functions/i18n.php

   1 <?php
   2
   3 /**
   4  * i18n.php
   5  *
   6  * Copyright (c) 1999-2003 The SquirrelMail Project Team
   7  * Licensed under the GNU GPL. For full terms see the file COPYING.
   8  *
   9  * This file contains variuos functions that are needed to do
  10  * internationalization of SquirrelMail.
  11  *
  12  * Internally the output character set is used. Other characters are
  13  * encoded using Unicode entities according to HTML 4.0.
  14  *
  15  * $Id$
  16  */
  17
  18 require_once(SM_PATH . 'functions/global.php');
  19
  20 /* Decodes a string to the internal encoding from the given charset */
  21 function charset_decode ($charset, $string) {
  22     global $languages, $squirrelmail_language, $default_charset;
  23
  24     if (isset($languages[$squirrelmail_language]['XTRA_CODE']) &&
  25         function_exists($languages[$squirrelmail_language]['XTRA_CODE'])) {
  26         $string = $languages[$squirrelmail_language]['XTRA_CODE']('decode', $string);
  27     }
  28
  29     $charset = strtolower($charset);
  30
  31     set_my_charset();
  32
  33     // Variables that allow to use functions without function_exist() calls
  34     $use_php_recode=false;
  35     $use_php_iconv=false;
  36
  37     // Don't do conversion if charset is the same.
  38     if ( $charset == strtolower($default_charset) )
  39           return htmlspecialchars($string);
  40
  41     // catch iso-8859-8-i thing
  42     if ( $charset == "iso-8859-8-i" )
  43               $charset = "iso-8859-8";
  44
  45     /*
  46      * Recode converts html special characters automatically if you use
  47      * 'charset..html' decoding. There is no documented way to put -d option
  48      * into php recode function call.
  49      */
  50     if ( $use_php_recode ) {
  51       if ( $default_charset == "utf-8" ) {
  52         // other charsets can be converted to utf-8 without loss.
  53         // and output string is smaller
  54         $string = recode_string($charset . "..utf-8",$string);
  55         return htmlspecialchars($string);
  56       } else {
  57         $string = recode_string($charset . "..html",$string);
  58         // recode does not convert single quote, htmlspecialchars does.
  59         $string = str_replace("'", '&#039;', $string);
  60         return $string;
  61       }
  62     }
  63
  64     // iconv functions does not have html target and can be used only with utf-8
  65     if ( $use_php_iconv && $default_charset=='utf-8') {
  66       $string = iconv($charset,$default_charset,$string);
  67       return htmlspecialchars($string);
  68     }
  69
  70     // If we don't use recode and iconv, we'll do it old way.
  71
  72     /* All HTML special characters are 7 bit and can be replaced first */
  73
  74     $string = htmlspecialchars ($string);
  75
  76     /* controls cpu and memory intensive decoding cycles */
  77     $agresive_decoding = false;
  78
  79     if (ereg('iso-8859-([[:digit:]]+)', $charset, $res)) {
  80         if ($res[1] == '1') {
  81             include_once(SM_PATH . 'functions/decode/iso8859-1.php');
  82             $ret = charset_decode_iso8859_1 ($string);
  83         } else if ($res[1] == '2') {
  84             include_once(SM_PATH . 'functions/decode/iso8859-2.php');
  85             $ret = charset_decode_iso8859_2 ($string);
  86         } else if ($res[1] == '3') {
  87             include_once(SM_PATH . 'functions/decode/iso8859-3.php');
  88             $ret = charset_decode_iso8859_3 ($string);
  89         } else if ($res[1] == '4') {
  90             include_once(SM_PATH . 'functions/decode/iso8859-4.php');
  91             $ret = charset_decode_iso8859_4 ($string);
  92         } else if ($res[1] == '5') {
  93             include_once(SM_PATH . 'functions/decode/iso8859-5.php');
  94             $ret = charset_decode_iso8859_5 ($string);
  95         } else if ($res[1] == '6') {
  96             include_once(SM_PATH . 'functions/decode/iso8859-6.php');
  97             $ret = charset_decode_iso8859_6 ($string);
  98         } else if ($res[1] == '7') {
  99             include_once(SM_PATH . 'functions/decode/iso8859-7.php');
 100             $ret = charset_decode_iso8859_7 ($string);
 101         } else if ($res[1] == '8') {
 102             include_once(SM_PATH . 'functions/decode/iso8859-8.php');
 103             $ret = charset_decode_iso8859_8 ($string);
 104         } else if ($res[1] == '9') {
 105             include_once(SM_PATH . 'functions/decode/iso8859-9.php');
 106             $ret = charset_decode_iso8859_9 ($string);
 107         } else if ($res[1] == '10') {
 108             include_once(SM_PATH . 'functions/decode/iso8859-10.php');
