X-Git-Url: https://vcs.fsf.org/?p=squirrelmail.git;a=blobdiff_plain;f=functions%2Fdecode%2Futf_8.php;h=263fb6e2bfcf9e79e8f4c32ee9297fb76d48d8ea;hp=968f8924267b5326a67f27768058c53ddf282c19;hb=8ed1923822b383ddb338e9eef75bb7f110cc47b4;hpb=e53c9681b4646b7ca1b0071434d360be56a80a6e

diff --git a/functions/decode/utf_8.php b/functions/decode/utf_8.php
index 968f8924..263fb6e2 100644
--- a/functions/decode/utf_8.php
+++ b/functions/decode/utf_8.php
@@ -1,21 +1,42 @@
 <?php
+
 /**
- * decode/utf-8.php
- *
- * Copyright (c) 2003-2004 The SquirrelMail Project Team
- * Licensed under the GNU GPL. For full terms see the file COPYING.
+ * functions/decode/utf-8.php - utf-8 decoding functions
  *
  * This file contains utf-8 decoding function that is needed to read
  * utf-8 encoded mails in non-utf-8 locale.
  *
  * Every decoded character consists of n bytes. First byte is octal
  * 300-375, other bytes - always octals 200-277.
+ *<pre>
+ * Ranges (first byte):
+ *                oct     dec    hex
+ * Two byte   - 300-337 192-223 C0-DF
+ * Three byte - 340-357 224-239 E0-EF
+ * Four byte  - 360-367 240-247 F0-F7
+ * Five byte  - 370-373 248-251 F8-FB
+ * Six byte   - 374-375 252-253 FC-FD
+ *
+ * \a\b characters are decoded to html code calculated with formula:
+ *  octdec(a-300)*64 + octdec(b-200)
  *
- * \a\b characters are decoded to html code octdec(a-300)*64 + octdec(b-200)
- * \a\b\c characters are decoded to html code octdec(a-340)*64*64 + octdec(b-200)*64 + octdec(c-200)
+ * \a\b\c characters are decoded to html code calculated with formula:
+ *  octdec(a-340)*64^2 + octdec(b-200)*64 + octdec(c-200)
  *
- * decoding cycle is unfinished. please test and report problems to tokul@users.sourceforge.net
- * 
+ * \a\b\c\d characters are decoded to html code calculated with formula:
+ *  octdec(a-360)*64^3 + octdec(b-200)*64^2 +
+ *  + octdec(c-200)*64 + octdec(d-200)
+ *
+ * \a\b\c\d\e characters are decoded to html code calculated with formula:
+ *  octdec(a-370)*64^4 + octdec(b-200)*64^3 +
+ *  + octdec(c-200)*64^2 + octdec(d-200)*64 + octdec(e-200)
+ *
+ * \a\b\c\d\e\f characters are decoded to html code calculated with formula:
+ *  octdec(a-374)*64^5 + octdec(b-200)*64^4 + octdec(c-200)*64^3 +
+ *  + octdec(d-200)*64^2 + octdec(e-200)*64 + octdec(f-200)
+ *</pre>
+ * @copyright 2003-2019 The SquirrelMail Project Team
+ * @license http://opensource.org/licenses/gpl-license.php GNU Public License
  * @version $Id$
  * @package squirrelmail
  * @subpackage decode
@@ -27,8 +48,9 @@
  * @return string Decoded string
  */
 function charset_decode_utf_8 ($string) {
-  global $default_charset,$squirrelmail_language;
+    global $squirrelmail_language;
 
+    // Japanese translation uses mbstring function to read utf-8
     if ($squirrelmail_language == 'ja_JP')
         return $string;
 
@@ -36,16 +58,37 @@ function charset_decode_utf_8 ($string) {
     if (! sq_is8bit($string,'utf-8'))
         return $string;
 
+    // decode six byte unicode characters
+    /* (i think currently there is no such symbol)
+    $string = preg_replace_callback("/([\374-\375])([\200-\277])([\200-\277])([\200-\277])([\200-\277])([\200-\277])/",
+    create_function ('$matches', 'return \'&#\'.((ord($matches[1])-252)*1073741824+(ord($matches[2])-200)*16777216+(ord($matches[3])-200)*262144+(ord($matches[4])-128)*4096+(ord($matches[5])-128)*64+(ord($matches[6])-128)).\';\';'),
+    $string);
+    */
+
+    // decode five byte unicode characters
+    /* (i think currently there is no such symbol)
+    $string = preg_replace_callback("/([\370-\373])([\200-\277])([\200-\277])([\200-\277])([\200-\277])/",
+    create_function ('$matches', 'return \'&#\'.((ord($matches[1])-248)*16777216+(ord($matches[2])-200)*262144+(ord($matches[3])-128)*4096+(ord($matches[4])-128)*64+(ord($matches[5])-128)).\';\';'),
+    $string);
+    */
+    
+    // decode four byte unicode characters
+    $string = preg_replace_callback("/([\360-\367])([\200-\277])([\200-\277])([\200-\277])/",
+    create_function ('$matches', 'return \'&#\'.((ord($matches[1])-240)*262144+(ord($matches[2])-128)*4096+(ord($matches[3])-128)*64+(ord($matches[4])-128)).\';\';'),
+    $string);
+
     // decode three byte unicode characters
-    $string = preg_replace("/([\340-\357])([\200-\277])([\200-\277])/e",
-    "'&#'.((ord('\\1')-224)*4096+(ord('\\2')-128)*64+(ord('\\3')-128)).';'",
+    $string = preg_replace_callback("/([\340-\357])([\200-\277])([\200-\277])/",
+    create_function ('$matches', 'return \'&#\'.((ord($matches[1])-224)*4096+(ord($matches[2])-128)*64+(ord($matches[3])-128)).\';\';'),
     $string);
 
     // decode two byte unicode characters
-    $string = preg_replace("/([\300-\337])([\200-\277])/e",
-    "'&#'.((ord('\\1')-192)*64+(ord('\\2')-128)).';'",
+    $string = preg_replace_callback("/([\300-\337])([\200-\277])/",
+    create_function ('$matches', 'return \'&#\'.((ord($matches[1])-192)*64+(ord($matches[2])-128)).\';\';'),
     $string);
 
+    // remove broken unicode
+    $string = preg_replace("/[\200-\237]|\240|[\241-\377]/",'?',$string);
+
     return $string;
 }
-?>
\ No newline at end of file