View Issue Details

IDProjectCategoryView StatusLast Update
0006491OXID eShop (all versions)1.01. Products (product, categories, manufacturer, promotions etc.)public2016-08-25 15:37
Status acknowledgedResolutionopen 
Product Version4.9.9 / 5.2.9 
Target VersionFixed in Version 
Summary0006491: Smarty Modifier 'smarty_modifier_truncate' is not UTF-8 compatible
Description5.2.9 has both modifierin package:
- smarty_modifier_oxtruncate
- smarty_modifier_truncate

and smarty_modifier_truncate seems to only for back-compatibility present.

But it still be a problem if in the template used old truncate modifier - it uses native substr() calls and not UTF-8 compatible. For example it breaks last symbol, if it's not ASCII ( ä, ö, ü, ß in German )
Steps To Reproducefunction smarty_modifier_truncate($string, $length = 80, $etc = '...',
                                  $break_words = false, $middle = false)
    if ($length == 0)
        return '';

    if (strlen($string) > $length) {
        $length -= min($length, strlen($etc));
        if (!$break_words && !$middle) {
            $string = preg_replace('/\s+?(\S+)?$/', '', substr($string, 0, $length+1));
        if(!$middle) {
            return substr($string, 0, $length) . $etc;
        } else {
            return substr($string, 0, $length/2) . $etc . substr($string, -$length/2);
    } else {
        return $string;
Additional InformationWhy not refactor it to Oxid string handler ( getStr()->substr() )?
TagsNo tags attached.
ThemeNot defined
BrowserNot defined
PHP VersionNot defined
Database VersionNot defined



2016-08-25 11:23

administrator   ~0011753

The bug is not reproducible. Kindly provide some information on where exactly this bug arises. A screenshot would also be helpful.


2016-08-25 14:14

reporter   ~0011754

If we truncating with modifier 'truncate' something like "Packung 200 Hüllen"|truncate:14

We getting "Packung 200 H?"

The reason why, may be the fact, that utf-8 symbols taking 2 bytes and substr() cuts second byte off...

This issue with substr() and utf-8 can be founded on stack overflow:

Correct would be mb_substr() which realised in oxStr class (calling handler oxstrmb)

As i said - it's works properly in oxtruncate, but truncate is still not utf-8 compatible, and i don't understand why.