utf8_compose()

Complexity

Conditions	18
Paths	27

Size

Total Lines	72
Code Lines	36

Duplication

Lines	0
Ratio	0 %

Importance

Changes	2
Bugs	0	Features	0

<?php
function utf8_compose($string)
{
	global $sourcedir;

	$substitutions = utf8_compose_maps();
	$combining_classes = utf8_combining_classes();

	$string = preg_split('/(.)/su', $string, 0, PREG_SPLIT_DELIM_CAPTURE | PREG_SPLIT_NO_EMPTY);

	for ($c = 0; $c < count($string); $c++)

It seems like you are calling the size function count() as part of the test condition. You might want to compute the size beforehand, and not on each iteration.

	{
		// Singleton replacements.
		if (isset($substitutions[$string[$c]]))
			$string[$c] = $substitutions[$string[$c]];

		// Hangul characters.
		// See "Hangul Syllable Composition" in the Unicode standard, ch. 3.12.
		if ($string[$c] >= "\xE1\x84\x80" && $string[$c] <= "\xE1\x84\x92" && $string[$c + 1] >= "\xE1\x85\xA1" && $string[$c + 1] <= "\xE1\x85\xB5")
		{
			if (!function_exists('mb_ord'))
				require_once($sourcedir . '/Subs-Compat.php');

			$l_part = $string[$c];
			$v_part = $string[$c + 1];
			$t_part = null;

			$l_index = mb_ord($l_part) - 0x1100;
			$v_index = mb_ord($v_part) - 0x1161;

			$lv_index = $l_index * 588 + $v_index * 28;
			$s = 0xAC00 + $lv_index;

			if ($string[$c + 2] >= "\xE1\x86\xA8" && $string[$c + 2] <= "\xE1\x87\x82")
			{
				$t_part = $string[$c + 2];
				$t_index = mb_ord($t_part) - 0x11A7;
				$s += $t_index;
			}

			$string[$c] = mb_chr($s);
			$string[++$c] = null;

			if (isset($t_part))
				$string[++$c] = null;

			continue;
		}

		if ($c > 0)
		{
			$ccc = isset($combining_classes[$string[$c]]) ? $combining_classes[$string[$c]] : 0;

			// Find the preceding starter character.
			$l = $c - 1;
			while ($l > 0 && (!isset($string[$l]) || (!empty($combining_classes[$string[$l]]) && $combining_classes[$string[$l]] < $ccc)))
				$l--;

			// Is there a composed form for this combination?
			if (isset($substitutions[$string[$l] . $string[$c]]))
			{
				// Replace the starter character with the composed character.
				$string[$l] = $substitutions[$string[$l] . $string[$c]];

				// Unset the current combining character.
				$string[$c] = null;
			}
		}
	}

	$string = implode('', $string);

	return $string;
}
?>