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  * @package squirrelmail
  17  */
  18
  19 /** Everything uses global.php... */
  20 require_once(SM_PATH . 'functions/global.php');
  21
  22 /**
  23  * Converts string from given charset to charset, that can be displayed by user translation.
  24  *
  25  * Function by default returns html encoded strings, if translation uses different encoding.
  26  * If Japanese translation is used - function returns string converted to euc-jp
  27  * If iconv or recode functions are enabled and translation uses utf-8 - function returns utf-8 encoded string.
  28  * If $charset is not supported - function returns unconverted string.
  29  *
  30  * sanitizing of html tags is also done by this function.
  31  *
  32  * @param string $charset
  33  * @param string $string Text to be decoded
  34  * @return string decoded string
  35  */
  36 function charset_decode ($charset, $string) {
  37     global $languages, $squirrelmail_language, $default_charset;
  38     global $use_php_recode, $use_php_iconv, $agresive_decoding;
  39
  40     if (isset($languages[$squirrelmail_language]['XTRA_CODE']) &&
  41         function_exists($languages[$squirrelmail_language]['XTRA_CODE'])) {
  42         $string = $languages[$squirrelmail_language]['XTRA_CODE']('decode', $string);
  43     }
  44
  45     $charset = strtolower($charset);
  46
  47     set_my_charset();
  48
  49     // Variables that allow to use functions without function_exist() calls
  50     if (! isset($use_php_recode) || $use_php_recode=="" ) {
  51              $use_php_recode=false; }
  52     if (! isset($use_php_iconv) || $use_php_iconv=="" ) {
  53          $use_php_iconv=false; }
  54
  55     // Don't do conversion if charset is the same.
  56     if ( $charset == strtolower($default_charset) )
  57           return htmlspecialchars($string);
  58
  59     // catch iso-8859-8-i thing
  60     if ( $charset == "iso-8859-8-i" )
  61               $charset = "iso-8859-8";
  62
  63     /*
  64      * Recode converts html special characters automatically if you use
  65      * 'charset..html' decoding. There is no documented way to put -d option
  66      * into php recode function call.
  67      */
  68     if ( $use_php_recode ) {
  69       if ( $default_charset == "utf-8" ) {
  70         // other charsets can be converted to utf-8 without loss.
  71         // and output string is smaller
  72         $string = recode_string($charset . "..utf-8",$string);
  73         return htmlspecialchars($string);
  74       } else {
  75         $string = recode_string($charset . "..html",$string);
  76         // recode does not convert single quote, htmlspecialchars does.
  77         $string = str_replace("'", '&#039;', $string);
  78         return $string;
  79       }
  80     }
  81
  82     // iconv functions does not have html target and can be used only with utf-8
  83     if ( $use_php_iconv && $default_charset=='utf-8') {
  84       $string = iconv($charset,$default_charset,$string);
  85       return htmlspecialchars($string);
  86     }
  87
  88     // If we don't use recode and iconv, we'll do it old way.
  89
  90     /* All HTML special characters are 7 bit and can be replaced first */
  91
  92     $string = htmlspecialchars ($string);
  93
  94     /* controls cpu and memory intensive decoding cycles */
  95     if (! isset($agresive_decoding) || $agresive_decoding=="" ) {
  96          $agresive_decoding=false; }
  97
  98     $decode=fixcharset($charset);
  99     $decodefile=SM_PATH . 'functions/decode/' . $decode . '.php';
 100     if (file_exists($decodefile)) {
 101         include_once($decodefile);
 102         $ret = call_user_func('charset_decode_'.$decode, $string);
 103     } else {
 104         $ret = $string;
 105     }
 106     return( $ret );
 107 }
 108
 109 /**
 110  * Makes charset name suitable for decoding cycles
 111  *
 112  * @param string $charset Name of charset
 113  * @return string $charset Adjusted name of charset
 114  */
 115 function fixcharset($charset) {
 116     // minus removed from function names
 117     $charset=str_replace('-','_',$charset);
 118
 119     // windows-125x and cp125x charsets
 120     $charset=str_replace('windows_','cp',$charset);
 121
 122     // ibm > cp
 123     $charset=str_replace('ibm','cp',$charset);
 124
 125     // iso-8859-8-i -> iso-8859-8
 126     // use same cycle until I'll find differences
 127     $charset=str_replace('iso_8859_8_i','iso_8859_8',$charset);
 128
 129     return $charset;
 130 }
 131
 132 /**
 133  * 8bit cleanup functions.
