoconfigure: true
 
-  Mummert\ContaoMeilisearchBundle\EventListener\MeilisearchPageMarkerListener:
-    autowire: true
-    autoconfigure: false
-    tags:
-      - { name: contao.hook, hook: outputFrontendTemplate, method: onOutputFrontendTemplate }
-
-  Mummert\ContaoMeilisearchBundle\EventListener\IndexPageListener:
+  MummertMedia\ContaoMeilisearchBundle\EventListener\IndexPageListener:
     autowire: true
     autoconfigure: false
     tags:
       - { name: contao.hook, hook: indexPage, method: onIndexPage }
 
-  Mummert\ContaoMeilisearchBundle\Controller\FrontendModule\MeilisearchSearchController:
+  MummertMedia\ContaoMeilisearchBundle\Cron\MeilisearchIndexCron:
+    autowire: true
+    autoconfigure: false
+    tags:
+      - { name: contao.cron, interval: daily, method: __invoke }
+
+  MummertMedia\ContaoMeilisearchBundle\Controller\FrontendModule\MeilisearchSearchController:
     autowire: true
     autoconfigure: false
     tags:

src/Resources/contao/config/config.php

@@ -1,6 +1,6 @@
 <?php
 
-use Mummert\ContaoMeilisearchBundle\EventListener\MeilisearchPageMarkerListener;
+use MummertMedia\ContaoMeilisearchBundle\EventListener\MeilisearchPageMarkerListener;
 
 
 $GLOBALS['TL_HOOKS']['outputFrontendTemplate'][] = [

src/Resources/contao/dca/tl_search_pdf.php

@@ -2,18 +2,16 @@
 
 use Contao\DC_Table;
 
-$GLOBALS['TL_DCA']['tl_search_files'] = [
+$GLOBALS['TL_DCA']['tl_search_pdf'] = [
     'config' => [
         'dataContainer' => DC_Table::class,
         'sql' => [
             'keys' => [
                 'id'       => 'primary',
+                'checksum' => 'unique',
                 'page_id'  => 'index',
-                'url'       => 'unique',
-                'type'      => 'index',
-                'checksum'  => 'index',
-                'uuid'      => 'index',
-                'last_seen' => 'index',
+                'url'      => 'index',
+                'type'     => 'index', // ⬅️ NEU
             ],
         ],
     ],
@@ -27,18 +25,10 @@ $GLOBALS['TL_DCA']['tl_search_files'] = [
             'sql' => "int(10) unsigned NOT NULL default 0",
         ],
 
-        /*
-         * Zeitpunkt, wann die Datei zuletzt beim Crawl gesehen wurde
-         * → Basis für Cleanup
-         */
-        'last_seen' => [ // ⬅️ NEU
-            'sql' => "int(10) unsigned NOT NULL default 0",
-        ],
-
         /*
          * Dateityp: pdf | docx | xlsx | pptx
          */
-        'type' => [
+        'type' => [ // ⬅️ NEU
             'sql' => "varchar(16) NOT NULL default 'pdf'",
         ],
 
@@ -64,10 +54,6 @@ $GLOBALS['TL_DCA']['tl_search_files'] = [
             'sql' => "mediumtext NULL",
         ],
 
-        'uuid' => [
-            'sql' => "binary(16) NULL",
-        ],
-
         /*
          * md5(url + filemtime)
          * → erkennt Änderungen zuverlässig
@@ -78,7 +64,7 @@ $GLOBALS['TL_DCA']['tl_search_files'] = [
 
         /*
          * Herkunftsseite (tl_page.id)
-         * → optional, Debug / Referenz
+         * → Cleanup / Referenz
          */
         'page_id' => [
             'sql' => "int(10) unsigned NOT NULL default 0",

src/Resources/contao/dca/tl_settings.php

@@ -4,11 +4,8 @@ use Contao\CoreBundle\DataContainer\PaletteManipulator;
 use Contao\System;
 
 /**
- * -------------------------------------------------
  * Fields
- * -------------------------------------------------
  */
-
 $GLOBALS['TL_DCA']['tl_settings']['fields']['meilisearch_host'] = [
     'inputType' => 'text',
     'eval' => [
@@ -62,9 +59,17 @@ $GLOBALS['TL_DCA']['tl_settings']['fields']['meilisearch_imagesize'] = [
         'chosen' => true,
         'includeBlankOption' => true,
     ],
+    // 🔥 DAS HAT GEFEHLT
     'sql' => "int(10) unsigned NOT NULL default 0",
 ];
 
+$GLOBALS['TL_DCA']['tl_settings']['fields']['meilisearch_index_past_events'] = [
+    'inputType' => 'checkbox',
+    'eval'      => [
+        'tl_class' => 'w50 clr',
+    ],
+];
+
 $GLOBALS['TL_DCA']['tl_settings']['fields']['meilisearch_fallback_image'] = [
     'inputType' => 'fileTree',
     'eval' => [
@@ -75,54 +80,25 @@ $GLOBALS['TL_DCA']['tl_settings']['fields']['meilisearch_fallback_image'] = [
     'sql' => "varbinary(16) NULL",
 ];
 
