mummert/meilisearch-bundle
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

Bugfix

Browse Source
This commit is contained in:
Jürgen Mummert
2025-12-23 13:45:10 +01:00
parent 4a09f8530a
commit 9fa59c4b1a
1 changed files with 86 additions and 52 deletions
Download Patch File Download Diff File Expand all files Collapse all files
src/EventListener/IndexPageListener.php
+86 -52
View File
@@ -2,6 +2,8 @@
namespace MummertMedia\ContaoMeilisearchBundle\EventListener;
use Contao\Database;
use Contao\StringUtil;
use Smalot\PdfParser\Parser;
class IndexPageListener
@@ -13,16 +15,15 @@ class IndexPageListener
return;
}
// JSON aus Kommentar extrahieren + parsen
// JSON aus Kommentar extrahieren
$parsed = $this->extractMeilisearchJson($content);
if ($parsed === null) {
return;
}
/*
* =====================
* PRIORITY (event > news > page)
* PRIORITY
* =====================
*/
$priority =
@@ -36,7 +37,7 @@ class IndexPageListener
/*
* =====================
* KEYWORDS (merge)
* KEYWORDS
* =====================
*/
$keywordSources = [
@@ -45,26 +46,26 @@ class IndexPageListener
$parsed['page']['keywords'] ?? null,
];
$kw = [];
foreach ($keywordSources as $s) {
if (!is_string($s) || trim($s) === '') {
$keywords = [];
foreach ($keywordSources as $src) {
if (!is_string($src) || trim($src) === '') {
continue;
}
foreach (preg_split('/\s+/', trim($s)) ?: [] as $p) {
if ($p !== '') {
$kw[] = $p;
foreach (preg_split('/\s+/', trim($src)) as $word) {
if ($word !== '') {
$keywords[] = $word;
}
}
}
if ($kw) {
$set['keywords'] = implode(' ', array_unique($kw));
if ($keywords) {
$set['keywords'] = implode(' ', array_unique($keywords));
}
/*
* =====================
* IMAGEPATH (event > news > page > custom)
* IMAGEPATH
* =====================
*/
$image =
@@ -79,7 +80,7 @@ class IndexPageListener
/*
* =====================
* STARTDATE (event/news)
* STARTDATE
* =====================
*/
$date =
@@ -95,20 +96,16 @@ class IndexPageListener
/*
* =====================
* PDF LINKS INDEXIEREN
* PDFS ALS EIGENE DOKUMENTE INDEXIEREN
* =====================
*/
$pdfText = $this->extractPdfTextFromContent($content);
if ($pdfText !== '') {
$set['text'] = trim(
($set['text'] ?? '') . "\n\n" . $pdfText
);
}
$this->indexPdfLinks($content);
}
/**
* Extrahiert das JSON aus <!-- MEILISEARCH_JSON {...} -->
* -------------------------------------
* JSON aus Kommentar extrahieren
* -------------------------------------
*/
private function extractMeilisearchJson(string $content): ?array
{
@@ -123,39 +120,36 @@ class IndexPageListener
}
/**
* Findet PDF-Links im HTML und extrahiert deren Text
* -------------------------------------
* PDF-Links finden und indexieren
* -------------------------------------
*/
private function extractPdfTextFromContent(string $content): string
private function indexPdfLinks(string $content): void
{
if (!preg_match_all(
'/<a\s+[^>]*href=["\']([^"\']+\.pdf[^"\']*)["\'][^>]*>/i',
'/<a\s+[^>]*href=["\']([^"\']+\.pdf[^"\']*)["\'][^>]*>(.*?)<\/a>/is',
$content,
$matches
)) {
return '';
return;
}
$texts = [];
foreach ($matches[1] as $url) {
$pdfText = $this->parsePdfFromUrl($url);
if ($pdfText !== '') {
$texts[] = $pdfText;
foreach ($matches[1] as $i => $url) {
$title = trim(strip_tags($matches[2][$i])) ?: basename($url);
$this->indexSinglePdf($url, $title);
}
}
return implode("\n\n", $texts);
}
/**
* Lädt und parsed ein PDF (nur /files/)
* -------------------------------------
* Einzelnes PDF indexieren
* -------------------------------------
*/
private function parsePdfFromUrl(string $url): string
private function indexSinglePdf(string $url, string $title): void
{
// Nur interne PDFs
// nur interne PDFs
if (!str_contains($url, '/files/')) {
return '';
return;
}
// relative URLs normalisieren
@@ -166,19 +160,62 @@ class IndexPageListener
. $url;
}
$checksum = md5($url);
$db = Database::getInstance();
// bereits indexiert?
$exists = $db
->prepare('SELECT id FROM tl_search WHERE checksum=?')
->execute($checksum);
if ($exists->numRows > 0) {
return;
}
$text = $this->parsePdf($url);
if ($text === '') {
return;
}
$db
->prepare('
INSERT INTO tl_search
(tstamp, title, url, text, checksum, protected, pid, type)
VALUES
(?, ?, ?, ?, ?, ?, ?, ?)
')
->execute(
time(),
StringUtil::decodeEntities($title),
$url,
$text,
$checksum,
'',
0,
'file'
);
}
/**
* -------------------------------------
* PDF parsen (Smalot)
* -------------------------------------
*/
private function parsePdf(string $url): string
{
try {
$pdfContent = @file_get_contents($url);
if (!$pdfContent) {
$content = @file_get_contents($url);
if (!$content) {
return '';
}
$parser = new Parser();
$pdf = $parser->parseContent($pdfContent);
$pdf = $parser->parseContent($content);
$text = $this->cleanPdfContent($pdf->getText());
// Begrenzung für Meilisearch
return mb_substr($text, 0, 2000);
// Begrenzen (Performance + Relevanz)
return mb_substr($text, 0, 5000);
} catch (\Throwable) {
return '';
@@ -186,17 +223,14 @@ class IndexPageListener
}
/**
* Bereinigt PDF-Text
* -------------------------------------
* PDF-Text bereinigen
* -------------------------------------
*/
private function cleanPdfContent(string $content): string
{
// UTF-8 normalisieren
$content = mb_convert_encoding($content, 'UTF-8', 'UTF-8');
// Steuerzeichen entfernen
$content = preg_replace('/[\x00-\x1F\x7F]/u', ' ', $content);
// Whitespaces normalisieren
$content = preg_replace('/\s+/u', ' ', $content);
return trim($content);