 109             $ret = charset_decode_iso8859_10 ($string);
 110         } else if ($res[1] == '11') {
 111             include_once(SM_PATH . 'functions/decode/iso8859-11.php');
 112             $ret = charset_decode_iso8859_11 ($string);
 113         } else if ($res[1] == '13') {
 114             include_once(SM_PATH . 'functions/decode/iso8859-13.php');
 115             $ret = charset_decode_iso8859_13 ($string);
 116         } else if ($res[1] == '14') {
 117             include_once(SM_PATH . 'functions/decode/iso8859-14.php');
 118             $ret = charset_decode_iso8859_14 ($string);
 119         } else if ($res[1] == '15') {
 120             include_once(SM_PATH . 'functions/decode/iso8859-15.php');
 121             $ret = charset_decode_iso8859_15 ($string);
 122         } else if ($res[1] == '16') {
 123             include_once(SM_PATH . 'functions/decode/iso8859-16.php');
 124             $ret = charset_decode_iso8859_16 ($string);
 125         } else {
 126             $ret = charset_decode_iso_8859_default ($string);
 127         }
 128     } else if ($charset == 'ns_4551-1') {
 129         $ret = charset_decode_ns_4551_1 ($string);
 130     } else if ($charset == 'koi8-r') {
 131         include_once(SM_PATH . 'functions/decode/koi8-r.php');
 132         $ret = charset_decode_koi8r ($string);
 133     } else if ($charset == 'koi8-u') {
 134         include_once(SM_PATH . 'functions/decode/koi8-u.php');
 135         $ret = charset_decode_koi8u ($string);
 136     } else if ($charset == 'windows-1250') {
 137         include_once(SM_PATH . 'functions/decode/cp1250.php');
 138         $ret = charset_decode_cp1250 ($string);
 139     } else if ($charset == 'windows-1251') {
 140         include_once(SM_PATH . 'functions/decode/cp1251.php');
 141         $ret = charset_decode_cp1251 ($string);
 142     } else if ($charset == 'windows-1252') {
 143         include_once(SM_PATH . 'functions/decode/cp1252.php');
 144         $ret = charset_decode_cp1252 ($string);
 145     } else if ($charset == 'windows-1253') {
 146         include_once(SM_PATH . 'functions/decode/cp1253.php');
 147         $ret = charset_decode_cp1253 ($string);
 148     } else if ($charset == 'windows-1254') {
 149         include_once(SM_PATH . 'functions/decode/cp1254.php');
 150         $ret = charset_decode_cp1254 ($string);
 151     } else if ($charset == 'windows-1255') {
 152         include_once(SM_PATH . 'functions/decode/cp1255.php');
 153         $ret = charset_decode_cp1255 ($string);
 154     } else if ($charset == 'windows-1256') {
 155         include_once(SM_PATH . 'functions/decode/cp1256.php');
 156         $ret = charset_decode_cp1256 ($string);
 157     } else if ($charset == 'windows-1257') {
 158         include_once(SM_PATH . 'functions/decode/cp1257.php');
 159         $ret = charset_decode_cp1257 ($string);
 160     } else if ($charset == 'windows-1258') {
 161         include_once(SM_PATH . 'functions/decode/cp1258.php');
 162         $ret = charset_decode_cp1258 ($string);
 163     } else if ($charset == 'big5' and $agresive_decoding ) {
 164         include_once(SM_PATH . 'functions/decode/big5.php');
 165         $ret = charset_decode_big5 ($string);
 166     } else if ($charset == 'gb2312' and $agresive_decoding ) {
 167         include_once(SM_PATH . 'functions/decode/gb2312.php');
 168         $ret = charset_decode_gb2312 ($string);
 169     } else if ($charset == 'utf-8') {
 170         include_once(SM_PATH . 'functions/decode/utf-8.php');
 171         $ret = charset_decode_utf8 ($string);
 172     } else {
 173         $ret = $string;
 174     }
 175     return( $ret );
 176 }
 177
 178
 179 /* Remove all 8 bit characters from all other ISO-8859 character sets */
 180 function charset_decode_iso_8859_default ($string) {
 181     return (strtr($string, "\240\241\242\243\244\245\246\247".
 182                     "\250\251\252\253\254\255\256\257".
 183                     "\260\261\262\263\264\265\266\267".
 184                     "\270\271\272\273\274\275\276\277".
 185                     "\300\301\302\303\304\305\306\307".
 186                     "\310\311\312\313\314\315\316\317".
 187                     "\320\321\322\323\324\325\326\327".
 188                     "\330\331\332\333\334\335\336\337".
 189                     "\340\341\342\343\344\345\346\347".
 190                     "\350\351\352\353\354\355\356\357".
 191                     "\360\361\362\363\364\365\366\367".
 192                     "\370\371\372\373\374\375\376\377",
 193                     "????????????????????????????????????????".
 194                     "????????????????????????????????????????".
 195                     "????????????????????????????????????????".
