functions/decode/utf_8.php

   1 <?php
   2 /**
   3  * decode/utf-8.php
   4  *
   5  * Copyright (c) 2003-2004 The SquirrelMail Project Team
   6  * Licensed under the GNU GPL. For full terms see the file COPYING.
   7  *
   8  * This file contains utf-8 decoding function that is needed to read
   9  * utf-8 encoded mails in non-utf-8 locale.
  10  *
  11  * Every decoded character consists of n bytes. First byte is octal
  12  * 300-375, other bytes - always octals 200-277.
  13  *
  14  * \a\b characters are decoded to html code octdec(a-300)*64 + octdec(b-200)
  15  * \a\b\c characters are decoded to html code octdec(a-340)*64*64 + octdec(b-200)*64 + octdec(c-200)
  16  *
  17  * decoding cycle is unfinished. please test and report problems to tokul@users.sourceforge.net
  18  *
  19  * @version $Id$
  20  * @package squirrelmail
  21  * @subpackage decode
  22  */
  23
  24 /**
  25  * Decode utf-8 strings
  26  * @param string $string Encoded string
  27  * @return string Decoded string
  28  */
  29 function charset_decode_utf_8 ($string) {
  30   global $default_charset,$squirrelmail_language;
  31
  32     if ($squirrelmail_language == 'ja_JP')
  33         return $string;
  34
  35     // don't do decoding when there are no 8bit symbols
  36     if (! sq_is8bit($string,'utf-8'))
  37         return $string;
  38
  39     // decode three byte unicode characters
  40     $string = preg_replace("/([\340-\357])([\200-\277])([\200-\277])/e",
  41     "'&#'.((ord('\\1')-224)*4096+(ord('\\2')-128)*64+(ord('\\3')-128)).';'",
  42     $string);
  43
  44     // decode two byte unicode characters
  45     $string = preg_replace("/([\300-\337])([\200-\277])/e",
  46     "'&#'.((ord('\\1')-192)*64+(ord('\\2')-128)).';'",
  47     $string);
  48
  49     return $string;
  50 }
  51 ?>