 134  *
 135  * Replaces all 8 bit characters from ISO-8859 character sets with '?'
 136  * Legacy function used for unsupported ISO-8859 charsets
 137  *
 138  * @param string $string string that has to be cleaned
 139  * @return string cleaned string
 140  */
 141 function charset_decode_iso_8859_default ($string) {
 142     return (strtr($string, "\240\241\242\243\244\245\246\247".
 143                     "\250\251\252\253\254\255\256\257".
 144                     "\260\261\262\263\264\265\266\267".
 145                     "\270\271\272\273\274\275\276\277".
 146                     "\300\301\302\303\304\305\306\307".
 147                     "\310\311\312\313\314\315\316\317".
 148                     "\320\321\322\323\324\325\326\327".
 149                     "\330\331\332\333\334\335\336\337".
 150                     "\340\341\342\343\344\345\346\347".
 151                     "\350\351\352\353\354\355\356\357".
 152                     "\360\361\362\363\364\365\366\367".
 153                     "\370\371\372\373\374\375\376\377",
 154                     "????????????????????????????????????????".
 155                     "????????????????????????????????????????".
 156                     "????????????????????????????????????????".
 157                     "????????"));
 158
 159 }
 160
 161 /**
 162  * ns_4551_1 decoding function
 163  *
 164  * This is the same as ISO-646-NO and is used by some
 165  * Microsoft programs when sending Norwegian characters
 166  *
 167  * @param string $string
 168  * @return string
 169  */
 170 function charset_decode_ns_4551_1 ($string) {
 171     /*
 172      * These characters are:
 173      * Latin capital letter AE
 174      * Latin capital letter O with stroke
 175      * Latin capital letter A with ring above
 176      * and the same as small letters
 177      */
 178     return strtr ($string, "[\\]{|}", "ÆØÅæøå");
 179 }
 180
 181
 182 /**
 183  * Set up the language to be output
 184  * if $do_search is true, then scan the browser information
 185  * for a possible language that we know
 186  *
 187  * Function sets system locale environment (LC_ALL, LANG, LANGUAGE),
 188  * gettext translation bindings and html header information.
 189  *
 190  * Function returns error codes, if there is some fatal error.
 191  *  0 = no error,
 192  *  1 = mbstring support is not present,
 193  *  2 = mbstring support is not present, user's translation reverted to en_US.
 194  *
 195  * @param string $sm_language translation used by user's interface
 196  * @param bool $do_search use browser's preferred language detection functions. Defaults to false.
 197  * @param bool $default set $sm_language to $squirrelmail_default_language if language detection fails or language is not set. Defaults to false.
 198  * @return int function execution error codes.
 199  */
 200 function set_up_language($sm_language, $do_search = false, $default = false) {
 201
 202     static $SetupAlready = 0;
 203     global $use_gettext, $languages,
 204            $squirrelmail_language, $squirrelmail_default_language,
 205            $sm_notAlias, $username, $data_dir;
 206
 207     if ($SetupAlready) {
 208         return;
 209     }
 210
 211     $SetupAlready = TRUE;
 212     sqgetGlobalVar('HTTP_ACCEPT_LANGUAGE',  $accept_lang, SQ_SERVER);
 213
 214     if ($do_search && ! $sm_language && isset($accept_lang)) {
 215         $sm_language = substr($accept_lang, 0, 2);
 216     }
 217
 218     if ((!$sm_language||$default) && isset($squirrelmail_default_language)) {
 219         $squirrelmail_language = $squirrelmail_default_language;
 220         $sm_language = $squirrelmail_default_language;
 221     }
 222     $sm_notAlias = $sm_language;
 223
 224     // Catching removed translation
 225     // System reverts to English translation if user prefs contain translation
 226     // that is not available in $languages array (admin removed directory
 227     // with that translation)
 228     if (!isset($languages[$sm_notAlias])) {
 229       $sm_notAlias="en_US";
 230     }
 231
 232     while (isset($languages[$sm_notAlias]['ALIAS'])) {
 233         $sm_notAlias = $languages[$sm_notAlias]['ALIAS'];
 234     }
 235
 236     if ( isset($sm_language) &&
 237          $use_gettext &&
 238          $sm_language != '' &&
 239          isset($languages[$sm_notAlias]['CHARSET']) ) {
 240         bindtextdomain( 'squirrelmail', SM_PATH . 'locale/' );
