X-Git-Url: https://vcs.fsf.org/?p=squirrelmail.git;a=blobdiff_plain;f=functions%2Fi18n.php;h=f594276c1588abded965100d5a9c70b015f468c7;hp=137b491bb3073ce7895113d703f6d36739b7617e;hb=2c86c02c16613d2e95d735e15b2b26d17c563453;hpb=b8b4ac4ac132a0f77e2e3136d4d0f90d7ee432a8

diff --git a/functions/i18n.php b/functions/i18n.php
index 137b491b..f594276c 100644
--- a/functions/i18n.php
+++ b/functions/i18n.php
@@ -3,7 +3,7 @@
 /**
  * i18n.php
  *
- * Copyright (c) 1999-2002 The SquirrelMail Project Team
+ * Copyright (c) 1999-2003 The SquirrelMail Project Team
  * Licensed under the GNU GPL. For full terms see the file COPYING.
  *
  * This file contains variuos functions that are needed to do
@@ -15,21 +15,40 @@
  * $Id$
  */
 
+require_once(SM_PATH . 'functions/global.php');
+
 /* Decodes a string to the internal encoding from the given charset */
 function charset_decode ($charset, $string) {
+    global $languages, $squirrelmail_language;
+
+    if (isset($languages[$squirrelmail_language]['XTRA_CODE']) &&
+        function_exists($languages[$squirrelmail_language]['XTRA_CODE'])) {
+        $string = $languages[$squirrelmail_language]['XTRA_CODE']('decode', $string);
+    }
 
     /* All HTML special characters are 7 bit and can be replaced first */
+    
     $string = htmlspecialchars ($string);
 
     $charset = strtolower($charset);
 
+    set_my_charset() ;
+
     if (ereg('iso-8859-([[:digit:]]+)', $charset, $res)) {
         if ($res[1] == '1') {
             $ret = charset_decode_iso_8859_1 ($string);
         } else if ($res[1] == '2') {
             $ret = charset_decode_iso_8859_2 ($string);
+        } else if ($res[1] == '4') {
+            $ret = charset_decode_iso_8859_4 ($string);
+        } else if ($res[1] == '5') {
+            $ret = charset_decode_iso_8859_5 ($string);
         } else if ($res[1] == '7') {
             $ret = charset_decode_iso_8859_7 ($string);
+        } else if ($res[1] == '9') {
+            $ret = charset_decode_iso_8859_9 ($string);
+        } else if ($res[1] == '13') {
+            $ret = charset_decode_iso_8859_13 ($string);
         } else if ($res[1] == '15') {
             $ret = charset_decode_iso_8859_15 ($string);
         } else {
@@ -40,7 +59,15 @@ function charset_decode ($charset, $string) {
     } else if ($charset == 'koi8-r') {
         $ret = charset_decode_koi8r ($string);
     } else if ($charset == 'windows-1251') {
-        $ret = charset_decode_koi8r ($string);
+        $ret = charset_decode_windows_1251 ($string);
+    } else if ($charset == 'windows-1253') {
+	$ret = charset_decode_windows_1253 ($string);
+    } else if ($charset == 'windows-1254') {
+	$ret = charset_decode_windows_1254 ($string);
+    } else if ($charset == 'windows-1257') {
+        $ret = charset_decode_windows_1257 ($string);
+    } else if ($charset == 'utf-8') {
+	$ret = charset_decode_utf8 ($string);
     } else {
         $ret = $string;
     }
@@ -397,6 +424,75 @@ function charset_decode_iso_8859_2 ($string) {
     return $string;
 }
 
