Standard string functions like strlen() count bytes, not characters. For UTF-8 text containing non-ASCII characters (accented letters, CJK, emoji), you need the mbstring extension.
mb_strlen() — character count.mb_strtolower() / mb_strtoupper() — Unicode-aware case conversion.mb_substr() — character-based slicing.mb_strpos() — character-based search.<?php
$str = 'Héllo Wörld'; // 11 characters, but >11 bytes in UTF-8
echo strlen($str); // e.g. 13 (bytes)
echo mb_strlen($str); // 11 (characters)
echo mb_strtoupper($str, 'UTF-8'); // HÉLLO WÖRLD
echo mb_strtolower($str, 'UTF-8'); // héllo wörld
echo mb_substr($str, 0, 5, 'UTF-8'); // Héllo
// Safe check for multibyte position
$pos = mb_strpos($str, 'Wö', 0, 'UTF-8');
echo $pos; // 6
// Convert encoding
$latin = mb_convert_encoding($str, 'ISO-8859-1', 'UTF-8');
// Set internal encoding once at bootstrap
mb_internal_encoding('UTF-8');
// After this, the encoding argument can be omitted
echo mb_strlen($str); // 11
Tip: Call mb_internal_encoding('UTF-8') once in your application bootstrap (or set mbstring.internal_encoding = UTF-8 in php.ini) so you never need to pass the encoding argument to every mb_* call.