 241         textdomain( 'squirrelmail' );
 242         if (function_exists('bind_textdomain_codeset')) {
 243             if ($sm_notAlias == 'ja_JP') {
 244                 bind_textdomain_codeset ("squirrelmail", 'EUC-JP');
 245             } else {
 246                 bind_textdomain_codeset ("squirrelmail", $languages[$sm_notAlias]['CHARSET'] );
 247             }
 248         }
 249         if (isset($languages[$sm_notAlias]['LOCALE'])){
 250           $longlocale=$languages[$sm_notAlias]['LOCALE'];
 251         } else {
 252           $longlocale=$sm_notAlias;
 253         }
 254         if ( !ini_get('safe_mode') &&
 255              getenv( 'LC_ALL' ) != $longlocale ) {
 256             putenv( "LC_ALL=$longlocale" );
 257             putenv( "LANG=$longlocale" );
 258             putenv( "LANGUAGE=$longlocale" );
 259         }
 260         setlocale(LC_ALL, $longlocale);
 261
 262         // Set text direction/alignment variables
 263         if (isset($languages[$sm_notAlias]['DIR']) &&
 264                 $languages[$sm_notAlias]['DIR'] == 'rtl') {
 265             $text_direction='rtl';
 266             $left_align='right';
 267             $right_align='left';
 268         } else {
 269             $text_direction='ltr';
 270             $left_align='left';
 271             $right_align='right';
 272         }
 273
 274         $squirrelmail_language = $sm_notAlias;
 275         if ($squirrelmail_language == 'ja_JP') {
 276             header ('Content-Type: text/html; charset=EUC-JP');
 277             if (!function_exists('mb_internal_encoding')) {
 278                 // Error messages can't be displayed here
 279                 $error = 1;
 280                 // Revert to English if possible.
 281                 if (function_exists('setPref')  && $username!='' && $data_dir!="") {
 282                     setPref($data_dir, $username, 'language', "en_US");
 283                     $error = 2;
 284                 }
 285                 // stop further execution in order not to get php errors on mb_internal_encoding().
 286                 return $error;
 287             }
 288             if (function_exists('mb_language')) {
 289                 mb_language('Japanese');
 290             }
 291             mb_internal_encoding('EUC-JP');
 292             mb_http_output('pass');
 293         } else {
 294         header( 'Content-Type: text/html; charset=' . $languages[$sm_notAlias]['CHARSET'] );
 295     }
 296 }
 297     return 0;
 298 }
 299
 300 /**
 301  * Sets default_charset variable according to the one that is used by user's translations.
 302  *
 303  * Function changes global $default_charset variable in order to be sure, that it
 304  * contains charset used by user's translation. Sanity of $squirrelmail_default_language
 305  * and $default_charset combination provided in SquirrelMail config is also tested.
 306  *
 307  * There can be a $default_charset setting in the
 308  * config.php file, but the user may have a different language
 309  * selected for a user interface. This function checks the
 310  * language selected by the user and tags the outgoing messages
 311  * with the appropriate charset corresponding to the language
 312  * selection. This is "more right" (tm), than just stamping the
 313  * message blindly with the system-wide $default_charset.
 314  */
 315 function set_my_charset(){
 316     global $data_dir, $username, $default_charset, $languages, $squirrelmail_default_language;
 317
 318     $my_language = getPref($data_dir, $username, 'language');
 319     if (!$my_language) {
 320         $my_language = $squirrelmail_default_language ;
 321     }
 322     // Catch removed translation
 323     if (!isset($languages[$my_language])) {
 324       $my_language="en_US";
 325     }
 326     while (isset($languages[$my_language]['ALIAS'])) {
 327         $my_language = $languages[$my_language]['ALIAS'];
 328     }
 329     $my_charset = $languages[$my_language]['CHARSET'];
 330     if ($my_charset) {
 331         $default_charset = $my_charset;
 332     }
 333 }
 334
 335 /* ------------------------------ main --------------------------- */
 336
 337 global $squirrelmail_language, $languages, $use_gettext;
 338
 339 if (! isset($squirrelmail_language)) {
 340     $squirrelmail_language = '';
 341 }
 342
 343 /**
 344  * Array specifies the available translations.