+/* 
+ ISO/IEC 8859-4:1998 Latin Alphabet No. 4
+*/
+
+function charset_decode_iso_8859_4 ($string) {
+    global $default_charset;
+
+    if (strtolower($default_charset) == 'iso-8859-4')
+        return $string;
+
+    /* Only do the slow convert if there are 8-bit characters */
+    if (! ereg("[\200-\377]", $string))
+        return $string;
+
+    $string = str_replace ("\241", 'Ą', $string);
+    $string = str_replace ("\242", 'ĸ', $string);
+    $string = str_replace ("\243", 'Ŗ', $string);
+    $string = str_replace ("\245", 'Ĩ', $string);
+    $string = str_replace ("\246", 'Ļ', $string);
+    $string = str_replace ("\251", 'Š', $string);
+    $string = str_replace ("\252", 'Ē', $string);
+    $string = str_replace ("\253", 'Ģ', $string);
+    $string = str_replace ("\254", 'Ŧ', $string);
+    $string = str_replace ("\256", 'Ž', $string);
+    $string = str_replace ("\261", 'ą', $string);
+    $string = str_replace ("\262", '˛', $string);
+    $string = str_replace ("\263", 'ŗ', $string);
+    $string = str_replace ("\265", 'ĩ', $string);
+    $string = str_replace ("\266", 'ļ', $string);
+    $string = str_replace ("\267", 'ˇ', $string);
+    $string = str_replace ("\271", 'š', $string);
+    $string = str_replace ("\272", 'ē', $string);
+    $string = str_replace ("\273", 'ģ', $string);
+    $string = str_replace ("\274", 'ŧ', $string);
+    $string = str_replace ("\275", 'Ŋ', $string);
+    $string = str_replace ("\276", 'ž', $string);
+    $string = str_replace ("\277", 'ŋ', $string);
+    $string = str_replace ("\300", 'Ā', $string);
+    $string = str_replace ("\307", 'Į', $string);
+    $string = str_replace ("\310", 'Č', $string);
+    $string = str_replace ("\312", 'Ę', $string);
+    $string = str_replace ("\314", 'Ė', $string);
+    $string = str_replace ("\317", 'Ī', $string);
+    $string = str_replace ("\320", 'Đ', $string);
+    $string = str_replace ("\321", 'Ņ', $string);
+    $string = str_replace ("\322", 'Ō', $string);
+    $string = str_replace ("\323", 'Ķ', $string);
+    $string = str_replace ("\331", 'Ų', $string);
+    $string = str_replace ("\335", 'Ũ', $string);
+    $string = str_replace ("\336", 'Ū', $string);
+    $string = str_replace ("\340", 'ā', $string);
+    $string = str_replace ("\347", 'į', $string);
+    $string = str_replace ("\350", 'č', $string);
+    $string = str_replace ("\352", 'ę', $string);
+    $string = str_replace ("\354", 'ė', $string);
+    $string = str_replace ("\357", 'ī', $string);
+    $string = str_replace ("\360", 'đ', $string);
+    $string = str_replace ("\361", 'ņ', $string);
+    $string = str_replace ("\362", 'ō', $string);
+    $string = str_replace ("\363", 'ķ', $string);
+    $string = str_replace ("\371", 'ų', $string);
+    $string = str_replace ("\375", 'ũ', $string);
+    $string = str_replace ("\376", 'ū', $string);
+    $string = str_replace ("\377", '˙', $string);
+
+    // rest of charset is the same as ISO-8859-1
+    return (charset_decode_iso_8859_1($string));
+}
+
 /* iso-8859-7 is Greek. */
 function charset_decode_iso_8859_7 ($string) {
     global $default_charset;
@@ -435,11 +531,8 @@ function charset_decode_iso_8859_7 ($string) {
      * ISO-8859-7 characters from 11/04 (0xB4) to 11/06 (0xB6)
      * These are Unicode 900-902
      */
-    while (ereg("([\264-\266])", $string, $res)) {
-        $replace = '&#' . (ord($res[1])+720) . ';';
-        $string = str_replace($res[1], $replace, $string);
-    }
-
+    $string = preg_replace("/([\264-\266])/e","'&#' . (ord('\\1')+720);",$string);
+    
     /* 11/07 (0xB7) Middle dot is the same in iso-8859-1 */
     $string = str_replace("\267", '·', $string);
 
@@ -447,10 +540,7 @@ function charset_decode_iso_8859_7 ($string) {
      * ISO-8859-7 characters from 11/08 (0xB8) to 11/10 (0xBA)
      * These are Unicode 900-902
      */
-    while (ereg("([\270-\272])", $string, $res)) {
-        $replace = '&#' . (ord($res[1])+720) . ";";
-        $string = str_replace($res[1], $replace, $string);
-    }
+    $string = preg_replace("/([\270-\272])/e","'&#' . (ord('\\1')+720);",$string);
 
     /*
      * 11/11 (0xBB) Right angle quotation mark is the same as in
@@ -459,14 +549,117 @@ function charset_decode_iso_8859_7 ($string) {
     $string = str_replace("\273", '»', $string);
 
     /* And now the rest of the charset */
-    while (ereg("([\274-\376])", $string, $res)) {
-        $replace = '&#' . (ord($res[1])+720) . ';';
-        $string = str_replace($res[1], $replace, $string);
-    }
+    $string = preg_replace("/([\274-\376])/e","'&#'.(ord('\\1')+720);",$string);
 