 196                     "????????"));
 197
 198 }
 199
 200 /*
 201  * This is the same as ISO-646-NO and is used by some
 202  * Microsoft programs when sending Norwegian characters
 203  */
 204 function charset_decode_ns_4551_1 ($string) {
 205     /*
 206      * These characters are:
 207      * Latin capital letter AE
 208      * Latin capital letter O with stroke
 209      * Latin capital letter A with ring above
 210      * and the same as small letters
 211      */
 212     return strtr ($string, "[\\]{|}", "ÆØÅæøå");
 213 }
 214
 215
 216 /*
 217  * Set up the language to be output
 218  * if $do_search is true, then scan the browser information
 219  * for a possible language that we know
 220  */
 221 function set_up_language($sm_language, $do_search = false, $default = false) {
 222
 223     static $SetupAlready = 0;
 224     global $use_gettext, $languages,
 225            $squirrelmail_language, $squirrelmail_default_language,
 226            $sm_notAlias;
 227
 228     if ($SetupAlready) {
 229         return;
 230     }
 231
 232     $SetupAlready = TRUE;
 233     sqgetGlobalVar('HTTP_ACCEPT_LANGUAGE',  $accept_lang, SQ_SERVER);
 234
 235     if ($do_search && ! $sm_language && isset($accept_lang)) {
 236         $sm_language = substr($accept_lang, 0, 2);
 237     }
 238
 239     if ((!$sm_language||$default) && isset($squirrelmail_default_language)) {
 240         $squirrelmail_language = $squirrelmail_default_language;
 241         $sm_language = $squirrelmail_default_language;
 242     }
 243     $sm_notAlias = $sm_language;
 244
 245     // Catching removed translation
 246     // System reverts to English translation if user prefs contain translation
 247     // that is not available in $languages array (admin removed directory
 248     // with that translation)
 249     if (!isset($languages[$sm_notAlias])) {
 250       $sm_notAlias="en_US";
 251     }
 252
 253     while (isset($languages[$sm_notAlias]['ALIAS'])) {
 254         $sm_notAlias = $languages[$sm_notAlias]['ALIAS'];
 255     }
 256
 257     if ( isset($sm_language) &&
 258          $use_gettext &&
 259          $sm_language != '' &&
 260          isset($languages[$sm_notAlias]['CHARSET']) ) {
 261         bindtextdomain( 'squirrelmail', SM_PATH . 'locale/' );
 262         textdomain( 'squirrelmail' );
 263         if (function_exists('bind_textdomain_codeset')) {
 264              bind_textdomain_codeset ("squirrelmail", $languages[$sm_notAlias]['CHARSET'] );
 265         }
 266         if (isset($languages[$sm_notAlias]['LOCALE'])){
 267           $longlocale=$languages[$sm_notAlias]['LOCALE'];
 268         } else {
 269           $longlocale=$sm_notAlias;
 270         }
 271         if ( !ini_get('safe_mode') &&
 272              getenv( 'LC_ALL' ) != $longlocale ) {
 273             putenv( "LC_ALL=$longlocale" );
 274             putenv( "LANG=$longlocale" );
 275             putenv( "LANGUAGE=$longlocale" );
 276         }
 277         setlocale(LC_ALL, $longlocale);
 278         $squirrelmail_language = $sm_notAlias;
 279         if ($squirrelmail_language == 'ja_JP' && function_exists('mb_detect_encoding') ) {
 280             header ('Content-Type: text/html; charset=EUC-JP');
 281             if (!function_exists('mb_internal_encoding')) {
 282                 echo _("You need to have php4 installed with the multibyte string function enabled (using configure option --enable-mbstring).");
 283             }
 284             if (function_exists('mb_language')) {
 285                 mb_language('Japanese');
 286             }
 287             mb_internal_encoding('EUC-JP');
 288             mb_http_output('pass');
 289         } else {
 290         header( 'Content-Type: text/html; charset=' . $languages[$sm_notAlias]['CHARSET'] );
 291     }
 292 }
 293 }
 294
 295 function set_my_charset(){
 296
 297     /*
 298      * There can be a $default_charset setting in the
 299      * config.php file, but the user may have a different language
 300      * selected for a user interface. This function checks the
 301      * language selected by the user and tags the outgoing messages
 302      * with the appropriate charset corresponding to the language
 303      * selection. This is "more right" (tm), than just stamping the
 304      * message blindly with the system-wide $default_charset.
 305      */
 306     global $data_dir, $username, $default_charset, $languages, $squirrelmail_default_language;
 307
 308     $my_language = getPref($data_dir, $username, 'language');
 309     if (!$my_language) {
 310         $my_language = $squirrelmail_default_language ;
 311     }
 312     // Catch removed translation
 313     if (!isset($languages[$my_language])) {
 314       $my_language="en_US";
 315     }
 316     while (isset($languages[$my_language]['ALIAS'])) {
 317         $my_language = $languages[$my_language]['ALIAS'];
 318     }
 319     $my_charset = $languages[$my_language]['CHARSET'];
 320     if ($my_charset) {
 321         $default_charset = $my_charset;
 322     }
 323 }
 324
 325 /* ------------------------------ main --------------------------- */
 326
 327 global $squirrelmail_language, $languages, $use_gettext;
 328
 329 if (! isset($squirrelmail_language)) {
 330     $squirrelmail_language = '';
 331 }
 332
 333 /* This array specifies the available languages. */
 334
 335 if ( file_exists( SM_PATH . 'locale/ca_ES') ) {
 336     // The glibc locale is ca_ES.