 345  *
 346  * Structure of array:
 347  * $languages['language']['variable'] = 'value'
 348  *
 349  * Possible 'variable' names:
 350  *  NAME      - Translation name in English
 351  *  CHARSET   - Encoding used by translation
 352  *  ALIAS     - used when 'language' is only short name and 'value' should provide long language name
 353  *  ALTNAME   - Native translation name. Any 8bit symbols must be html encoded.
 354  *  LOCALE    - Full locale name (in xx_XX.charset format)
 355  *  DIR       - Text direction. Used to define Right-to-Left languages. Possible values 'rtl' or 'ltr'. If undefined - defaults to 'ltr'
 356  *  XTRA_CODE - translation uses special functions. 'value' provides name of that extra function
 357  *
 358  * Each 'language' definition requires NAME+CHARSET or ALIAS variables.
 359  *
 360  * @name $languages
 361  * @global $languages
 362  */
 363 $languages['bg_BG']['NAME']    = 'Bulgarian';
 364 $languages['bg_BG']['ALTNAME'] = '&#1041;&#1098;&#1083;&#1075;&#1072;&#1088;&#1089;&#1082;&#1080;';
 365 $languages['bg_BG']['CHARSET'] = 'windows-1251';
 366 $languages['bg']['ALIAS'] = 'bg_BG';
 367
 368 $languages['ca_ES']['NAME']    = 'Catalan';
 369 $languages['ca_ES']['CHARSET'] = 'iso-8859-1';
 370 $languages['ca']['ALIAS'] = 'ca_ES';
 371
 372 $languages['cs_CZ']['NAME']    = 'Czech';
 373 $languages['cs_CZ']['ALTNAME'] = '&#268;e&scaron;tina';
 374 $languages['cs_CZ']['CHARSET'] = 'iso-8859-2';
 375 $languages['cs']['ALIAS']      = 'cs_CZ';
 376
 377 $languages['cy_GB']['NAME']    = 'Welsh';
 378 $languages['cy_GB']['ALTNAME'] = 'Cymraeg';
 379 $languages['cy_GB']['CHARSET'] = 'iso-8859-1';
 380 $languages['cy']['ALIAS'] = 'cy_GB';
 381
 382 // Danish locale is da_DK.
 383 $languages['da_DK']['NAME']    = 'Danish';
 384 $languages['da_DK']['ALTNAME'] = 'Dansk';
 385 $languages['da_DK']['CHARSET'] = 'iso-8859-1';
 386 $languages['da']['ALIAS'] = 'da_DK';
 387
 388 $languages['de_DE']['NAME']    = 'German';
 389 $languages['de_DE']['ALTNAME']    = 'Deutsch';
 390 $languages['de_DE']['CHARSET'] = 'iso-8859-1';
 391 $languages['de']['ALIAS'] = 'de_DE';
 392
 393 $languages['el_GR']['NAME']    = 'Greek';
 394 $languages['el_GR']['ALTNAME'] = '&Epsilon;&lambda;&lambda;&eta;&nu;&iota;&kappa;&#940;';
 395 $languages['el_GR']['CHARSET'] = 'iso-8859-7';
 396 $languages['el']['ALIAS'] = 'el_GR';
 397
 398 // There is no en_EN! There is en_US, en_BR, en_AU, and so forth,
 399 // but who cares about !US, right? Right? :)
 400 $languages['en_US']['NAME']    = 'English';
 401 $languages['en_US']['CHARSET'] = 'iso-8859-1';
 402 $languages['en']['ALIAS'] = 'en_US';
 403
 404 $languages['es_ES']['NAME']    = 'Spanish';
 405 $languages['es_ES']['ALTNAME'] = 'Espa&ntilde;ol';
 406 $languages['es_ES']['CHARSET'] = 'iso-8859-1';
 407 $languages['es']['ALIAS'] = 'es_ES';
 408