     return $string;
 }
 
+/*
+ ISOIEC 8859-9:1999 Latin Alphabet No. 5
+
+*/
+function charset_decode_iso_8859_9 ($string) {
+    global $default_charset;
+
+    if (strtolower($default_charset) == 'iso-8859-9')
+        return $string;
+
+    /* Only do the slow convert if there are 8-bit characters */
+    if (! ereg("[\200-\377]", $string))
+        return $string;
+
+    // latin capital letter g with breve 208->286
+    $string = str_replace("\320", 'Ğ', $string);
+    // latin capital letter i with dot above 221->304
+    $string = str_replace("\335", 'İ', $string);
+    // latin capital letter s with cedilla 222->350
+    $string = str_replace("\336", 'Ş', $string);
+    // latin small letter g with breve 240->287
+    $string = str_replace("\360", 'ğ', $string);
+    // latin small letter dotless i 253->305
+    $string = str_replace("\375", 'ı', $string);
+    // latin small letter s with cedilla 254->351
+    $string = str_replace("\376", 'ş', $string);
+
+    // rest of charset is the same as ISO-8859-1
+    return (charset_decode_iso_8859_1($string));
+}
+
+
+/*
+ ISO/IEC 8859-13:1998 Latin Alphabet No. 7 (Baltic Rim) 
+*/
+function charset_decode_iso_8859_13 ($string) {
+    global $default_charset;
+
+    if (strtolower($default_charset) == 'iso-8859-13')
+        return $string;
+
+    /* Only do the slow convert if there are 8-bit characters */
+    if (! ereg("[\200-\377]", $string))
+        return $string;
+
+    $string = str_replace ("\241", '”', $string);
+    $string = str_replace ("\245", '„', $string);
+    $string = str_replace ("\250", 'Ø', $string);
+    $string = str_replace ("\252", 'Ŗ', $string);
+    $string = str_replace ("\257", 'Æ', $string);
+    $string = str_replace ("\264", '“', $string);
+    $string = str_replace ("\270", 'ø', $string);
+    $string = str_replace ("\272", 'ŗ', $string);
+    $string = str_replace ("\277", 'æ', $string);
+    $string = str_replace ("\300", 'Ą', $string);
+    $string = str_replace ("\301", 'Į', $string);
+    $string = str_replace ("\302", 'Ā', $string);
+    $string = str_replace ("\303", 'Ć', $string);
+    $string = str_replace ("\306", 'Ę', $string);
+    $string = str_replace ("\307", 'Ē', $string);
+    $string = str_replace ("\310", 'Č', $string);
+    $string = str_replace ("\312", 'Ź', $string);
+    $string = str_replace ("\313", 'Ė', $string);
+    $string = str_replace ("\314", 'Ģ', $string);
+    $string = str_replace ("\315", 'Ķ', $string);
+    $string = str_replace ("\316", 'Ī', $string);
+    $string = str_replace ("\317", 'Ļ', $string);
+    $string = str_replace ("\320", 'Š', $string);
+    $string = str_replace ("\321", 'Ń', $string);
+    $string = str_replace ("\322", 'Ņ', $string);
+    $string = str_replace ("\324", 'Ō', $string);
+    $string = str_replace ("\330", 'Ų', $string);
+    $string = str_replace ("\331", 'Ł', $string);
+    $string = str_replace ("\332", 'Ś', $string);
+    $string = str_replace ("\333", 'Ū', $string);
+    $string = str_replace ("\335", 'Ż', $string);
+    $string = str_replace ("\336", 'Ž', $string);
+    $string = str_replace ("\340", 'ą', $string);
+    $string = str_replace ("\341", 'į', $string);
+    $string = str_replace ("\342", 'ā', $string);
+    $string = str_replace ("\343", 'ć', $string);
+    $string = str_replace ("\346", 'ę', $string);
+    $string = str_replace ("\347", 'ē', $string);
+    $string = str_replace ("\350", 'č', $string);
+    $string = str_replace ("\352", 'ź', $string);
+    $string = str_replace ("\353", 'ė', $string);
+    $string = str_replace ("\354", 'ģ', $string);
+    $string = str_replace ("\355", 'ķ', $string);
+    $string = str_replace ("\356", 'ī', $string);
+    $string = str_replace ("\357", 'ļ', $string);
+    $string = str_replace ("\360", 'š', $string);
+    $string = str_replace ("\361", 'ń', $string);
+    $string = str_replace ("\362", 'ņ', $string);
+    $string = str_replace ("\364", 'ō', $string);
+    $string = str_replace ("\370", 'ų', $string);
+    $string = str_replace ("\371", 'ł', $string);
+    $string = str_replace ("\372", 'ś', $string);
+    $string = str_replace ("\373", 'ū', $string);    
+    $string = str_replace ("\375", 'ż', $string);
+    $string = str_replace ("\376", 'ž', $string);
+    $string = str_replace ("\377", '’', $string);
+
+    // rest of charset is the same as ISO-8859-1
+    return (charset_decode_iso_8859_1($string));
+}
+
 /*
  * iso-8859-15 is Latin 9 and has very much the same use as Latin 1
  * but has the Euro symbol and some characters needed for French.
@@ -475,7 +668,7 @@ function charset_decode_iso_8859_15 ($string) {
     // Euro sign
     $string = str_replace ("\244", '€', $string);
     // Latin capital letter S with caron
-    $string = str_replace ("\244", 'Š', $string);
+    $string = str_replace ("\246", 'Š', $string);
     // Latin small letter s with caron
     $string = str_replace ("\250", 'š', $string);
     // Latin capital letter Z with caron
@@ -492,7 +685,7 @@ function charset_decode_iso_8859_15 ($string) {
     return (charset_decode_iso_8859_1($string));
 }
 
-/* ISO-8859-15 is Cyrillic */
+/* ISO-8859-5 is Cyrillic */
 function charset_decode_iso_8859_5 ($string) {
     // Convert to KOI8-R, then return this decoded.
     $string = convert_cyr_string($string, 'i', 'k');
@@ -682,6 +875,235 @@ function charset_decode_koi8r ($string) {
     return $string;
 }
 
