PHP: UTF-8 multi-byte string splitter?

Question

Can this work for splitting a multi-byte string every ten characters?

$string = 'Star Wars Episode Seven Sucked';    
mb_split('.', $string, 10);

The PHP manual says that str_split() works on bytes, not characters in multi-byte strings. That means mb_split() would seem to be a natural "overloaded" substitution, but the two functions (str_split() and mb_split()) have different function signatures and are not "overload buddies," so to speak. Then, I had a thought, what about this?

mb_internal_encoding("UTF-8");
$string = 'Star Wars Episode Seven Sucked';  
$tokens = [];

for($i = 0, $length = mb_strlen($string); $i < $length; $i += 10)
{
    $tokens[] = mb_substr($string, $i, 10, 'UTF-8');
}

print_r($tokens);

Are you searching for: http://stackoverflow.com/q/3825226/3933332 ? — Rizier123, Jan 01 '16 at 03:37
I'm confused why you're asking this instead of just test it yourself? Or reading [mb_split()](http://php.net/manual/en/function.mb-split.php) ? — devlin carnate, Jan 01 '16 at 03:39
@devlincarnate I see. I refer you to the definition of the Stackoverflow site. — Melinda, Jan 01 '16 at 03:52
Why should she? Maybe she's not somewhere she cannot code right now? — Anthony Rutledge, Jan 01 '16 at 03:57
@RyanVincent You need to read this. http://stackoverflow.com/help/on-topic — Melinda, Jan 01 '16 at 04:01
@I am on a train and I had a thought. Sorry if that is not good enough for some. — Melinda, Jan 01 '16 at 04:02

Melinda · Answer 1 · 2016-01-01T08:52:11.957

-1

function mb_utf8_split($string, $interval)
{
     $tokens = [];
     mb_internal_encoding('UTF-8');

     $stringEncoding = mb_detect_encoding($string, 'UTF-8, ISO-8859-1', true);

     if(!$stringEncoding)
     {
          throw new RuntimeException("Unable to identify character encoding.");
     }

     if($stringEncoding !== 'UTF-8')
     {
          $string = mb_convert_encoding($string, 'UTF-8', $stringEncoding);
     }

     for($i = 0, $length = mb_strlen($string); $i < $length; $i += $interval)
     {
         $tokens[] = mb_substr($string, $i, $interval, 'UTF-8');
     }

     return $tokens;
}

edited Jan 01 '16 at 08:52

answered Jan 01 '16 at 08:43

Melinda

76
8

This could probably be better. – Melinda Jan 01 '16 at 08:58
1

This is a good attempt. I'm not sure about the `mb_detect_encoding()` line's second parameter (`'UTF-8, ISO-8859-1'`), but it looks like a good start to me. – Anthony Rutledge Jan 01 '16 at 09:05

PHP: UTF-8 multi-byte string splitter?

1 Answers1