[BUGFIX] Unicode characters vs. statistics feature

kitzberger · dkd-kaehm · commit 5df9deacffc6 · 2025-08-18T11:50:34.000+02:00
In case someone enters multi-byte characters into the search truncating that string requires mb_substr() instead of substr(). Resolves: #4072
diff --git a/Classes/Domain/Search/Statistics/StatisticsWriterProcessor.php b/Classes/Domain/Search/Statistics/StatisticsWriterProcessor.php
@@ -107,7 +107,7 @@ protected function getProcessedKeywords(
         $keywords = $query->getQuery();
         $keywords = $this->sanitizeString($keywords);
         // Ensure string does not exceed database field length
-        $keywords = substr($keywords, 0, 128);
+        $keywords = mb_substr($keywords, 0, 128, 'UTF-8');
         if ($lowerCaseQuery) {
             $keywords = mb_strtolower($keywords);
         }

Original file line number	Diff line number	Diff line change
`@@ -107,7 +107,7 @@ protected function getProcessedKeywords(`
`107`	`107`	`$keywords = $query->getQuery();`
`108`	`108`	`$keywords = $this->sanitizeString($keywords);`
`109`	`109`	`// Ensure string does not exceed database field length`
`110`		`- $keywords = substr($keywords, 0, 128);`
	`110`	`+ $keywords = mb_substr($keywords, 0, 128, 'UTF-8');`
`111`	`111`	`if ($lowerCaseQuery) {`
`112`	`112`	`$keywords = mb_strtolower($keywords);`
`113`	`113`	`}`