 409 $languages['et_EE']['NAME']    = 'Estonian';
 410 $languages['et_EE']['CHARSET'] = 'iso-8859-15';
 411 $languages['et']['ALIAS'] = 'et_EE';
 412
 413 $languages['fo_FO']['NAME']    = 'Faroese';
 414 $languages['fo_FO']['CHARSET'] = 'iso-8859-1';
 415 $languages['fo']['ALIAS'] = 'fo_FO';
 416
 417 $languages['fi_FI']['NAME']    = 'Finnish';
 418 $languages['fi_FI']['ALTNAME'] = 'Suomi';
 419 $languages['fi_FI']['CHARSET'] = 'iso-8859-1';
 420 $languages['fi']['ALIAS'] = 'fi_FI';
 421
 422 $languages['fr_FR']['NAME']    = 'French';
 423 $languages['fr_FR']['ALTNAME'] = 'Fran&#231;ais';
 424 $languages['fr_FR']['CHARSET'] = 'iso-8859-1';
 425 $languages['fr']['ALIAS'] = 'fr_FR';
 426
 427 $languages['hr_HR']['NAME']    = 'Croatian';
 428 $languages['hr_HR']['CHARSET'] = 'iso-8859-2';
 429 $languages['hr']['ALIAS'] = 'hr_HR';
 430
 431 $languages['hu_HU']['NAME']    = 'Hungarian';
 432 $languages['hu_HU']['ALTNAME'] = 'Magyar';
 433 $languages['hu_HU']['CHARSET'] = 'iso-8859-2';
 434 $languages['hu']['ALIAS'] = 'hu_HU';
 435
 436 $languages['id_ID']['NAME']    = 'Indonesian';
 437 $languages['id_ID']['ALTNAME'] = 'Bahasa Indonesia';
 438 $languages['id_ID']['CHARSET'] = 'iso-8859-1';
 439 $languages['id']['ALIAS'] = 'id_ID';
 440
 441 $languages['is_IS']['NAME']    = 'Icelandic';
 442 $languages['is_IS']['ALTNAME'] = '&Iacute;slenska';
 443 $languages['is_IS']['CHARSET'] = 'iso-8859-1';
 444 $languages['is']['ALIAS'] = 'is_IS';
 445
 446 $languages['it_IT']['NAME']    = 'Italian';
 447 $languages['it_IT']['CHARSET'] = 'iso-8859-1';
 448 $languages['it']['ALIAS'] = 'it_IT';
 449
 450 $languages['ja_JP']['NAME']    = 'Japanese';
 451 $languages['ja_JP']['ALTNAME'] = '&#26085;&#26412;&#35486;';
 452 $languages['ja_JP']['CHARSET'] = 'iso-2022-jp';
 453 $languages['ja_JP']['LOCALE'] = 'ja_JP.EUC-JP';
 454 $languages['ja_JP']['XTRA_CODE'] = 'japanese_charset_xtra';
 455 $languages['ja']['ALIAS'] = 'ja_JP';
 456
 457 $languages['ko_KR']['NAME']    = 'Korean';
 458 $languages['ko_KR']['CHARSET'] = 'euc-KR';
 459 $languages['ko_KR']['XTRA_CODE'] = 'korean_charset_xtra';
 460 $languages['ko']['ALIAS'] = 'ko_KR';
 461
 462 $languages['lt_LT']['NAME']    = 'Lithuanian';
 463 $languages['lt_LT']['ALTNAME'] = 'Lietuvi&#371;';
 464 $languages['lt_LT']['CHARSET'] = 'utf-8';
 465 $languages['lt_LT']['LOCALE'] = 'lt_LT.UTF-8';
 466 $languages['lt']['ALIAS'] = 'lt_LT';
 467
 468 $languages['nl_NL']['NAME']    = 'Dutch';
 469 $languages['nl_NL']['ALTNAME'] = 'Nederlands';
 470 $languages['nl_NL']['CHARSET'] = 'iso-8859-1';
 471 $languages['nl']['ALIAS'] = 'nl_NL';
 472
 473 $languages['ms_MY']['NAME']    = 'Malay';
 474 $languages['ms_MY']['ALTNAME'] = 'Bahasa Melayu';
 475 $languages['ms_MY']['CHARSET'] = 'iso-8859-1';
 476 $languages['my']['ALIAS'] = 'ms_MY';
 477
 478 $languages['no_NO']['NAME']    = 'Norwegian (Bokm&aring;l)';