+/*
+ windows-1254 (Turks)
+ */
+function charset_decode_windows_1254 ($string) {
+    global $default_charset;
+
+    if (strtolower($default_charset) == 'windows-1254')
+        return $string;
+
+    /* Only do the slow convert if there are 8-bit characters */
+    if (! ereg("[\200-\377]", $string))
+        return $string;
+
+    // Euro sign 128 -> 8364
+    $string = str_replace("\200", '€', $string);
+    // Single low-9 quotation mark 130 -> 8218
+    $string = str_replace("\202", '‚', $string);
+    // latin small letter f with hook 131 -> 402
+    $string = str_replace("\203", 'ƒ', $string);
+    // Double low-9 quotation mark 132 -> 8222
+    $string = str_replace("\204", '„', $string);
+    // horizontal ellipsis 133 -> 8230
+    $string = str_replace("\205", '…', $string);
+    // dagger 134 -> 8224
+    $string = str_replace("\206", '†', $string);
+    // double dagger 135 -> 8225
+    $string = str_replace("\207", '‡', $string);
+    // modifier letter circumflex accent 136->710
+    $string = str_replace("\210", 'ˆ', $string);
+    // per mille sign 137 -> 8240
+    $string = str_replace("\211", '‰', $string);
+    // latin capital letter s with caron 138 -> 352
+    $string = str_replace("\212", 'Š', $string);
+    // single left-pointing angle quotation mark 139 -> 8249
+    $string = str_replace("\213", '‹', $string);
+    // latin capital ligature oe 140 -> 338
+    $string = str_replace("\214", 'Œ', $string);
+    // left single quotation mark 145 -> 8216
+    $string = str_replace("\221", '‘', $string);
+    // right single quotation mark 146 -> 8217
+    $string = str_replace("\222", '’', $string);
+    // left double quotation mark 147 -> 8220
+    $string = str_replace("\223", '“', $string);
+    // right double quotation mark 148 -> 8221
+    $string = str_replace("\224", '”', $string);
+    // bullet 149 -> 8226
+    $string = str_replace("\225", '•', $string);
+    // en dash 150 -> 8211
+    $string = str_replace("\226", '–', $string);
+    // em dash 151 -> 8212
+    $string = str_replace("\227", '—', $string);
+    // small tilde 152 -> 732
+    $string = str_replace("\230", '˜', $string);
+    // trade mark sign 153 -> 8482
+    $string = str_replace("\231", '™', $string);
+    // latin small letter s with caron 154 -> 353
+    $string = str_replace("\232", 'š', $string);
+    // single right-pointing angle quotation mark 155 -> 8250
+    $string = str_replace("\233", '›', $string);
+    // latin small ligature oe 156 -> 339
+    $string = str_replace("\234", 'œ', $string);
+    // latin capital letter y with diaresis 159->376
+    $string = str_replace("\237", 'Ÿ', $string);
+    // latin capital letter g with breve 208->286
+    $string = str_replace("\320", 'Ğ', $string);
+    // latin capital letter i with dot above 221->304
+    $string = str_replace("\335", 'İ', $string);
+    // latin capital letter s with cedilla 222->350
+    $string = str_replace("\336", 'Ş', $string);
+    // latin small letter g with breve 240->287
+    $string = str_replace("\360", 'ğ', $string);
+    // latin small letter dotless i 253->305
+    $string = str_replace("\375", 'ı', $string);
+    // latin small letter s with cedilla 254->351
+    $string = str_replace("\376", 'ş', $string);
+
+    // Rest of charset is like iso-8859-1
+    return (charset_decode_iso_8859_1($string));
+}
+
+/*
+ windows-1253 (Greek)
+ */
+function charset_decode_windows_1253 ($string) {
+    global $default_charset;
+
+    if (strtolower($default_charset) == 'windows-1253')
+        return $string;
+
+    /* Only do the slow convert if there are 8-bit characters */
+    if (! ereg("[\200-\377]", $string))
+        return $string;
+
+    $string = str_replace("\200", '€', $string);
+    $string = str_replace("\202", '‚', $string);
+    $string = str_replace("\203", 'ƒ', $string);
+    $string = str_replace("\204", '„', $string);
+    $string = str_replace("\205", '…', $string);
+    $string = str_replace("\206", '†', $string);
+    $string = str_replace("\207", '‡', $string);
+    $string = str_replace("\211", '‰', $string);
+    $string = str_replace("\213", '‹', $string);
+    $string = str_replace("\221", '‘', $string);
+    $string = str_replace("\222", '’', $string);
+    $string = str_replace("\223", '“', $string);
+    $string = str_replace("\224", '”', $string);
+    $string = str_replace("\225", '•', $string);
+    $string = str_replace("\226", '–', $string);
+    $string = str_replace("\227", '—', $string);
+    $string = str_replace("\231", '™', $string);
+    $string = str_replace("\233", '›', $string);
+    $string = str_replace("\241", '΅', $string);
+    $string = str_replace("\242", 'Ά', $string);
+    $string = str_replace ("\257", '―', $string);
+    $string = str_replace("\264", '΄', $string);
+    $string = str_replace("\270", 'Έ', $string);
+    $string = str_replace ("\271", 'Ή', $string);
+    $string = str_replace ("\272", 'Ί', $string);
+    $string = str_replace ("\274", 'Ό', $string);