 337     $languages['ca_ES']['NAME']    = 'Catalan';
 338     $languages['ca_ES']['CHARSET'] = 'iso-8859-1';
 339     $languages['ca']['ALIAS'] = 'ca_ES';
 340 }
 341
 342 if ( file_exists( SM_PATH . 'locale/cs_CZ') ) {
 343     $languages['cs_CZ']['NAME']    = 'Czech';
 344     $languages['cs_CZ']['CHARSET'] = 'iso-8859-2';
 345     $languages['cs']['ALIAS']      = 'cs_CZ';
 346 }
 347
 348 if ( file_exists( SM_PATH . 'locale/da_DK') ) {
 349     // Danish locale is da_DK.
 350     $languages['da_DK']['NAME']    = 'Danish';
 351     $languages['da_DK']['CHARSET'] = 'iso-8859-1';
 352     $languages['da']['ALIAS'] = 'da_DK';
 353 }
 354
 355 if ( file_exists( SM_PATH . 'locale/de_DE') ) {
 356     $languages['de_DE']['NAME']    = 'Deutsch';
 357     $languages['de_DE']['CHARSET'] = 'iso-8859-1';
 358     $languages['de']['ALIAS'] = 'de_DE';
 359 }
 360
 361 // There is no en_EN! There is en_US, en_BR, en_AU, and so forth,
 362 // but who cares about !US, right? Right? :)
 363
 364 if ( file_exists( SM_PATH . 'locale/el_GR') ) {
 365     $languages['el_GR']['NAME']    = 'Greek';
 366     $languages['el_GR']['CHARSET'] = 'iso-8859-7';
 367     $languages['el']['ALIAS'] = 'el_GR';
 368 }
 369
 370 $languages['en_US']['NAME']    = 'English';
 371 $languages['en_US']['CHARSET'] = 'iso-8859-1';
 372 $languages['en']['ALIAS'] = 'en_US';
 373
 374 if ( file_exists( SM_PATH . 'locale/es_ES') ) {
 375     $languages['es_ES']['NAME']    = 'Spanish';
 376     $languages['es_ES']['CHARSET'] = 'iso-8859-1';
 377     $languages['es']['ALIAS'] = 'es_ES';
 378 }
 379 if ( file_exists( SM_PATH . 'locale/et_EE') ) {
 380     $languages['et_EE']['NAME']    = 'Estonian';
 381     $languages['et_EE']['CHARSET'] = 'iso-8859-15';
 382     $languages['et']['ALIAS'] = 'et_EE';
 383 }
 384 if ( file_exists( SM_PATH . 'locale/fo_FO') ) {
 385     $languages['fo_FO']['NAME']    = 'Faroese';
 386     $languages['fo_FO']['CHARSET'] = 'iso-8859-1';
 387     $languages['fo']['ALIAS'] = 'fo_FO';
 388 }
 389 if ( file_exists( SM_PATH . 'locale/fi_FI') ) {
 390     $languages['fi_FI']['NAME']    = 'Finnish';
 391     $languages['fi_FI']['CHARSET'] = 'iso-8859-1';
 392     $languages['fi']['ALIAS'] = 'fi_FI';
 393 }
 394 if ( file_exists( SM_PATH . 'locale/fr_FR') ) {
 395     $languages['fr_FR']['NAME']    = 'French';
 396     $languages['fr_FR']['CHARSET'] = 'iso-8859-1';
 397     $languages['fr']['ALIAS'] = 'fr_FR';
 398 }
 399 if ( file_exists( SM_PATH . 'locale/hr_HR') ) {
 400     $languages['hr_HR']['NAME']    = 'Croatian';
 401     $languages['hr_HR']['CHARSET'] = 'iso-8859-2';
 402     $languages['hr']['ALIAS'] = 'hr_HR';
 403 }
 404 if ( file_exists( SM_PATH . 'locale/hu_HU') ) {
 405     $languages['hu_HU']['NAME']    = 'Hungarian';
 406     $languages['hu_HU']['CHARSET'] = 'iso-8859-2';
 407     $languages['hu']['ALIAS'] = 'hu_HU';
 408 }
 409 if ( file_exists( SM_PATH . 'locale/id_ID') ) {
 410     $languages['id_ID']['NAME']    = 'Bahasa Indonesia';
 411     $languages['id_ID']['CHARSET'] = 'iso-8859-1';
 412     $languages['id']['ALIAS'] = 'id_ID';
 413 }
 414 if ( file_exists( SM_PATH . 'locale/is_IS') ) {
 415     $languages['is_IS']['NAME']    = 'Icelandic';
 416     $languages['is_IS']['CHARSET'] = 'iso-8859-1';
 417     $languages['is']['ALIAS'] = 'is_IS';
 418 }
 419 if ( file_exists( SM_PATH . 'locale/it_IT') ) {
 420     $languages['it_IT']['NAME']    = 'Italian';
 421     $languages['it_IT']['CHARSET'] = 'iso-8859-1';
 422     $languages['it']['ALIAS'] = 'it_IT';
 423 }