-$GLOBALS['TL_DCA']['tl_settings']['fields']['meilisearch_index_past_events'] = [
-    'inputType' => 'checkbox',
-    'eval' => [
-        'tl_class' => 'w50 clr',
-    ],
-];
-
-/**
- * -------------------------------------------------
- * Datei-Indexierung (Tika)
- * -------------------------------------------------
- */
-
-$GLOBALS['TL_DCA']['tl_settings']['fields']['meilisearch_index_files'] = [
+$GLOBALS['TL_DCA']['tl_settings']['fields']['meilisearch_index_pdfs'] = [
+    'label'     => &$GLOBALS['TL_LANG']['tl_settings']['meilisearch_index_pdfs'],
     'inputType' => 'checkbox',
     'eval'      => [
         'tl_class' => 'w50',
-        'submitOnChange' => true,
     ],
+    'sql'       => "char(1) NOT NULL default '1'",
+];
+
+$GLOBALS['TL_DCA']['tl_settings']['fields']['meilisearch_index_office'] = [
+    'label'     => &$GLOBALS['TL_LANG']['tl_settings']['meilisearch_index_office'],
+    'inputType' => 'checkbox',
+    'eval'      => ['tl_class' => 'w50'],
     'sql'       => "char(1) NOT NULL default '0'",
 ];
 
-$GLOBALS['TL_DCA']['tl_settings']['fields']['meilisearch_tika_url'] = [
-    'inputType' => 'text',
-    'eval' => [
-        'rgxp'      => 'url',
-        'mandatory' => true,
-        'tl_class'  => 'w50 clr',
-    ],
-];
-
 /**
- * -------------------------------------------------
- * Selector / Subpalette
- * -------------------------------------------------
- */
-
-$GLOBALS['TL_DCA']['tl_settings']['palettes']['__selector__'][] = 'meilisearch_index_files';
-
-$GLOBALS['TL_DCA']['tl_settings']['subpalettes']['meilisearch_index_files']
-    = 'meilisearch_tika_url';
-
-/**
- * -------------------------------------------------
  * Palette
- * -------------------------------------------------
  */
-
 PaletteManipulator::create()
     ->addLegend('meilisearch_legend', null, PaletteManipulator::POSITION_AFTER, true)
     ->addField('meilisearch_host', 'meilisearch_legend')
@@ -132,5 +108,6 @@ PaletteManipulator::create()
     ->addField('meilisearch_imagesize', 'meilisearch_legend')
     ->addField('meilisearch_fallback_image', 'meilisearch_legend')
     ->addField('meilisearch_index_past_events', 'meilisearch_legend')
-    ->addField('meilisearch_index_files', 'meilisearch_legend')
+    ->addField('meilisearch_index_pdfs', 'meilisearch_legend')
+    ->addField('meilisearch_index_office', 'meilisearch_legend')
     ->applyToPalette('default', 'tl_settings');

src/Resources/contao/languages/de/tl_settings.php

@@ -28,10 +28,10 @@ $GLOBALS['TL_LANG']['tl_settings']['meilisearch_index_past_events'][0]
 $GLOBALS['TL_LANG']['tl_settings']['meilisearch_index_past_events'][1]
     = 'Vergangene Kalender-Events werden ebenfalls in Meilisearch indexiert.';
 
-$GLOBALS['TL_LANG']['tl_settings']['meilisearch_index_files'] = [
-    'Dateien indexieren',
-    'Aktiviert die Indexierung von PDF-Dateien sowie DOCX, XLSX und PPTX.',
+$GLOBALS['TL_LANG']['tl_settings']['meilisearch_index_pdfs'] = [
+    'PDFs indexieren',
+    'Aktiviert die Indexierung von PDF-Dateien für die Suche.',
 ];
 
-$GLOBALS['TL_LANG']['tl_settings']['meilisearch_tika_url']
-    = ['Apache Tika URL', 'URL der Apache Tika Instanz (z. B. https://tika.domain.tld).'];
+$GLOBALS['TL_LANG']['tl_settings']['meilisearch_index_office']
+    = ['Office-Dateien indexieren', 'DOCX, XLSX und PPTX in die Suche aufnehmen.'];

src/Resources/contao/templates/frontend_module/mod_meilisearch_search.html.twig

@@ -4,7 +4,6 @@ Contao 5 – Frontend Module Template
 #}
 
 <!-- indexer::stop -->
-{% block meilisearch %}
 <div
         id="topsearch"
         class="meilisearch-search"
@@ -60,44 +59,10 @@ Contao 5 – Frontend Module Template
     </div>
 </div>
 
-<script>
-    (function () {
-        const CDN_URLS = [
-            'https://cdn.jsdelivr.net/npm/meilisearch@0.39.0/dist/bundles/meilisearch.umd.min.js',
-            'https://unpkg.com/meilisearch@0.39.0/dist/bundles/meilisearch.umd.min.js'
-        ];
+<script type="module">
+    import MeiliSearch from 'https://cdn.jsdelivr.net/npm/meilisearch@latest/dist/bundles/meilisearch.esm.js';
 