+    // cycle for 190-254 symbols
+    $string = preg_replace("/([\274-\376])/e","'&#' . (ord('\\1')+720);",$string);
+
+    // Rest of charset is like iso-8859-1
+    return (charset_decode_iso_8859_1($string));
+}
+
+
+/*
+ windows-1257 (BaltRim)
+ */
+function charset_decode_windows_1257 ($string) {
+    global $default_charset;
+
+    if (strtolower($default_charset) == 'windows-1257')
+        return $string;
+
+    /* Only do the slow convert if there are 8-bit characters */
+    if (! ereg("[\200-\377]", $string))
+        return $string;
+
+    $string = str_replace("\200", '€', $string);
+    $string = str_replace("\202", '‚', $string);
+    $string = str_replace("\204", '„', $string);
+    $string = str_replace("\205", '…', $string);
+    $string = str_replace("\206", '†', $string);
+    $string = str_replace("\207", '‡', $string);
+    $string = str_replace("\211", '‰', $string);
+    $string = str_replace("\213", '‹', $string);
+    $string = str_replace("\215", '¨', $string);
+    $string = str_replace("\216", 'ˇ', $string);
+    $string = str_replace("\217", '¸', $string);
+    $string = str_replace("\221", '‘', $string);
+    $string = str_replace("\222", '’', $string);
+    $string = str_replace("\223", '“', $string);
+    $string = str_replace("\224", '”', $string);
+    $string = str_replace("\225", '•', $string);
+    $string = str_replace("\226", '–', $string);
+    $string = str_replace("\227", '—', $string);
+    $string = str_replace("\231", '™', $string);
+    $string = str_replace("\233", '›', $string);
+    $string = str_replace("\235", '¯', $string);
+    $string = str_replace("\236", '˛', $string);
+    $string = str_replace ("\250", 'Ø', $string);
+    $string = str_replace ("\252", 'Ŗ', $string);
+    $string = str_replace ("\257", 'Æ', $string);
+    $string = str_replace ("\270", 'ø', $string);
+    $string = str_replace ("\272", 'ŗ', $string);
+    $string = str_replace ("\277", 'æ', $string);
+    $string = str_replace ("\300", 'Ą', $string);
+    $string = str_replace ("\301", 'Į', $string);
+    $string = str_replace ("\302", 'Ā', $string);
+    $string = str_replace ("\303", 'Ć', $string);
+    $string = str_replace ("\306", 'Ę', $string);
+    $string = str_replace ("\307", 'Ē', $string);
+    $string = str_replace ("\310", 'Č', $string);
+    $string = str_replace ("\312", 'Ź', $string);
+    $string = str_replace ("\313", 'Ė', $string);
+    $string = str_replace ("\314", 'Ģ', $string);
+    $string = str_replace ("\315", 'Ķ', $string);
+    $string = str_replace ("\316", 'Ī', $string);
+    $string = str_replace ("\317", 'Ļ', $string);
+    $string = str_replace ("\320", 'Š', $string);
+    $string = str_replace ("\321", 'Ń', $string);
+    $string = str_replace ("\322", 'Ņ', $string);
+    $string = str_replace ("\324", 'Ō', $string);
+    $string = str_replace ("\330", 'Ų', $string);
+    $string = str_replace ("\331", 'Ł', $string);
+    $string = str_replace ("\332", 'Ŕ', $string);
+    $string = str_replace ("\333", 'Ū', $string);
+    $string = str_replace ("\335", 'Ż', $string);
+    $string = str_replace ("\336", 'Ž', $string);
+    $string = str_replace ("\340", 'ą', $string);
+    $string = str_replace ("\341", 'į', $string);
+    $string = str_replace ("\342", 'ā', $string);
+    $string = str_replace ("\343", 'ć', $string);
+    $string = str_replace ("\346", 'ę', $string);
+    $string = str_replace ("\347", 'ē', $string);
+    $string = str_replace ("\350", 'č', $string);
+    $string = str_replace ("\352", 'ź', $string);
+    $string = str_replace ("\353", 'ė', $string);
+    $string = str_replace ("\354", 'ģ', $string);
+    $string = str_replace ("\355", 'ķ', $string);
+    $string = str_replace ("\356", 'ī', $string);
+    $string = str_replace ("\357", 'ļ', $string);
+    $string = str_replace ("\360", 'š', $string);
+    $string = str_replace ("\361", 'ń', $string);
+    $string = str_replace ("\362", 'ņ', $string);
+    $string = str_replace ("\364", 'ō', $string);
+    $string = str_replace ("\370", 'ų', $string);
+    $string = str_replace ("\371", 'ł', $string);
+    $string = str_replace ("\372", 'ś', $string);
+    $string = str_replace ("\373", 'ū', $string);    
+    $string = str_replace ("\375", 'ż', $string);
+    $string = str_replace ("\376", 'ž', $string);
+    $string = str_replace ("\377", '˙', $string);
+
+    // Rest of charset is like iso-8859-1
+    return (charset_decode_iso_8859_1($string));
+}
+
+/* windows-1251 is Microsoft Cyrillic encoding */
+function charset_decode_windows_1251 ($string) {
+    // Convert to KOI8-R, then return this decoded.
+    $string = convert_cyr_string($string, 'w', 'k');
+    return charset_decode_koi8r($string);
+}
+
+
+
 /*
  * Set up the language to be output
  * if $do_search is true, then scan the browser information
@@ -690,17 +1112,19 @@ function charset_decode_koi8r ($string) {
 function set_up_language($sm_language, $do_search = false) {
 