 424 if ( file_exists( SM_PATH . 'locale/ja_JP') ) {
 425     $languages['ja_JP']['NAME']    = 'Japanese';
 426     $languages['ja_JP']['CHARSET'] = 'iso-2022-jp';
 427     $languages['ja_JP']['XTRA_CODE'] = 'japanese_charset_xtra';
 428     $languages['ja']['ALIAS'] = 'ja_JP';
 429 }
 430 if ( file_exists( SM_PATH . 'locale/ko_KR') ) {
 431     $languages['ko_KR']['NAME']    = 'Korean';
 432     $languages['ko_KR']['CHARSET'] = 'euc-KR';
 433     $languages['ko_KR']['XTRA_CODE'] = 'korean_charset_xtra';
 434     $languages['ko']['ALIAS'] = 'ko_KR';
 435 }
 436 if ( file_exists( SM_PATH . 'locale/nl_NL') ) {
 437     $languages['nl_NL']['NAME']    = 'Dutch';
 438     $languages['nl_NL']['CHARSET'] = 'iso-8859-1';
 439     $languages['nl']['ALIAS'] = 'nl_NL';
 440 }
 441 if ( file_exists( SM_PATH . 'locale/no_NO') ) {
 442     $languages['no_NO']['NAME']    = 'Norwegian (Bokm&aring;l)';
 443     $languages['no_NO']['CHARSET'] = 'iso-8859-1';
 444     $languages['no']['ALIAS'] = 'no_NO';
 445 }
 446 if ( file_exists( SM_PATH . 'locale/nn_NO') ) {
 447     $languages['nn_NO']['NAME']    = 'Norwegian (Nynorsk)';
 448     $languages['nn_NO']['CHARSET'] = 'iso-8859-1';
 449 }
 450 if ( file_exists( SM_PATH . 'locale/pl_PL') ) {
 451     $languages['pl_PL']['NAME']    = 'Polish';
 452     $languages['pl_PL']['CHARSET'] = 'iso-8859-2';
 453     $languages['pl']['ALIAS'] = 'pl_PL';
 454 }
 455 if ( file_exists( SM_PATH . 'locale/pt_PT') ) {
 456     $languages['pt_PT']['NAME'] = 'Portuguese (Portugal)';
 457     $languages['pt_PT']['CHARSET'] = 'iso-8859-1';
 458     $languages['pt']['ALIAS'] = 'pt_PT';
 459 }
 460 if ( file_exists( SM_PATH . 'locale/pt_BR') ) {
 461     $languages['pt_BR']['NAME']    = 'Portuguese (Brazil)';
 462     $languages['pt_BR']['CHARSET'] = 'iso-8859-1';
 463 }
 464 if ( file_exists( SM_PATH . 'locale/ru_RU') ) {
 465     $languages['ru_RU']['NAME']    = 'Russian';
 466     $languages['ru_RU']['CHARSET'] = 'koi8-r';
 467     $languages['ru_RU']['LOCALE'] = 'ru_RU.KOI8-R';
 468     $languages['ru']['ALIAS'] = 'ru_RU';
 469 }
 470 if ( file_exists( SM_PATH . 'locale/sr_YU') ) {
 471     $languages['sr_YU']['NAME']    = 'Serbian';
 472     $languages['sr_YU']['CHARSET'] = 'iso-8859-2';
 473     $languages['sr']['ALIAS'] = 'sr_YU';
 474 }
 475 if ( file_exists( SM_PATH . 'locale/sv_SE') ) {
 476     $languages['sv_SE']['NAME']    = 'Swedish';
 477     $languages['sv_SE']['CHARSET'] = 'iso-8859-1';
 478     $languages['sv']['ALIAS'] = 'sv_SE';
 479 }
 480 if ( file_exists( SM_PATH . 'locale/tr_TR') ) {
 481     $languages['tr_TR']['NAME']    = 'Turkish';
 482     $languages['tr_TR']['CHARSET'] = 'iso-8859-9';
 483     $languages['tr']['ALIAS'] = 'tr_TR';
 484 }
 485 if ( file_exists( SM_PATH . 'locale/zh_TW') ) {
 486     $languages['zh_TW']['NAME']    = 'Chinese Trad';
 487     $languages['zh_TW']['CHARSET'] = 'big5';
 488     $languages['tw']['ALIAS'] = 'zh_TW';
 489 }
 490 if ( file_exists( SM_PATH . 'locale/zh_CN') ) {
 491     $languages['zh_CN']['NAME']    = 'Chinese Simp';
 492     $languages['zh_CN']['CHARSET'] = 'gb2312';
 493     $languages['cn']['ALIAS'] = 'zh_CN';
 494 }
 495 if ( file_exists( SM_PATH . 'locale/sk_SK') ) {
 496     $languages['sk_SK']['NAME']     = 'Slovak';
 497     $languages['sk_SK']['CHARSET']  = 'iso-8859-2';
 498     $languages['sk']['ALIAS']       = 'sk_SK';
 499 }
 500 if ( file_exists( SM_PATH . 'locale/ro_RO') ) {
 501     $languages['ro_RO']['NAME']    = 'Romanian';
 502     $languages['ro_RO']['CHARSET'] = 'iso-8859-2';
 503     $languages['ro']['ALIAS'] = 'ro_RO';