-        function loadClient(urls, onDone) {
-            if (typeof MeiliSearch !== 'undefined') {
-                onDone(true, null);
-                return;
-            }
-
-            if (!urls.length) {
-                onDone(false, 'Alle CDN-Quellen fehlgeschlagen (mögliche CSP-Blockierung von script-src).');
-                return;
-            }
-
-            const url = urls.shift();
-            const script = document.createElement('script');
-            script.src = url;
-            script.async = true;
-            script.crossOrigin = 'anonymous';
-
-            script.onload = () => {
-                if (typeof MeiliSearch !== 'undefined') {
-                    onDone(true, null);
-                } else {
-                    loadClient(urls, onDone);
-                }
-            };
-
-            script.onerror = () => loadClient(urls, onDone);
-
-            document.head.appendChild(script);
-        }
-
-        function initSearch() {
+    document.addEventListener('DOMContentLoaded', () => {
 
         const wrapper  = document.querySelector('.meilisearch-search');
         if (!wrapper) return;
@@ -130,10 +95,6 @@ Contao 5 – Frontend Module Template
             input.value = '';
             results.innerHTML = '';
             clear.classList.add('is-hidden');
-
-            // ✅ WICHTIG: Suchmodus verlassen
-            document.body.classList.remove('search-active');
-
             input.focus();
         });
 
@@ -255,19 +216,6 @@ Contao 5 – Frontend Module Template
                 results.appendChild(node);
             }
         }
-        }
-
-        document.addEventListener('DOMContentLoaded', () => {
-            loadClient([...CDN_URLS], (ok, reason) => {
-                if (!ok) {
-                    console.error('[Meilisearch] Browser client konnte nicht geladen werden. ' + reason);
-                    return;
-                }
-
-                initSearch();
     });
-        });
-    })();
 </script>
-{% endblock %}
 <!-- indexer::continue -->

