/**
* Take a series of space-separated hexadecimal numbers representing
* Unicode code points and return a UTF-8 string composed of those
* characters. Used by UTF-8 data generation and testing routines.
*
* @param string $sequence
* @return string
* @throws InvalidArgumentException if fed out of range data.
* @private
* @deprecated since 1.25, use UtfNormal\Utils directly
*/
function hexSequenceToUtf8($sequence)
{
return Utils::hexSequenceToUtf8($sequence);
}
$combiningClass[$source] = intval($canonicalCombiningClass);
}
if ($decompositionMapping === '') {
continue;
}
if (preg_match('/^<(.+)> (.*)$/', $decompositionMapping, $matches)) {
# Compatibility decomposition
$canonical = false;
$decompositionMapping = $matches[2];
$compat++;
} else {
$canonical = true;
$canon++;
}
$total++;
$dest = Utils::hexSequenceToUtf8($decompositionMapping);
$compatibilityDecomp[$source] = $dest;
if ($canonical) {
$canonicalDecomp[$source] = $dest;
if (empty($exclude[$source])) {
$canonicalComp[$dest] = $source;
}
}
# print "$codepoint | $canonicalCombiningClasses | $decompositionMapping\n";
}
fclose($in);
print "Recursively expanding canonical mappings...\n";
$changed = 42;
$pass = 1;
while ($changed > 0) {
print "pass {$pass}\n";