 504 }
 505 if ( file_exists( SM_PATH . 'locale/th_TH') ) {
 506     $languages['th_TH']['NAME']    = 'Thai';
 507     $languages['th_TH']['CHARSET'] = 'tis-620';
 508     $languages['th']['ALIAS'] = 'th_TH';
 509 }
 510 if ( file_exists( SM_PATH . 'locale/lt_LT') ) {
 511     $languages['lt_LT']['NAME']    = 'Lithuanian';
 512     $languages['lt_LT']['CHARSET'] = 'iso-8859-4';
 513     $languages['lt_LT']['LOCALE'] = 'lt_LT.ISO-8859-4';
 514     $languages['lt']['ALIAS'] = 'lt_LT';
 515 }
 516 if ( file_exists( SM_PATH . 'locale/sl_SI') ) {
 517     $languages['sl_SI']['NAME']    = 'Slovenian';
 518     $languages['sl_SI']['CHARSET'] = 'iso-8859-2';
 519     $languages['sl']['ALIAS'] = 'sl_SI';
 520 }
 521 if ( file_exists( SM_PATH . 'locale/bg_BG') ) {
 522     $languages['bg_BG']['NAME']    = 'Bulgarian';
 523     $languages['bg_BG']['CHARSET'] = 'windows-1251';
 524     $languages['bg']['ALIAS'] = 'bg_BG';
 525 }
 526 if ( file_exists( SM_PATH . 'locale/uk_UA') ) {
 527     $languages['uk_UA']['NAME']    = 'Ukrainian';
 528     $languages['uk_UA']['CHARSET'] = 'koi8-u';
 529     $languages['uk']['ALIAS'] = 'uk_UA';
 530 }
 531 if ( file_exists( SM_PATH . 'locale/cy_GB') ) {
 532     $languages['cy_GB']['NAME']    = 'Welsh';
 533     $languages['cy_GB']['CHARSET'] = 'iso-8859-1';
 534     $languages['cy']['ALIAS'] = 'cy_GB';
 535 }
 536 if ( file_exists( SM_PATH . 'locale/vi_VN') ) {
 537     $languages['vi_VN']['NAME']    = 'Vietnamese';
 538     $languages['vi_VN']['CHARSET'] = 'utf-8';
 539     $languages['vi']['ALIAS'] = 'vi_VN';
 540 }
 541 // Right to left languages
 542 if ( file_exists( SM_PATH . 'locale/ar') ) {
 543     $languages['ar']['NAME']    = 'Arabic';
 544     $languages['ar']['CHARSET'] = 'windows-1256';
 545     $languages['ar']['DIR']     = 'rtl';
 546 }
 547 if ( file_exists( SM_PATH . 'locale/he_IL') ) {
 548     $languages['he_IL']['NAME']    = 'Hebrew';
 549     $languages['he_IL']['CHARSET'] = 'windows-1255';
 550     $languages['he_IL']['DIR']     = 'rtl';
 551     $languages['he']['ALIAS']      = 'he_IL';
 552 }
 553
 554 /* Detect whether gettext is installed. */
 555 $gettext_flags = 0;
 556 if (function_exists('_')) {
 557     $gettext_flags += 1;
 558 }
 559 if (function_exists('bindtextdomain')) {
 560     $gettext_flags += 2;
 561 }
 562 if (function_exists('textdomain')) {
 563     $gettext_flags += 4;
 564 }
 565
 566 /* If gettext is fully loaded, cool */
 567 if ($gettext_flags == 7) {
 568     $use_gettext = true;
 569 }
 570 /* If we can fake gettext, try that */
 571 elseif ($gettext_flags == 0) {
 572     $use_gettext = true;
 573     include_once(SM_PATH . 'functions/gettext.php');
 574 } else {
 575     /* Uh-ho.  A weird install */
 576     if (! $gettext_flags & 1) {
 577         function _($str) {
 578             return $str;
 579         }
 580     }
 581     if (! $gettext_flags & 2) {
 582         function bindtextdomain() {
 583             return;
 584         }
 585     }
 586     if (! $gettext_flags & 4) {
 587         function textdomain() {
 588             return;
 589         }
 590     }
 591 }
 592
 593
 594 /*
 595  * Japanese charset extra function
 596  *
 597  */
 598 function japanese_charset_xtra() {
 599     $ret = func_get_arg(1);  /* default return value */
 600     if (function_exists('mb_detect_encoding')) {
 601         switch (func_get_arg(0)) { /* action */
 602         case 'decode':
 603             $detect_encoding = @mb_detect_encoding($ret);
 604             if ($detect_encoding == 'JIS' ||
