Get UTF-8 length of the text.

utf8_length.php

<?php
//
// Get UTF-8 length of the text.
// Return false when the text not UTF-8 encoding.
//
function utf8_length($text)
{
	$text_len = strlen($text);
	$utf8_len = 0;
	
	// UTF-8 format：
	// < 0x80 | 0xxxxxxx
	// < 0xE0 | 110xxxxx 10xxxxxx
	// < 0xF0 | 1110xxxx 10xxxxxx 10xxxxxx
	// < 0xF8 | 11110xxx 10xxxxxx 10xxxxxx 10xxxxxx
	// < 0xFC | 111110xx 10xxxxxx 10xxxxxx 10xxxxxx 10xxxxxx
	// < 0xFE | 1111110x 10xxxxxx 10xxxxxx 10xxxxxx 10xxxxxx 10xxxxxx
	
	$need_check = 0;
	$have_check = 0;
	
	for ($i = 0; $i < $text_len; $i ++) {
		$c = ord($text[$i]);
 
		if ($need_check > 0) {
			$c = ord($text[$i]);
			$c = ($c >> 6) << 6;
			
			$have_check ++;
			
			// 10xxxxxx ~ 10111111
			if ($c != 0x80) {
				$i -= $have_check;
				$need_check = 0;
				$have_check = 0;
				return false;
			}
			else if ($need_check == $have_check) {
				$need_check = 0;
				$have_check = 0;
				$utf8_len ++;
			}
			
			continue;
		}
		
		if ($c < 0x80)      // 0xxxxxxx
			$utf8_len ++;
		else if ($c < 0xE0) // 110xxxxx
			$need_check = 1;
		else if ($c < 0xF0) // 1110xxxx
			$need_check = 2;
		else if ($c < 0xF8) // 11110xxx
			$need_check = 3;
		else if ($c < 0xFC) // 111110xx
			$need_check = 4;
		else if ($c < 0xFE) // 1111110x
			$need_check = 5;
		else
			return false;
	}
	
	return $utf8_len;
}
?>