/**
* Make regular expression for case insensitive match
* Example (non ASCII): "123_слово_test" => "123_(с|С)(л|Л)(о|О)(в|В)(о|О)_[tT][eE][sS][tT]"
* Example (only ASCII): "123_test" => "(?i:123_test)"
*
* @param string $s
* @param string|null $delimiter If the optional delimiter is specified, it will also be escaped.
* This is useful for escaping the delimiter that is required by the PCRE functions.
* The / is the most commonly used delimiter.
* @return string|bool|null Returns FALSE if error occurred
*/
public static function preg_quote_case_insensitive($s, $delimiter = null)
{
if (is_null($s)) {
return $s;
}
if (self::is_ascii($s)) {
return '(?i:' . preg_quote($s, $delimiter) . ')';
}
#speed improve
$s_re = '';
$s_lc = PHPUTF8::lowercase($s);
if ($s_lc === false) {
return false;
}
$s_uc = PHPUTF8::uppercase($s);
if ($s_uc === false) {
return false;
}
$chars_lc = PHPUTF8::str_split($s_lc);
if ($chars_lc === false) {
return false;
}
$chars_uc = PHPUTF8::str_split($s_uc);
if ($chars_uc === false) {
return false;
}
foreach ($chars_lc as $i => $char) {
if ($chars_lc[$i] === $chars_uc[$i]) {
$s_re .= preg_quote($chars_lc[$i], $delimiter);
} elseif (self::is_ascii($chars_lc[$i])) {
$s_re .= '[' . preg_quote($chars_lc[$i] . $chars_uc[$i], $delimiter) . ']';
} else {
$s_re .= '(' . preg_quote($chars_lc[$i], $delimiter) . '|' . preg_quote($chars_uc[$i], $delimiter) . ')';
}
}
return $s_re;
}
private static function _make_regexp_callback(array $m)
{
#$re_holes = '[\x00-\x20\-_\*\~\.\'"\^=`:]';
#$re_holes = '[\x00-\x2f\x3a-\x40\x5b-\x60\x7b-\x7f]';
$re_holes = '(?!/\\\\)[^\\p{L}\\d]';
#non letter, non digit, non '/\'
if ($m[0] === 'а') {
$re = '[@аА]++ (?>[:holes:]|[@аА]+)*+';
} elseif ($m[0] === 'з') {
$re = '[3зЗ]++ (?>[:holes:]|[3зЗ]+)*+';
} elseif ($m[0] === 'б') {
$re = '[6бБ]++ (?>[:holes:]|[6бБ]+)*+';
} elseif ($m[0] === 'л') {
$re = '(?>[лЛ]+|/\\\\)++ (?>[:holes:]|[лЛ]+|/\\\\)*+';
} else {
#в PCRE-7.2 флаг /i в комбинации с /u в регулярном выражении почему-то не работает (BUG?)
#поэтому делаем класс символов с буквами в обоих регистрах
$char = '[' . preg_quote($m[0] . PHPUTF8::uppercase($m[0]), '~') . ']';
$re = str_replace('$0', $char, '$0++ (?>[:holes:]|$0+)*+');
}
return str_replace('[:holes:]', $re_holes, $re . "\r\n");
}