src/Service/MeilisearchFileHelper.php

@@ -1,259 +0,0 @@
-<?php
-
-namespace Mummert\ContaoMeilisearchBundle\Service;
-
-use Contao\FilesModel;
-use Contao\StringUtil;
-use Contao\System;
-use Doctrine\DBAL\Connection;
-
-class MeilisearchFileHelper
-{
-    public function __construct(
-        private readonly Connection $connection,
-    ) {
-    }
-
-    /**
-     * Zentrale Datei-Verarbeitung
-     */
-    public function collect(string $url, string $type, int $pageId): void
-    {
-        $this->log('collect() start', [
-            'url'    => $url,
-            'type'   => $type,
-            'pageId' => $pageId,
-        ]);
-
-        // -------------------------------------------------
-        // 1. URL normalisieren
-        // -------------------------------------------------
-        $cleanUrl = strtok($url, '#');
-        $parts    = parse_url($cleanUrl);
-
-        if (!$parts) {
-            $this->log('Invalid URL, skip');
-            return;
-        }
-
-        // -------------------------------------------------
-        // 2. Externe Datei? → skip
-        // -------------------------------------------------
-        if (!empty($parts['host'])) {
-            $currentRequest = System::getContainer()
-                ->get('request_stack')
-                ->getCurrentRequest();
-
-            $pageHost = $currentRequest
-                ? parse_url($currentRequest->getSchemeAndHttpHost(), PHP_URL_HOST)
-                : null;
-
-            if ($pageHost && $parts['host'] !== $pageHost) {
-                $this->log('External file detected, skip', [
-                    'host' => $parts['host'],
-                ]);
-                return;
-            }
-        }
-
-        // -------------------------------------------------
-        // 3. Pfad-Kandidaten sammeln (ohne Annahmen!)
-        // -------------------------------------------------
-        $query = [];
-        if (!empty($parts['query'])) {
-            parse_str($parts['query'], $query);
-        }
-
-        $pathCandidates = [];
-
-        // direkter Pfad
-        if (!empty($parts['path'])) {
-            $pathCandidates[] = $parts['path'];
-        }
-
-        // Download-Parameter
-        foreach (['file', 'f', 'p'] as $param) {
-            if (!empty($query[$param])) {
-                $pathCandidates[] = $query[$param];
-            }
-        }
-
-        // normalisieren
-        $pathCandidates = array_values(array_unique(array_filter(array_map(
-            static function ($candidate) {
-                $candidate = rawurldecode(html_entity_decode((string) $candidate, ENT_QUOTES));
-                return ltrim($candidate, '/') ?: null;
-            },
-            $pathCandidates
-        ))));
-
-        $this->log('Path candidates (normalized)', [
-            'candidates' => $pathCandidates,
-        ]);
-
-        // -------------------------------------------------
-        // 4. FilesModel (DBAFS) auflösen → UUID
-        // -------------------------------------------------
-        $fileModel = null;
-
-        foreach ($pathCandidates as $candidate) {
-
-            // 1) direkt
-            $model = FilesModel::findByPath($candidate);
-            if ($model && $model->uuid) {
-                $fileModel = $model;
-                $this->log('Resolved via FilesModel (direct)', [
-                    'candidate' => $candidate,
-                    'path'      => $model->path,
-                ]);
-                break;
-            }
-
-            // 2) fallback: files/ davor
-            if (!str_starts_with($candidate, 'files/')) {
-                $model = FilesModel::findByPath('files/' . $candidate);
-                if ($model && $model->uuid) {
-                    $fileModel = $model;
-                    $this->log('Resolved via FilesModel (files/ prefix)', [
-                        'candidate' => $candidate,
-                        'path'      => $model->path,
-                    ]);
-                    break;
-                }
-            }
-        }
-
-        if (!$fileModel) {
-            $this->log('No Contao file model found, skip', [
-                'candidates' => $pathCandidates,
-            ]);
-            return;
-        }
-
-        $normalizedPath = (string) $fileModel->path;
-        $uuidBin        = $fileModel->uuid;
-        $uuid           = StringUtil::binToUuid($uuidBin);
-        $canonicalUrl   = '/' . ltrim($normalizedPath, '/');
-
-        $this->log('UUID resolved', [
-            'path' => $canonicalUrl,
-            'uuid' => $uuid,
-        ]);
-
-        // -------------------------------------------------
-        // 5. Datei im Filesystem prüfen
-        // -------------------------------------------------
-        $projectDir = System::getContainer()->getParameter('kernel.project_dir');
-        $abs        = $projectDir . '/public/' . $normalizedPath;
-
-        if (!is_file($abs)) {
-            $this->log('Resolved model but file missing on filesystem, skip', [
-                'path' => $normalizedPath,
-                'abs'  => $abs,
-            ]);
-            return;
-        }
-
-// -------------------------------------------------
-// 6. Redaktionellen Titel aus tl_files.meta
-// -------------------------------------------------
-        $title = null;
-        $meta  = StringUtil::deserialize($fileModel->meta, true);
-
-// 1) bevorzugte Sprache (falls vorhanden)
-        $lang = $GLOBALS['TL_LANGUAGE'] ?? null;
-        if ($lang && !empty($meta[$lang]['title'])) {
-            $title = trim((string) $meta[$lang]['title']);
-        }
-
-// 2) Fallback: erste verfügbare Sprache
-        if ($title === null && is_array($meta)) {
-            foreach ($meta as $langKey => $langMeta) {
-                if (!empty($langMeta['title'])) {
-                    $title = trim((string) $langMeta['title']);
-                    break;
-                }
-            }
-        }
-
-        if ($title) {
-            $this->log('Title resolved from tl_files', [
-                'title' => $title,
-            ]);
-        }
-
-        // -------------------------------------------------
-        // 7. Datei-Infos
-        // -------------------------------------------------
-        $mtime    = filemtime($abs) ?: 0;
-        $checksum = md5($normalizedPath . '|' . $mtime);
-        $now      = time();
-
-        // -------------------------------------------------
-        // 8. Upsert über UUID
-        // -------------------------------------------------
-        $existing = $this->connection->fetchAssociative(
-            'SELECT id FROM tl_search_files WHERE uuid = ?',
-            [$uuidBin]
-        );
-
-        if ($existing) {
-            $data = [
-                'tstamp'     => $now,
-                'last_seen'  => $now,
-                'type'       => $type,
-                'url'        => $canonicalUrl,
-                'page_id'    => $pageId,
-                'file_mtime' => $mtime,
-                'checksum'   => $checksum,
-            ];
-
-            if ($title !== null) {
-                $data['title'] = $title;
-            }
-
-            $this->connection->update(
-                'tl_search_files',
-                $data,
-                ['id' => $existing['id']]
-            );
-
-            $this->log('File updated by UUID', [
-                'uuid' => $uuid,
-            ]);
-        } else {
-            $this->connection->insert(
-                'tl_search_files',
-                [
-                    'tstamp'     => $now,
-                    'last_seen'  => $now,
-                    'type'       => $type,
-                    'url'        => $canonicalUrl,
-                    'title'      => $title ?? basename($normalizedPath),
-                    'page_id'    => $pageId,
-                    'file_mtime' => $mtime,
-                    'checksum'   => $checksum,
-                    'uuid'       => $uuidBin,
-                ]
-            );
-
-            $this->log('File inserted by UUID', [
-                'uuid' => $uuid,
-            ]);
-        }
-
-        $this->log('collect() end');
-    }
-
-    // -------------------------------------------------
-    // Logging
-    // -------------------------------------------------
-    private function log(string $message, array $context = []): void
-    {
-        $ctx = $context
-            ? ' | ' . json_encode($context, JSON_UNESCAPED_SLASHES | JSON_UNESCAPED_UNICODE)
-            : '';
-
-        error_log('[ContaoMeilisearch][MeilisearchFileHelper] ' . $message . $ctx);
-    }
-}

src/Service/MeilisearchImageHelper.php

@@ -1,6 +1,6 @@
 <?php
 
-namespace Mummert\ContaoMeilisearchBundle\Service;
+namespace MummertMedia\ContaoMeilisearchBundle\Service;
 
 use Contao\Config;
 use Contao\CoreBundle\Framework\ContaoFramework;

src/Service/MeilisearchIndexService.php

@@ -1,6 +1,6 @@
 <?php
 
-namespace Mummert\ContaoMeilisearchBundle\Service;
+namespace MummertMedia\ContaoMeilisearchBundle\Service;
 
 use Contao\Config;
 use Contao\CoreBundle\Framework\ContaoFramework;
@@ -72,7 +72,7 @@ class MeilisearchIndexService
         }
 