     static $SetupAlready = 0;
-    global $HTTP_ACCEPT_LANGUAGE, $use_gettext, $languages,
+    global $use_gettext, $languages,
            $squirrelmail_language, $squirrelmail_default_language,
            $sm_notAlias;
 
     if ($SetupAlready) {
         return;
     }
+
     $SetupAlready = TRUE;
+    sqgetGlobalVar('HTTP_ACCEPT_LANGUAGE',  $accept_lang, SQ_SERVER);
 
-    if ($do_search && ! $sm_language && isset($HTTP_ACCEPT_LANGUAGE)) {
-        $sm_language = substr($HTTP_ACCEPT_LANGUAGE, 0, 2);
+    if ($do_search && ! $sm_language && isset($accept_lang)) {
+        $sm_language = substr($accept_lang, 0, 2);
     }
     
     if (!$sm_language && isset($squirrelmail_default_language)) {
@@ -712,23 +1136,32 @@ function set_up_language($sm_language, $do_search = false) {
         $sm_notAlias = $languages[$sm_notAlias]['ALIAS'];
     }
 
-    if ( isset($sm_language) && 
+    if ( isset($sm_language) &&
          $use_gettext &&
          $sm_language != '' &&
          isset($languages[$sm_notAlias]['CHARSET']) ) {
-        bindtextdomain( 'squirrelmail', '../locale/' );
-        textdomain( 'squirrelmail' );         
-        if ( !ini_get('safe_mode') && 
+        bindtextdomain( 'squirrelmail', SM_PATH . 'locale/' );
+        textdomain( 'squirrelmail' );
+        if ( !ini_get('safe_mode') &&
              getenv( 'LC_ALL' ) != $sm_notAlias ) {
             putenv( "LC_ALL=$sm_notAlias" );
             putenv( "LANG=$sm_notAlias" );
             putenv( "LANGUAGE=$sm_notAlias" );
         }
-        setlocale('LC_ALL', $sm_notAlias);
+        setlocale(LC_ALL, $sm_notAlias);
         $squirrelmail_language = $sm_notAlias;
+        if ($squirrelmail_language == 'ja_JP' && function_exists('mb_detect_encoding') ) {
+            header ('Content-Type: text/html; charset=EUC-JP');
+            if (!function_exists('mb_internal_encoding')) {
+                echo _("You need to have php4 installed with the multibyte string function enabled (using configure option --with-mbstring).");
+            }
+            mb_internal_encoding('EUC-JP');
+            mb_http_output('pass');
+        } else {
         header( 'Content-Type: text/html; charset=' . $languages[$sm_notAlias]['CHARSET'] );
     }
 }
+}
 
 function set_my_charset(){
 
@@ -741,11 +1174,11 @@ function set_my_charset(){
      * selection. This is "more right" (tm), than just stamping the
      * message blindly with the system-wide $default_charset.
      */
-    global $data_dir, $username, $default_charset, $languages;
-    
+    global $data_dir, $username, $default_charset, $languages, $squirrelmail_default_language;
+
     $my_language = getPref($data_dir, $username, 'language');
     if (!$my_language) {
-        return;
+        $my_language = $squirrelmail_default_language ;
     }
     while (isset($languages[$my_language]['ALIAS'])) {
         $my_language = $languages[$my_language]['ALIAS'];
@@ -789,6 +1222,10 @@ $languages['de']['ALIAS'] = 'de_DE';
 // There is no en_EN! There is en_US, en_BR, en_AU, and so forth, 
 // but who cares about !US, right? Right? :)
 
+$languages['el_GR']['NAME']    = 'Greek';
+$languages['el_GR']['CHARSET'] = 'iso-8859-7';
+$languages['el']['ALIAS'] = 'el_GR';
+
 $languages['en_US']['NAME']    = 'English';
 $languages['en_US']['CHARSET'] = 'iso-8859-1';
 $languages['en']['ALIAS'] = 'en_US';
@@ -829,8 +1266,14 @@ $languages['it_IT']['NAME']    = 'Italian';
 $languages['it_IT']['CHARSET'] = 'iso-8859-1';
 $languages['it']['ALIAS'] = 'it_IT';
 
