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

Tika Title encoding

Browse Source
This commit is contained in:
Jürgen Mummert
2026-01-10 18:57:13 +01:00
parent 17da2a8434
commit 2d3ddac945
1 changed files with 16 additions and 9 deletions
Download Patch File Download Diff File Expand all files Collapse all files
src/Command/MeilisearchFilesParseCommand.php
+16 -9
View File
@@ -145,7 +145,7 @@ class MeilisearchFilesParseCommand extends Command
}
// -------------------------------------------------
// Tika BODY
// Tika BODY (roher Plaintext)
// -------------------------------------------------
try {
$this->log('Parsing file', ['url' => $normalized]);
@@ -210,19 +210,26 @@ class MeilisearchFilesParseCommand extends Command
}
// -------------------------------------------------
// TITLE FALLBACK
// TITLE → ASCII SAFE (DELIBERATE DATA LOSS)
// -------------------------------------------------
if (!$title) {
$title = pathinfo($normalized, PATHINFO_FILENAME);
$title = str_replace(['_', '-'], ' ', $title);
if ($title) {
// UTF-8 → ASCII, Unbekanntes verwerfen
$title = iconv('UTF-8', 'ASCII//TRANSLIT//IGNORE', $title);
// Normalisieren
$title = preg_replace('/\s+/', ' ', $title);
$title = trim($title);
}
// -------------------------------------------------
// 🔑 CRITICAL FIX: remove invalid UTF-8 bytes
// FALLBACK: Dateiname
// -------------------------------------------------
$title = iconv('UTF-8', 'UTF-8//IGNORE', $title);
$title = preg_replace('/\s+/u', ' ', $title);
$title = trim($title);
if (!$title || strlen($title) < 5) {
$title = pathinfo($normalized, PATHINFO_FILENAME);
$title = str_replace(['_', '-'], ' ', $title);
$title = preg_replace('/\s+/', ' ', $title);
$title = trim($title);
}
// -------------------------------------------------
// Store result