         $this->indexTlSearch($index);
-        $this->indexTlSearchFiles($index);
+        $this->indexTlSearchPdf($index);
     }
 
     private function ensureIndexSettings(Indexes $index): void
@@ -80,7 +80,6 @@ class MeilisearchIndexService
         $index->updateSettings([
             'searchableAttributes' => ['title', 'keywords', 'text'],
             'sortableAttributes'   => ['priority', 'startDate'],
-            'filterableAttributes' => ['type', 'filetype'],
         ]);
     }
 
@@ -133,7 +132,7 @@ class MeilisearchIndexService
     }
 
     /**
-     * tl_search indexieren (Seiten / News / Events)
+     * tl_search indexieren
      */
     private function indexTlSearch(Indexes $index): void
     {
@@ -165,11 +164,13 @@ class MeilisearchIndexService
                     }
                 }
 
+                $cleanText = $this->stripMeilisearchMeta((string) $row['text']);
+
                 $doc = [
                     'id'        => $type . '_' . $row['id'],
                     'type'      => $type,
                     'title'     => $row['title'],
-                    'text'      => $this->stripMeilisearchMeta((string) $row['text']),
+                    'text'      => $cleanText,
                     'url'       => $row['url'],
                     'protected' => (bool) $row['protected'],
                     'checksum'  => $row['checksum'],
@@ -191,24 +192,31 @@ class MeilisearchIndexService
                 $documents[] = $doc;
 
             } catch (\Throwable $e) {
-                error_log('[ContaoMeilisearch] Failed to build tl_search document: ' . $e->getMessage());
+                error_log(
+                    '[ContaoMeilisearch] Failed to build document for tl_search ID '
+                    . ($row['id'] ?? '?') . ': ' . $e->getMessage()
+                );
             }
         }
 
         if ($documents !== []) {
+            try {
                 $index->addDocuments($documents);
+            } catch (\Throwable $e) {
+                error_log('[ContaoMeilisearch] Failed to add tl_search documents: ' . $e->getMessage());
+            }
         }
     }
 
     /**
-     * tl_search_files indexieren (PDF / Office)
+     * tl_search_pdf indexieren
      */
-    private function indexTlSearchFiles(Indexes $index): void
+    private function indexTlSearchPdf(Indexes $index): void
     {
         try {
-            $rows = $this->connection->fetchAllAssociative('SELECT * FROM tl_search_files');
+            $rows = $this->connection->fetchAllAssociative('SELECT * FROM tl_search_pdf');
         } catch (\Throwable $e) {
-            error_log('[ContaoMeilisearch] Failed to read tl_search_files: ' . $e->getMessage());
+            error_log('[ContaoMeilisearch] Failed to read tl_search_pdf: ' . $e->getMessage());
             return;
         }
 
@@ -225,11 +233,10 @@ class MeilisearchIndexService
                     : 'pdf';
 
                 $documents[] = [
-                    'id'        => 'file_' . $row['id'],
-                    'type'      => 'file',
-                    'filetype'  => $fileType,
-                    'title'     => $row['title'] ?: basename($row['url']),
-                    'text'      => (string) $row['text'],
+                    'id'       => $fileType . '_' . $row['id'],
+                    'type'     => $fileType,
+                    'title'    => $row['title'],
+                    'text'     => $this->stripMeilisearchMeta((string) $row['text']),
                     'url'      => $row['url'],
                     'checksum' => $row['checksum'],
                     'poster'   => self::FILETYPE_ICON_MAP[$fileType]
@@ -237,12 +244,19 @@ class MeilisearchIndexService
                 ];
 
             } catch (\Throwable $e) {
-                error_log('[ContaoMeilisearch] Failed to build file document: ' . $e->getMessage());
+                error_log(
+                    '[ContaoMeilisearch] Failed to build PDF document for ID '
+                    . ($row['id'] ?? '?') . ': ' . $e->getMessage()
+                );
             }
         }
 
         if ($documents !== []) {
+            try {
                 $index->addDocuments($documents);
+            } catch (\Throwable $e) {
+                error_log('[ContaoMeilisearch] Failed to add tl_search_pdf documents: ' . $e->getMessage());
+            }
         }
     }
 