 605                 $detect_encoding == 'EUC-JP' ||
 606                 $detect_encoding == 'SJIS' ||
 607                 $detect_encoding == 'UTF-8') {
 608
 609                 $ret = mb_convert_kana(mb_convert_encoding($ret, 'EUC-JP', 'AUTO'), "KV");
 610             }
 611             break;
 612         case 'encode':
 613             $detect_encoding = @mb_detect_encoding($ret);
 614             if ($detect_encoding == 'JIS' ||
 615                 $detect_encoding == 'EUC-JP' ||
 616                 $detect_encoding == 'SJIS' ||
 617                 $detect_encoding == 'UTF-8') {
 618
 619                 $ret = mb_convert_encoding(mb_convert_kana($ret, "KV"), 'JIS', 'AUTO');
 620             }
 621             break;
 622         case 'strimwidth':
 623             $width = func_get_arg(2);
 624             $ret = mb_strimwidth($ret, 0, $width, '...');
 625             break;
 626         case 'encodeheader':
 627             $result = '';
 628             if (strlen($ret) > 0) {
 629                 $tmpstr = mb_substr($ret, 0, 1);
 630                 $prevcsize = strlen($tmpstr);
 631                 for ($i = 1; $i < mb_strlen($ret); $i++) {
 632                     $tmp = mb_substr($ret, $i, 1);
 633                     if (strlen($tmp) == $prevcsize) {
 634                         $tmpstr .= $tmp;
 635                     } else {
 636                         if ($prevcsize == 1) {
 637                             $result .= $tmpstr;
 638                         } else {
 639                             $result .= str_replace(' ', '',
 640                                                    mb_encode_mimeheader($tmpstr,'iso-2022-jp','B',''));
 641                         }
 642                         $tmpstr = $tmp;
 643                         $prevcsize = strlen($tmp);
 644                     }
 645                 }
 646                 if (strlen($tmpstr)) {
 647                     if (strlen(mb_substr($tmpstr, 0, 1)) == 1)
 648                         $result .= $tmpstr;
 649                     else
 650                         $result .= str_replace(' ', '',
 651                                                mb_encode_mimeheader($tmpstr,'iso-2022-jp','B',''));
 652                 }
 653             }
 654             $ret = $result;
 655             break;
 656         case 'decodeheader':
 657             $ret = str_replace("\t", "", $ret);
 658             if (eregi('=\\?([^?]+)\\?(q|b)\\?([^?]+)\\?=', $ret))
 659                 $ret = @mb_decode_mimeheader($ret);
 660             $ret = @mb_convert_encoding($ret, 'EUC-JP', 'AUTO');
 661             break;
 662         case 'downloadfilename':
 663             $useragent = func_get_arg(2);
 664             if (strstr($useragent, 'Windows') !== false ||
 665                 strstr($useragent, 'Mac_') !== false) {
 666                 $ret = mb_convert_encoding($ret, 'SJIS', 'AUTO');
 667             } else {
 668                 $ret = mb_convert_encoding($ret, 'EUC-JP', 'AUTO');
 669 }
 670             break;
 671         case 'wordwrap':
 672             $no_begin = "\x21\x25\x29\x2c\x2e\x3a\x3b\x3f\x5d\x7d\xa1\xf1\xa1\xeb\xa1" .
 673                 "\xc7\xa1\xc9\xa2\xf3\xa1\xec\xa1\xed\xa1\xee\xa1\xa2\xa1\xa3\xa1\xb9" .
 674                 "\xa1\xd3\xa1\xd5\xa1\xd7\xa1\xd9\xa1\xdb\xa1\xcd\xa4\xa1\xa4\xa3\xa4" .
 675                 "\xa5\xa4\xa7\xa4\xa9\xa4\xc3\xa4\xe3\xa4\xe5\xa4\xe7\xa4\xee\xa1\xab" .
 676                 "\xa1\xac\xa1\xb5\xa1\xb6\xa5\xa1\xa5\xa3\xa5\xa5\xa5\xa7\xa5\xa9\xa5" .
