Move rcube_addressbook::normalize_string() to rcube_utils::normalize_string() for general purpose
| | |
| | | * |
| | | * @param string Input string (UTF-8) |
| | | * @return string Normalized string |
| | | * @deprecated since 0.9-beta |
| | | */ |
| | | protected static function normalize_string($str) |
| | | { |
| | | // split by words |
| | | $arr = explode(" ", preg_replace( |
| | | array('/[\s;\+\-\/]+/i', '/(\d)[-.\s]+(\d)/', '/\s\w{1,3}\s/'), |
| | | array(' ', '\\1\\2', ' '), |
| | | $str)); |
| | | |
| | | foreach ($arr as $i => $part) { |
| | | if (utf8_encode(utf8_decode($part)) == $part) { // is latin-1 ? |
| | | $arr[$i] = utf8_encode(strtr(strtolower(strtr(utf8_decode($part), |
| | | 'ÇçäâàåéêëèïîìÅÉöôòüûùÿøØáíóúñÑÁÂÀãÃÊËÈÍÎÏÓÔõÕÚÛÙýÝ', |
| | | 'ccaaaaeeeeiiiaeooouuuyooaiounnaaaaaeeeiiioooouuuyy')), |
| | | array('ß' => 'ss', 'ae' => 'a', 'oe' => 'o', 'ue' => 'u'))); |
| | | } |
| | | else |
| | | $arr[$i] = mb_strtolower($part); |
| | | } |
| | | |
| | | return join(" ", $arr); |
| | | return rcbe_utils::normalize_string($str); |
| | | } |
| | | |
| | | |
| | | /** |
| | | * Compose a valid display name from the given structured contact data |
| | |
| | | // fulltext search in all fields |
| | | else if ($col == '*') { |
| | | $words = array(); |
| | | foreach (explode($WS, self::normalize_string($value)) as $word) { |
| | | foreach (explode($WS, rcube_utils::normalize_string($value)) as $word) { |
| | | switch ($mode) { |
| | | case 1: // strict |
| | | $words[] = '(' . $this->db->ilike('words', $word . '%') |
| | |
| | | // vCard field |
| | | else { |
| | | if (in_array($col, $this->fulltext_cols)) { |
| | | foreach (explode(" ", self::normalize_string($val)) as $word) { |
| | | foreach (rcube_utils::normalize_string($val, true) as $word) { |
| | | switch ($mode) { |
| | | case 1: // strict |
| | | $words[] = '(' . $this->db->ilike('words', $word . $WS . '%') |
| | |
| | | if (isset($value)) |
| | | $vcard->set($field, $value, $section); |
| | | if ($fulltext && is_array($value)) |
| | | $words .= ' ' . self::normalize_string(join(" ", $value)); |
| | | $words .= ' ' . rcube_utils::normalize_string(join(" ", $value)); |
| | | else if ($fulltext && strlen($value) >= 3) |
| | | $words .= ' ' . self::normalize_string($value); |
| | | $words .= ' ' . rcube_utils::normalize_string($value); |
| | | } |
| | | } |
| | | $out['vcard'] = $vcard->export(false); |
| | |
| | | return $at ? $user . '@' . $domain : $domain; |
| | | } |
| | | |
| | | /** |
| | | * Split the given string into word tokens |
| | | * |
| | | * @param string Input to tokenize |
| | | * @return array List of tokens |
| | | */ |
| | | public static function tokenize_string($str) |
| | | { |
| | | return explode(" ", preg_replace( |
| | | array('/[\s;\/+-]+/i', '/(\d)[-.\s]+(\d)/', '/\s\w{1,3}\s/u'), |
| | | array(' ', '\\1\\2', ' '), |
| | | $str)); |
| | | } |
| | | |
| | | /** |
| | | * Normalize the given string for fulltext search. |
| | | * Currently only optimized for Latin-1 characters; to be extended |
| | | * |
| | | * @param string Input string (UTF-8) |
| | | * @param boolean True to return list of words as array |
| | | * @return mixed Normalized string or a list of normalized tokens |
| | | */ |
| | | public static function normalize_string($str, $as_array = false) |
| | | { |
| | | // split by words |
| | | $arr = self::tokenize_string($str); |
| | | |
| | | foreach ($arr as $i => $part) { |
| | | if (utf8_encode(utf8_decode($part)) == $part) { // is latin-1 ? |
| | | $arr[$i] = utf8_encode(strtr(strtolower(strtr(utf8_decode($part), |
| | | 'ÇçäâàåéêëèïîìÅÉöôòüûùÿøØáíóúñÑÁÂÀãÃÊËÈÍÎÏÓÔõÕÚÛÙýÝ', |
| | | 'ccaaaaeeeeiiiaeooouuuyooaiounnaaaaaeeeiiioooouuuyy')), |
| | | array('ß' => 'ss', 'ae' => 'a', 'oe' => 'o', 'ue' => 'u'))); |
| | | } |
| | | else |
| | | $arr[$i] = mb_strtolower($part); |
| | | } |
| | | |
| | | return $as_array ? $arr : join(" ", $arr); |
| | | } |
| | | |
| | | } |