src/Service/OfficeIndexService.php

@@ -0,0 +1,273 @@
+<?php
+
+namespace MummertMedia\ContaoMeilisearchBundle\Service;
+
+use Contao\Database;
+use PhpOffice\PhpWord\IOFactory as WordIOFactory;
+use PhpOffice\PhpSpreadsheet\IOFactory as SpreadsheetIOFactory;
+use PhpOffice\PhpPresentation\IOFactory as PresentationIOFactory;
+use Symfony\Component\DependencyInjection\ParameterBag\ParameterBagInterface;
+
+class OfficeIndexService
+{
+    private string $projectDir;
+
+    // pro Crawl-Durchlauf: doppelte Verarbeitung vermeiden
+    private array $seenThisCrawl = [];
+
+    public function __construct(ParameterBagInterface $params)
+    {
+        $this->projectDir = rtrim((string) $params->get('kernel.project_dir'), '/');
+    }
+
+    /**
+     * @param array<int,array{url:string,linkText:?string}> $officeLinks
+     */
+    public function handleOfficeLinks(array $officeLinks): void
+    {
+        foreach ($officeLinks as $row) {
+            $url = (string) ($row['url'] ?? '');
+            $linkText = $row['linkText'] ?? null;
+
+            if ($url === '') {
+                continue;
+            }
+
+            try {
+                // innerhalb des Crawls gleiche URL nicht mehrfach parsen
+                $seenKey = md5($url);
+                if (isset($this->seenThisCrawl[$seenKey])) {
+                    continue;
+                }
+                $this->seenThisCrawl[$seenKey] = true;
+
+                $normalized = $this->normalizeOfficeUrl($url);
+                if ($normalized === null) {
+                    continue;
+                }
+
+                [$relativePath, $type] = $normalized;
+
+                $absolutePath = $this->getAbsolutePath($relativePath);
+                if (!is_file($absolutePath)) {
+                    continue;
+                }
+
+                $mtime = (int) (filemtime($absolutePath) ?: 0);
+                $checksum = md5($relativePath . '|' . $mtime);
+
+                $title = $linkText ?: basename($absolutePath);
+
+                $text = $this->parseOfficeFile($absolutePath, $type);
+                if ($text === '') {
+                    continue;
+                }
+
+                $this->upsertOffice(
+                    $relativePath,
+                    $title,
+                    $text,
+                    $checksum,
+                    $mtime,
+                    $type
+                );
+
+            } catch (\Throwable $e) {
+                error_log(
+                    '[ContaoMeilisearch] Office indexing failed for "' . $url . '": ' . $e->getMessage()
+                );
+            }
+        }
+    }
+
+    /**
+     * @return array{string,string}|null [relativePath, type]
+     */
+    private function normalizeOfficeUrl(string $url): ?array
+    {
+        $decoded = html_entity_decode($url);
+        $parts = parse_url($decoded);
+
+        // 1) files/... (ohne führenden Slash)
+        if (!empty($parts['path']) && str_starts_with($parts['path'], 'files/')) {
+            $ext = strtolower(pathinfo($parts['path'], PATHINFO_EXTENSION));
+            if (in_array($ext, ['docx', 'xlsx', 'pptx'], true)) {
+                return ['/' . $parts['path'], $ext];
+            }
+        }
+
+        // 2) /files/...
+        if (!empty($parts['path']) && str_starts_with($parts['path'], '/files/')) {
+            $ext = strtolower(pathinfo($parts['path'], PATHINFO_EXTENSION));
+            if (in_array($ext, ['docx', 'xlsx', 'pptx'], true)) {
+                return [$parts['path'], $ext];
+            }
+        }
+
+        if (empty($parts['query'])) {
+            return null;
+        }
+
+        parse_str($parts['query'], $query);
+
+        // 3) Contao 4: ?file=files/...
+        if (!empty($query['file'])) {
+            $file = urldecode((string) $query['file']);
+            $file = ltrim($file, '/');
+            $ext = strtolower(pathinfo($file, PATHINFO_EXTENSION));
+
+            if (
+                str_starts_with($file, 'files/')
+                && in_array($ext, ['docx', 'xlsx', 'pptx'], true)
+            ) {
+                return ['/' . $file, $ext];
+            }
+        }
+
+        // 4) Contao 5: ?p=...
+        if (!empty($query['p'])) {
+            $p = urldecode((string) $query['p']);
+            $ext = strtolower(pathinfo($p, PATHINFO_EXTENSION));
+
+            if (in_array($ext, ['docx', 'xlsx', 'pptx'], true)) {
+                return ['/files/' . ltrim($p, '/'), $ext];
