/**
* Takes an array of integers representing the Unicode characters and returns
* a UTF-8 string.
*
* @param array $unicode The array of Unicode code points
* @param bool $strict Should the function throw an error if something goes wrong
* @return string The UTF-8 encoded string or false on failure
*/
function utf8_from_unicode($unicode, $strict = false)
{
$rv = '';
$count = count($unicode);
for ($i = 0; $i < $count; $i++) {
$cp = $unicode[$i];
if (utf8_is_ascii($cp)) {
$rv .= chr($cp);
continue;
}
if (!unicode_is_valid($cp)) {
if ($strict) {
$msg = unicode_is_surrogate($cp) ? 'Illegal surrogate in UTF-8' : 'Codepoint out of Unicode range';
trigger_error(sprintf('utf8_from_unicode: %s at index: %d, value: 0x%X', $msg, $i, $cp));
return false;
}
continue;
}
$seqlen = utf8_sequence_length($cp);
$rv .= chr(utf8_encode_identifier($cp, $seqlen));
$offset = $seqlen - 2;
do {
$rv .= chr(utf8_encode_trail($cp, $offset));
} while ($offset--);
}
return $rv;
}
/**
* @dataProvider providerFailingUtf8EncodeTrail
*/
public function testUtf8FailingEncodeTrailReturnValue($ord, $offset)
{
$this->assertFalse(@utf8_encode_trail($ord, $offset));
}
/**
* UTF-8 aware replacement for chr().
*
* Returns a string containing the character specified by UTF-8.
*
* @param int $cp The Unicode code point
* @param bool $strict Should the function throw an error if something goes wrong
* @return string The character specified by UTF-8 or empty string/false on
* failure
*/
function utf8_chr($cp, $strict = false)
{
if (utf8_is_ascii($cp)) {
return chr($cp);
}
if (!unicode_is_valid($cp)) {
if ($strict) {
$msg = unicode_is_surrogate($cp) ? 'Illegal surrogate' : 'Codepoint out of Unicode range';
trigger_error(sprintf('utf8_chr: %s, value: 0x%X', $msg, $cp));
return false;
}
return '';
}
$seqlen = utf8_sequence_length($cp);
$char = chr(utf8_encode_identifier($cp, $seqlen));
$offset = $seqlen - 2;
do {
$char .= chr(utf8_encode_trail($cp, $offset));
} while ($offset--);
return $char;
}