 677                 "\xc3\xa5\xe3\xa5\xe5\xa5\xe7\xa5\xee\xa5\xf5\xa5\xf6\xa1\xa6\xa1\xbc" .
 678                 "\xa1\xb3\xa1\xb4\xa1\xaa\xa1\xf3\xa1\xcb\xa1\xa4\xa1\xa5\xa1\xa7\xa1" .
 679                 "\xa8\xa1\xa9\xa1\xcf\xa1\xd1";
 680             $no_end = "\x5c\x24\x28\x5b\x7b\xa1\xf2\x5c\xa1\xc6\xa1\xc8\xa1\xd2\xa1" .
 681                 "\xd4\xa1\xd6\xa1\xd8\xa1\xda\xa1\xcc\xa1\xf0\xa1\xca\xa1\xce\xa1\xd0\xa1\xef";
 682             $wrap = func_get_arg(2);
 683
 684             if (strlen($ret) >= $wrap &&
 685                 substr($ret, 0, 1) != '>' &&
 686                 strpos($ret, 'http://') === FALSE &&
 687                 strpos($ret, 'https://') === FALSE &&
 688                 strpos($ret, 'ftp://') === FALSE) {
 689
 690                 $ret = mb_convert_kana($ret, "KV");
 691
 692                 $line_new = '';
 693                 $ptr = 0;
 694
 695                 while ($ptr < strlen($ret) - 1) {
 696                     $l = mb_strcut($ret, $ptr, $wrap);
 697                     $ptr += strlen($l);
 698                     $tmp = $l;
 699
 700                     $l = mb_strcut($ret, $ptr, 2);
 701                     while (strlen($l) != 0 && mb_strpos($no_begin, $l) !== FALSE ) {
 702                         $tmp .= $l;
 703                         $ptr += strlen($l);
 704                         $l = mb_strcut($ret, $ptr, 1);
 705                     }
 706                     $line_new .= $tmp;
 707                     if ($ptr < strlen($ret) - 1)
 708                         $line_new .= "\n";
 709                 }
 710                 $ret = $line_new;
 711             }
 712             break;
 713         case 'utf7-imap_encode':
 714             $ret = mb_convert_encoding($ret, 'UTF7-IMAP', 'EUC-JP');
 715             break;
 716         case 'utf7-imap_decode':
 717             $ret = mb_convert_encoding($ret, 'EUC-JP', 'UTF7-IMAP');
 718             break;
 719         }
 720     }
 721     return $ret;
 722 }
 723
 724
 725 /*
 726  * Korean charset extra function
 727  * Hangul(Korean Character) Attached File Name Fix.
 728  */
 729 function korean_charset_xtra() {
 730
 731     $ret = func_get_arg(1);  /* default return value */
 732     if (func_get_arg(0) == 'downloadfilename') { /* action */
 733         $ret = str_replace("\x0D\x0A", '', $ret);  /* Hanmail's CR/LF Clear */
 734         for ($i=0;$i<strlen($ret);$i++) {
 735             if ($ret[$i] >= "\xA1" && $ret[$i] <= "\xFE") {   /* 0xA1 - 0XFE are Valid */
 736                 $i++;
 737                 continue;
 738             } else if (($ret[$i] >= 'a' && $ret[$i] <= 'z') || /* From Original ereg_replace in download.php */
 739                        ($ret[$i] >= 'A' && $ret[$i] <= 'Z') ||
 740                        ($ret[$i] == '.') || ($ret[$i] == '-')) {
 741                 continue;
 742             } else {
 743                 $ret[$i] = '_';
 744             }
 745         }
 746
 747     }
 748
 749     return $ret;
 750 }
 751
 752 /*
 753  * This function can be used to replace non-braking space symbols
 754  * that are inserted in forms by some browsers instead of normal
 755  * space symbol.
 756  */
 757 function cleanup_nbsp($string,$charset) {
 758
 759   // reduce number of case statements
 760   if (stristr('iso-8859-',substr($charset,0,9))){
 761     $output_charset="iso-8859-x";
 762   }
 763   if (stristr('windows-125',substr($charset,0,11))){
 764     $output_charset="cp125x";
 765   }
 766   if (stristr('koi8',substr($charset,0,4))){
 767     $output_charset="koi8-x";
 768   }
 769   if (! isset($output_charset)){
 770     $output_charset=strtolower($charset);
 771   }
 772
 773 // where is non-braking space symbol
 774 switch($output_charset):
 775  case "iso-8859-x":
 776  case "cp125x":
 777  case "iso-2022-jp":
 778   $nbsp="\xA0";
 779   break;
 780  case "koi8-x":
 781    $nbsp="\x9A";
 782    break;
 783  case "utf-8":
 784    $nbsp="\xC2\xA0";
 785    break;
 786  default:
 787    // don't change string if charset is unmatched
 788    return $string;
 789 endswitch;
 790
 791 // return space instead of non-braking space.
 792  return str_replace($nbsp,' ',$string);
 793 }
 794 ?>