+            }
+        }
+
+        return null;
+    }
+
+    private function getAbsolutePath(string $relativePath): string
+    {
+        return $this->projectDir . '/' . ltrim($relativePath, '/');
+    }
+
+    private function upsertOffice(
+        string $url,
+        string $title,
+        string $text,
+        string $checksum,
+        int $mtime,
+        string $type
+    ): void {
+        try {
+            Database::getInstance()
+                ->prepare('
+                    INSERT INTO tl_search_pdf
+                        (tstamp, type, url, title, text, checksum, file_mtime)
+                    VALUES
+                        (?, ?, ?, ?, ?, ?, ?)
+                    ON DUPLICATE KEY UPDATE
+                        tstamp=VALUES(tstamp),
+                        type=VALUES(type),
+                        url=VALUES(url),
+                        title=VALUES(title),
+                        text=VALUES(text),
+                        file_mtime=VALUES(file_mtime)
+                ')
+                ->execute(
+                    time(),
+                    $type,
+                    $url,
+                    $title,
+                    $text,
+                    $checksum,
+                    $mtime
+                );
+        } catch (\Throwable $e) {
+            error_log(
+                '[ContaoMeilisearch] Failed to write Office index entry (' . $url . '): ' . $e->getMessage()
+            );
+        }
+    }
+
+    private function parseOfficeFile(string $absolutePath, string $type): string
+    {
+        return match ($type) {
+            'docx' => $this->parseDocx($absolutePath),
+            'xlsx' => $this->parseXlsx($absolutePath),
+            'pptx' => $this->parsePptx($absolutePath),
+            default => '',
+        };
+    }
+
+    private function parseDocx(string $absolutePath): string
+    {
+        try {
+            $phpWord = WordIOFactory::load($absolutePath);
+            $text = '';
+
+            foreach ($phpWord->getSections() as $section) {
+                foreach ($section->getElements() as $element) {
+                    if (method_exists($element, 'getText')) {
+                        $text .= ' ' . $element->getText();
+                    }
+                }
+            }
+
+            return $this->cleanText($text);
+        } catch (\Throwable $e) {
+            error_log(
+                '[ContaoMeilisearch] Failed to parse DOCX "' . $absolutePath . '": ' . $e->getMessage()
+            );
+            return '';
+        }
+    }
+
+    private function parseXlsx(string $absolutePath): string
+    {
+        try {
+            $spreadsheet = SpreadsheetIOFactory::load($absolutePath);
+            $text = '';
+
+            foreach ($spreadsheet->getAllSheets() as $sheet) {
+                foreach ($sheet->toArray() as $row) {
+                    $text .= ' ' . implode(' ', array_filter($row, 'is_scalar'));
+                }
+            }
+
+            return $this->cleanText($text);
+        } catch (\Throwable $e) {
+            error_log(
+                '[ContaoMeilisearch] Failed to parse XLSX "' . $absolutePath . '": ' . $e->getMessage()
+            );
+            return '';
+        }
+    }
+
+    private function parsePptx(string $absolutePath): string
+    {
+        try {
+            $presentation = PresentationIOFactory::load($absolutePath);
+            $text = '';
+
+            foreach ($presentation->getAllSlides() as $slide) {
+                foreach ($slide->getShapeCollection() as $shape) {
+                    if (method_exists($shape, 'getPlainText')) {
+                        $text .= ' ' . $shape->getPlainText();
+                    }
+                }
+            }
+
+            return $this->cleanText($text);
+        } catch (\Throwable $e) {
+            error_log(
+                '[ContaoMeilisearch] Failed to parse PPTX "' . $absolutePath . '": ' . $e->getMessage()
+            );
+            return '';
+        }
+    }
+
+    private function cleanText(string $text): string
+    {
+        if (class_exists(\Normalizer::class)) {
+            $text = \Normalizer::normalize($text, \Normalizer::FORM_C) ?? $text;
+        }
+
+        $text = str_replace(["\r\n", "\r"], "\n", $text);
+        $text = preg_replace('/[^\p{L}\p{N}\p{P}\p{Z}\n]/u', ' ', $text);
+        $text = preg_replace('/\s+/u', ' ', $text);
+
+        return trim(mb_substr($text, 0, 20000));
+    }
+}