+$languages['ja_JP']['NAME']    = 'Japanese';
+$languages['ja_JP']['CHARSET'] = 'iso-2022-jp';
+$languages['ja_JP']['XTRA_CODE'] = 'japanese_charset_xtra';
+$languages['ja']['ALIAS'] = 'ja_JP';
+
 $languages['ko_KR']['NAME']    = 'Korean';
 $languages['ko_KR']['CHARSET'] = 'euc-KR';
+$languages['ko_KR']['XTRA_CODE'] = 'korean_charset_xtra';
 $languages['ko']['ALIAS'] = 'ko_KR';
 
 $languages['nl_NL']['NAME']    = 'Dutch';
@@ -857,9 +1300,9 @@ $languages['ru_RU']['NAME']    = 'Russian';
 $languages['ru_RU']['CHARSET'] = 'koi8-r';
 $languages['ru']['ALIAS'] = 'ru_RU';
 
-$languages['sr']['NAME']    = 'Serbian';
-$languages['sr']['CHARSET'] = 'iso-8859-2';
-$languages['sr_YU']['ALIAS'] = 'sr';
+$languages['sr_YU']['NAME']    = 'Serbian';
+$languages['sr_YU']['CHARSET'] = 'iso-8859-2';
+$languages['sr']['ALIAS'] = 'sr_YU';
 
 $languages['sv_SE']['NAME']    = 'Swedish';
 $languages['sv_SE']['CHARSET'] = 'iso-8859-1';
@@ -869,12 +1312,14 @@ $languages['tr_TR']['NAME']    = 'Turkish';
 $languages['tr_TR']['CHARSET'] = 'iso-8859-9';
 $languages['tr']['ALIAS'] = 'tr_TR';
 
-// it's zh_TW, actually.
-
-$languages['zh_TW']['NAME']    = 'Taiwan';
+$languages['zh_TW']['NAME']    = 'Chinese Trad';
 $languages['zh_TW']['CHARSET'] = 'big5';
 $languages['tw']['ALIAS'] = 'zh_TW';
 
+$languages['zh_CN']['NAME']    = 'Chinese Simp';
+$languages['zh_CN']['CHARSET'] = 'gb2312';
+$languages['cn']['ALIAS'] = 'zh_CN';
+
 $languages['sk_SK']['NAME']     = 'Slovak';
 $languages['sk_SK']['CHARSET']  = 'iso-8859-2';
 $languages['sk']['ALIAS']       = 'sk_SK';
@@ -888,7 +1333,7 @@ $languages['th_TH']['CHARSET'] = 'tis-620';
 $languages['th']['ALIAS'] = 'th_TH';
 
 $languages['lt_LT']['NAME']    = 'Lithuanian';
-$languages['lt_LT']['CHARSET'] = 'iso-8859-13';
+$languages['lt_LT']['CHARSET'] = 'windows-1257';
 $languages['lt']['ALIAS'] = 'lt_LT';
 
 $languages['sl_SI']['NAME']    = 'Slovenian';
@@ -899,6 +1344,25 @@ $languages['bg_BG']['NAME']    = 'Bulgarian';
 $languages['bg_BG']['CHARSET'] = 'windows-1251';
 $languages['bg']['ALIAS'] = 'bg_BG';
 
+$languages['uk_UA']['NAME']    = 'Ukrainian';
+$languages['uk_UA']['CHARSET'] = 'koi8-u';
+$languages['uk']['ALIAS'] = 'uk_UA';
+
+// Right to left languages
+
+$languages['ar']['NAME']    = 'Arabic';
+$languages['ar']['CHARSET'] = 'windows-1256';
+$languages['ar']['DIR']     = 'rtl';
+
+$languages['he_IL']['NAME']    = 'Hebrew';
+$languages['he_IL']['CHARSET'] = 'windows-1255';
+$languages['he_IL']['DIR']     = 'rtl';
+$languages['he']['ALIAS']      = 'he_IL';
+
+$languages['vi_VN']['NAME']    = 'Vietnamese';
+$languages['vi_VN']['CHARSET'] = 'utf-8';
+$languages['vi']['ALIAS'] = 'vi_VN';
+
 /* Detect whether gettext is installed. */
 $gettext_flags = 0;
 if (function_exists('_')) {
@@ -918,7 +1382,7 @@ if ($gettext_flags == 7) {
 /* If we can fake gettext, try that */
 elseif ($gettext_flags == 0) {
     $use_gettext = true;
-    include_once('../functions/gettext.php');
+    include_once(SM_PATH . 'functions/gettext.php');
 } else {
     /* Uh-ho.  A weird install */
     if (! $gettext_flags & 1) {
@@ -938,4 +1402,145 @@ elseif ($gettext_flags == 0) {
     }
 }
 
