<?php
$string = '#️⃣';
echo '=== UTF-8 =============================', PHP_EOL;
echo 'STRING: ', $string, PHP_EOL;
echo 'CHARACTERS: ', mb_strlen($string), PHP_EOL; // 3
echo 'BYTE COUNT: ', strlen($string), PHP_EOL; // 7
foreach(range(0, mb_strlen($string) - 1) as $index) {
echo PHP_EOL;
echo 'CHARACTER #', $index + 1, ' CODE POINT: ', '0x', dechex(mb_ord(mb_substr($string, $index, 1))), PHP_EOL; // [0x23, 0xfe0f, 0x20e3]
echo 'CHARACTER #', $index + 1, ' BYTE COUNT: ', mb_strlen(mb_substr($string, $index, 1), '8bit'), PHP_EOL; // [1, 3, 3]
}
echo '=======================================', PHP_EOL, PHP_EOL;
$string = mb_convert_encoding($string, 'UTF-16', 'UTF-8');
echo '=== UTF-16 ============================', PHP_EOL;
echo 'STRING: ', $string, PHP_EOL;
echo 'CHARACTERS: ', mb_strlen($string), PHP_EOL; // should be 3, is 6
echo 'BYTE COUNT: ', strlen($string), PHP_EOL; // should 6, is 6
foreach(range(0, mb_strlen($string) - 1) as $index) {
echo PHP_EOL;
echo 'CHARACTER #', $index + 1, ' CODE POINT: ', '0x', dechex(mb_ord(mb_substr($string, $index, 1))), PHP_EOL; // should be [0x23, 0xfe0f, 0x20e3], is [0x0, 0x23, 0x0, 0xf, 0x20, 0x0]; it actually gets kinda close here, if you recompose two by two bytes, but no cigar
echo 'CHARACTER #', $index + 1, ' BYTE COUNT: ', mb_strlen(mb_substr($string, $index, 1), '8bit'), PHP_EOL; // should be [2, 2, 2], is [1, 1, 1, 1, 1, 1]
}
echo '=======================================', PHP_EOL;
preferences:
56.54 ms | 402 KiB | 5 Q