src/Service/PdfIndexService.php

@@ -0,0 +1,224 @@
+<?php
+
+namespace MummertMedia\ContaoMeilisearchBundle\Service;
+
+use Contao\Database;
+use Smalot\PdfParser\Parser;
+use Symfony\Component\DependencyInjection\ParameterBag\ParameterBagInterface;
+
+class PdfIndexService
+{
+    private string $projectDir;
+
+    private bool $didReset = false;
+    private array $seenThisCrawl = [];
+
+    public function __construct(ParameterBagInterface $params)
+    {
+        $this->projectDir = rtrim((string) $params->get('kernel.project_dir'), '/');
+    }
+
+    /**
+     * Wird aus dem Listener beim ersten Hook-Call pro Crawl aufgerufen.
+     */
+    public function resetTableOnce(): void
+    {
+        if ($this->didReset) {
+            return;
+        }
+
+        $this->didReset = true;
+        $this->seenThisCrawl = [];
+
+        Database::getInstance()->execute('TRUNCATE tl_search_pdf');
+    }
+
+    /**
+     * @param array<int,array{url:string,linkText:?string}> $pdfLinks
+     */
+    public function handlePdfLinks(array $pdfLinks): void
+    {
+        foreach ($pdfLinks as $row) {
+            $url = (string) ($row['url'] ?? '');
+            $linkText = $row['linkText'] ?? null;
+
+            if ($url === '') {
+                continue;
+            }
+
+            // innerhalb eines Crawls doppelte URLs vermeiden
+            $seenKey = md5($url);
+            if (isset($this->seenThisCrawl[$seenKey])) {
+                continue;
+            }
+            $this->seenThisCrawl[$seenKey] = true;
+
+            $normalizedPath = $this->normalizePdfUrl($url);
+            if ($normalizedPath === null) {
+                continue;
+            }
+
+            $absolutePath = $this->getAbsolutePath($normalizedPath);
+            if (!is_file($absolutePath)) {
+                continue;
+            }
+
+            $mtime = (int) (filemtime($absolutePath) ?: 0);
+            $checksum = md5($normalizedPath . '|' . $mtime);
+
+            // Titel-Priorität:
+            // 1) Linktext
+            // 2) PDF-Metadaten
+            // 3) Dateiname
+            $pdfMetaTitle = $this->readPdfMetaTitle($absolutePath);
+            $title = $linkText ?: ($pdfMetaTitle ?: basename($absolutePath));
+
+            $text = $this->parsePdf($absolutePath);
+            if ($text === '') {
+                continue;
+            }
+
+            $this->upsertPdf(
+                $normalizedPath,
+                $title,
+                $text,
+                $checksum,
+                $mtime
+            );
+        }
+    }
+
+    private function normalizePdfUrl(string $url): ?string
+    {
+        $decoded = html_entity_decode($url);
+        $parts = parse_url($decoded);
+
+        // 1) files/...pdf (ohne führenden Slash)
+        if (
+            !empty($parts['path'])
+            && str_starts_with($parts['path'], 'files/')
+            && str_ends_with(strtolower($parts['path']), '.pdf')
+        ) {
+            return '/' . $parts['path'];
+        }
+
+        // 2) /files/...pdf
+        if (
+            !empty($parts['path'])
+            && str_starts_with($parts['path'], '/files/')
+            && str_ends_with(strtolower($parts['path']), '.pdf')
+        ) {
+            return $parts['path'];
+        }
+
+        if (empty($parts['query'])) {
+            return null;
+        }
+
+        parse_str($parts['query'], $query);
+
+        // 3) Contao 4: ?file=files/...
+        if (!empty($query['file'])) {
+            $file = urldecode((string) $query['file']);
+            $file = ltrim($file, '/');
+
+            if (
+                str_starts_with($file, 'files/')
+                && str_ends_with(strtolower($file), '.pdf')
+            ) {
+                return '/' . $file;
+            }
+        }
+
+        // 4) Contao 5: ?p=...
+        if (!empty($query['p'])) {
+            $p = urldecode((string) $query['p']);
+            return '/files/' . ltrim($p, '/');
+        }
+
+        return null;
+    }
+
+    private function getAbsolutePath(string $relativePath): string
+    {
+        return $this->projectDir . '/' . ltrim($relativePath, '/');
+    }
+
+    private function upsertPdf(
+        string $url,
+        string $title,
+        string $text,
+        string $checksum,
+        int $mtime
+    ): void {
+        Database::getInstance()
+            ->prepare('
+                INSERT INTO tl_search_pdf
+                    (tstamp, url, title, text, checksum, file_mtime)
+                VALUES
+                    (?, ?, ?, ?, ?, ?)
+                ON DUPLICATE KEY UPDATE
+                    tstamp=VALUES(tstamp),
+                    url=VALUES(url),
+                    title=VALUES(title),
+                    text=VALUES(text),
+                    file_mtime=VALUES(file_mtime)
+            ')
+            ->execute(
+                time(),
+                $url,
+                $title,
+                $text,
+                $checksum,
+                $mtime
+            );
+    }
+
+    private function parsePdf(string $absolutePath): string
+    {
+        try {
+            $parser = new Parser();
+            $pdf = $parser->parseFile($absolutePath);
+            $text = $this->cleanPdfContent($pdf->getText());
+
+            return mb_substr($text, 0, 20000);
+        } catch (\Throwable) {
+            return '';
+        }
+    }
+
+    private function readPdfMetaTitle(string $absolutePath): ?string
+    {
+        try {
+            $parser = new Parser();
+            $pdf = $parser->parseFile($absolutePath);
+            $details = $pdf->getDetails();
+
+            foreach (['Title', 'title'] as $key) {
+                if (!empty($details[$key]) && is_string($details[$key])) {
+                    $t = trim($details[$key]);
+                    if ($t !== '') {
+                        return $t;
+                    }
+                }
+            }
+        } catch (\Throwable) {
+        }
+
+        return null;
+    }
+
+    private function cleanPdfContent(string $text): string
+    {
+        if (class_exists(\Normalizer::class)) {
+            $text = \Normalizer::normalize($text, \Normalizer::FORM_C) ?? $text;
+        }
+
+        $text = str_replace(["\r\n", "\r"], "\n", $text);
+        $text = preg_replace('/[^\p{L}\p{N}\p{P}\p{Z}\n]/u', ' ', $text);
+        $text = preg_replace('/(?<=\p{L})\s+(?=\p{L})/u', ' ', $text);
+        $text = preg_replace('/\s+/u', ' ', $text);
+
+        return trim($text);
+    }
+}