| | |
| | | } |
| | | |
| | | foreach ($domain_array as $part) { |
| | | if (!preg_match('/^(([A-Za-z0-9][A-Za-z0-9-]{0,61}[A-Za-z0-9])|([A-Za-z0-9]))$/', $part)) { |
| | | if (!preg_match('/^((xn--)?([A-Za-z0-9][A-Za-z0-9-]{0,61}[A-Za-z0-9])|([A-Za-z0-9]))$/', $part)) { |
| | | return false; |
| | | } |
| | | } |
| | | |
| | | // last domain part |
| | | if (preg_match('/[^a-zA-Z]/', array_pop($domain_array))) { |
| | | $last_part = array_pop($domain_array); |
| | | if (strpos($last_part, 'xn--') !== 0 && preg_match('/[^a-zA-Z]/', $last_part)) { |
| | | return false; |
| | | } |
| | | |
| | |
| | | * Improved equivalent to strtotime() |
| | | * |
| | | * @param string $date Date string |
| | | * @param object DateTimeZone to use for DateTime object |
| | | * |
| | | * @return int Unix timestamp |
| | | */ |
| | | public static function strtotime($date) |
| | | public static function strtotime($date, $timezone = null) |
| | | { |
| | | $date = self::clean_datestr($date); |
| | | $tzname = $timezone ? ' ' . $timezone->getName() : ''; |
| | | |
| | | // unix timestamp |
| | | if (is_numeric($date)) { |
| | |
| | | |
| | | // if date parsing fails, we have a date in non-rfc format. |
| | | // remove token from the end and try again |
| | | while ((($ts = @strtotime($date)) === false) || ($ts < 0)) { |
| | | while ((($ts = @strtotime($date . $tzname)) === false) || ($ts < 0)) { |
| | | $d = explode(' ', $date); |
| | | array_pop($d); |
| | | if (!$d) { |
| | |
| | | * Date parsing function that turns the given value into a DateTime object |
| | | * |
| | | * @param string $date Date string |
| | | * @param object DateTimeZone to use for DateTime object |
| | | * |
| | | * @return object DateTime instance or false on failure |
| | | */ |
| | |
| | | // try to parse string with DateTime first |
| | | if (!empty($date)) { |
| | | try { |
| | | $dt = new DateTime($date, $timezone); |
| | | $dt = $timezone ? new DateTime($date, $timezone) : new DateTime($date); |
| | | } |
| | | catch (Exception $e) { |
| | | // ignore |
| | |
| | | } |
| | | |
| | | // try our advanced strtotime() method |
| | | if (!$dt && ($timestamp = self::strtotime($date))) { |
| | | if (!$dt && ($timestamp = self::strtotime($date, $timezone))) { |
| | | try { |
| | | $dt = new DateTime("@".$timestamp); |
| | | if ($timezone) { |
| | | $dt->setTimezone($timezone); |
| | | } |
| | | } |
| | | catch (Exception $e) { |
| | | // ignore |
| | |
| | | * Split the given string into word tokens |
| | | * |
| | | * @param string Input to tokenize |
| | | * @param integer Minimum length of a single token |
| | | * @return array List of tokens |
| | | */ |
| | | public static function tokenize_string($str) |
| | | public static function tokenize_string($str, $minlen = 2) |
| | | { |
| | | return explode(" ", preg_replace( |
| | | array('/[\s;\/+-]+/i', '/(\d)[-.\s]+(\d)/', '/\s\w{1,3}\s/u'), |
| | | array(' ', '\\1\\2', ' '), |
| | | $str)); |
| | | $expr = array('/[\s;\/+-]+/ui', '/(\d)[-.\s]+(\d)/u'); |
| | | $repl = array(' ', '\\1\\2'); |
| | | |
| | | if ($minlen > 1) { |
| | | $minlen--; |
| | | $expr[] = "/(^|\s+)\w{1,$minlen}(\s+|$)/u"; |
| | | $repl[] = ' '; |
| | | } |
| | | |
| | | return array_filter(explode(" ", preg_replace($expr, $repl, $str))); |
| | | } |
| | | |
| | | /** |
| | |
| | | * |
| | | * @param string Input string (UTF-8) |
| | | * @param boolean True to return list of words as array |
| | | * @param integer Minimum length of tokens |
| | | * |
| | | * @return mixed Normalized string or a list of normalized tokens |
| | | */ |
| | | public static function normalize_string($str, $as_array = false) |
| | | public static function normalize_string($str, $as_array = false, $minlen = 2) |
| | | { |
| | | // replace 4-byte unicode characters with '?' character, |
| | | // these are not supported in default utf-8 charset on mysql, |
| | |
| | | . ')/', '?', $str); |
| | | |
| | | // split by words |
| | | $arr = self::tokenize_string($str); |
| | | $arr = self::tokenize_string($str, $minlen); |
| | | |
| | | // detect character set |
| | | if (utf8_encode(utf8_decode($str)) == $str) { |
| | |
| | | } |
| | | |
| | | return $as_array ? $arr : join(" ", $arr); |
| | | } |
| | | |
| | | /** |
| | | * Compare two strings for matching words (order not relevant) |
| | | * |
| | | * @param string Haystack |
| | | * @param string Needle |
| | | * @return boolen True if match, False otherwise |
| | | */ |
| | | public static function words_match($haystack, $needle) |
| | | { |
| | | $a_needle = self::tokenize_string($needle, 1); |
| | | $_haystack = join(" ", self::tokenize_string($haystack, 1)); |
| | | $valid = strlen($_haystack) > 0; |
| | | $hits = 0; |
| | | |
| | | foreach ($a_needle as $w) { |
| | | if ($valid) { |
| | | if (stripos($_haystack, $w) !== false) { |
| | | $hits++; |
| | | } |
| | | } |
| | | else if (stripos($haystack, $w) !== false) { |
| | | $hits++; |
| | | } |
| | | } |
| | | |
| | | return $hits >= count($a_needle); |
| | | } |
| | | |
| | | /** |
| | |
| | | |
| | | return $url; |
| | | } |
| | | |
| | | /** |
| | | * Generate a ramdom string |
| | | * |
| | | * @param int String length |
| | | * |
| | | * @return string The generated random string |
| | | */ |
| | | public static function random_bytes($length) |
| | | { |
| | | if (function_exists('openssl_random_pseudo_bytes')) { |
| | | $random = openssl_random_pseudo_bytes(ceil($length / 2)); |
| | | $random = bin2hex($random); |
| | | |
| | | // if the length wasn't even... |
| | | if ($length < strlen($random)) { |
| | | $random = substr($random, 0, $length); |
| | | } |
| | | } |
| | | else { |
| | | $alpha = 'ABCDEFGHIJKLMNOPQERSTUVXYZabcdefghijklmnopqrtsuvwxyz0123456789+*%&?!$-_='; |
| | | $random = ''; |
| | | |
| | | for ($i = 0; $i < $length; $i++) { |
| | | $random .= $alpha[rand(0, strlen($alpha)-1)]; |
| | | } |
| | | } |
| | | |
| | | return $random; |
| | | } |
| | | } |