+function charset_decode_utf8 ($string) {
+/*
+    Every decoded character consists of n bytes. First byte is octal
+    300-375, other bytes - always octals 200-277.
+
+    \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)
+    
+    decoding cycle is unfinished. please test and report problems to tokul@users.sourceforge.net
+*/
+    global $default_charset, $languages, $sm_notAlias;
+
+    if (strtolower($default_charset) == 'utf-8')
+        return $string;
+    if (strtolower($languages[$sm_notAlias]['CHARSET']) == 'utf-8')
+        return $string;
+
+    /* Only do the slow convert if there are 8-bit characters */
+    if (! ereg("[\200-\377]", $string))
+        return $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);
+
+    // decode two byte unicode characters
+    $string = preg_replace("/([\300-\337])([\200-\277])/e",
+    "'&#'.((ord('\\1')-192)*64+(ord('\\2')-128)).';'",
+    $string);
+
+    return $string;
+}
+
+/*
+ * Japanese charset extra function
+ *
+ */
+function japanese_charset_xtra() {
+    $ret = func_get_arg(1);  /* default return value */
+    if (function_exists('mb_detect_encoding')) {
+        switch (func_get_arg(0)) { /* action */
+        case 'decode':
+            $detect_encoding = mb_detect_encoding($ret);
+            if ($detect_encoding == 'JIS' ||
+                $detect_encoding == 'EUC-JP' ||
+                $detect_encoding == 'SJIS') {
+                
+                $ret = mb_convert_encoding($ret, 'EUC-JP', 'AUTO');
+            }
+            break;
+        case 'encode':
+            $detect_encoding = mb_detect_encoding($ret);
+            if ($detect_encoding == 'JIS' ||
+                $detect_encoding == 'EUC-JP' ||
+                $detect_encoding == 'SJIS') {
+                
+                $ret = mb_convert_encoding($ret, 'JIS', 'AUTO');
+            }
+            break;
+        case 'strimwidth':
+            $width = func_get_arg(2);
+            $ret = mb_strimwidth($ret, 0, $width, '...'); 
+            break;
+        case 'encodeheader':
+            $result = '';
+            if (strlen($ret) > 0) {
+                $tmpstr = mb_substr($ret, 0, 1);
+                $prevcsize = strlen($tmpstr);
+                for ($i = 1; $i < mb_strlen($ret); $i++) {
+                    $tmp = mb_substr($ret, $i, 1);
+                    if (strlen($tmp) == $prevcsize) {
+                        $tmpstr .= $tmp;
+                    } else {
+                        if ($prevcsize == 1) {
+                            $result .= $tmpstr;
+                        } else {
+                            $result .= mb_encode_mimeheader($tmpstr);
+                        }
+                        $tmpstr = $tmp;
+                        $prevcsize = strlen($tmp);
+                    }
+                }
+                if (strlen($tmpstr)) {
+                    if (strlen(mb_substr($tmpstr, 0, 1)) == 1)
+                        $result .= $tmpstr;
+                    else
+                        $result .= mb_encode_mimeheader($tmpstr);
+                }
+            }
+            $ret = $result;
+            //$ret = mb_encode_mimeheader($ret);
+            break;
+        case 'decodeheader':
+            $ret = str_replace("\t", "", $ret);
+            if (eregi('=\\?([^?]+)\\?(q|b)\\?([^?]+)\\?=', $ret))
+                $ret = mb_decode_mimeheader($ret);
+            $ret = mb_convert_encoding($ret, 'EUC-JP', 'AUTO');
+            break;
+        case 'downloadfilename':
+            $useragent = func_get_arg(2);
+            if (strstr($useragent, 'Windows') !== false ||
+                strstr($useragent, 'Mac_') !== false) {
+                $ret = mb_convert_encoding($ret, 'SJIS', 'AUTO');
+            } else {
+                $ret = mb_convert_encoding($ret, 'EUC-JP', 'AUTO');
+}
+            break;
+        }
+    }
+    return $ret;
+}
+
+
+/*
+ * Korean charset extra function
+ * Hangul(Korean Character) Attached File Name Fix.
+ */
+function korean_charset_xtra() {
+    
+    $ret = func_get_arg(1);  /* default return value */
+    if (func_get_arg(0) == 'downloadfilename') { /* action */
+        $ret = str_replace("\x0D\x0A", '', $ret);  /* Hanmail's CR/LF Clear */
+        for ($i=0;$i<strlen($ret);$i++) {
+            if ($ret[$i] >= "\xA1" && $ret[$i] <= "\xFE") {   /* 0xA1 - 0XFE are Valid */
+                $i++;
+                continue;
+            } else if (($ret[$i] >= 'a' && $ret[$i] <= 'z') || /* From Original ereg_replace in download.php */
+                       ($ret[$i] >= 'A' && $ret[$i] <= 'Z') ||
+                       ($ret[$i] == '.') || ($ret[$i] == '-')) {
+                continue;
+            } else {
+                $ret[$i] = '_';
+            }
+        }
+
+    }
+
+    return $ret;
+}
+
 ?>