 479 $languages['no_NO']['ALTNAME'] = 'Norsk (Bokm&aring;l)';
 480 $languages['no_NO']['CHARSET'] = 'iso-8859-1';
 481 $languages['no']['ALIAS'] = 'no_NO';
 482
 483 $languages['nn_NO']['NAME']    = 'Norwegian (Nynorsk)';
 484 $languages['nn_NO']['ALTNAME'] = 'Norsk (Nynorsk)';
 485 $languages['nn_NO']['CHARSET'] = 'iso-8859-1';
 486
 487 $languages['pl_PL']['NAME']    = 'Polish';
 488 $languages['pl_PL']['ALTNAME'] = 'Polski';
 489 $languages['pl_PL']['CHARSET'] = 'iso-8859-2';
 490 $languages['pl']['ALIAS'] = 'pl_PL';
 491
 492 $languages['pt_PT']['NAME'] = 'Portuguese (Portugal)';
 493 $languages['pt_PT']['CHARSET'] = 'iso-8859-1';
 494 $languages['pt']['ALIAS'] = 'pt_PT';
 495
 496 $languages['pt_BR']['NAME']    = 'Portuguese (Brazil)';
 497 $languages['pt_BR']['ALTNAME'] = 'Portugu&ecirc;s do Brasil';
 498 $languages['pt_BR']['CHARSET'] = 'iso-8859-1';
 499
 500 $languages['ro_RO']['NAME']    = 'Romanian';
 501 $languages['ro_RO']['ALTNAME'] = 'Rom&acirc;n&#259;';
 502 $languages['ro_RO']['CHARSET'] = 'iso-8859-2';
 503 $languages['ro']['ALIAS'] = 'ro_RO';
 504
 505 $languages['ru_RU']['NAME']    = 'Russian';
 506 $languages['ru_RU']['ALTNAME'] = '&#1056;&#1091;&#1089;&#1089;&#1082;&#1080;&#1081;';
 507 $languages['ru_RU']['CHARSET'] = 'utf-8';
 508 $languages['ru_RU']['LOCALE'] = 'ru_RU.UTF-8';
 509 $languages['ru']['ALIAS'] = 'ru_RU';
 510
 511 $languages['sk_SK']['NAME']     = 'Slovak';
 512 $languages['sk_SK']['CHARSET']  = 'iso-8859-2';
 513 $languages['sk']['ALIAS']       = 'sk_SK';
 514
 515 $languages['sl_SI']['NAME']    = 'Slovenian';
 516 $languages['sl_SI']['ALTNAME'] = 'Sloven&scaron;&#269;ina';
 517 $languages['sl_SI']['CHARSET'] = 'iso-8859-2';
 518 $languages['sl']['ALIAS'] = 'sl_SI';
 519
 520 $languages['sr_YU']['NAME']    = 'Serbian';
 521 $languages['sr_YU']['ALTNAME'] = 'Srpski';
 522 $languages['sr_YU']['CHARSET'] = 'iso-8859-2';
 523 $languages['sr']['ALIAS'] = 'sr_YU';
 524
 525 $languages['sv_SE']['NAME']    = 'Swedish';
 526 $languages['sv_SE']['ALTNAME'] = 'Svenska';
 527 $languages['sv_SE']['CHARSET'] = 'iso-8859-1';
 528 $languages['sv']['ALIAS'] = 'sv_SE';
 529
 530 $languages['th_TH']['NAME']    = 'Thai';
 531 $languages['th_TH']['CHARSET'] = 'tis-620';
 532 $languages['th']['ALIAS'] = 'th_TH';
 533
 534 $languages['tr_TR']['NAME']    = 'Turkish';
 535 $languages['tr_TR']['CHARSET'] = 'iso-8859-9';
 536 $languages['tr']['ALIAS'] = 'tr_TR';
 537
 538 $languages['zh_TW']['NAME']    = 'Chinese Trad';
 539 $languages['zh_TW']['CHARSET'] = 'big5';
 540 $languages['tw']['ALIAS'] = 'zh_TW';
 541
 542 $languages['zh_CN']['NAME']    = 'Chinese Simp';
 543 $languages['zh_CN']['CHARSET'] = 'gb2312';
 544 $languages['cn']['ALIAS'] = 'zh_CN';
 545
 546 /*
 547 $languages['uk_UA']['NAME']    = 'Ukrainian';
 548 $languages['uk_UA']['CHARSET'] = 'koi8-u';
 549 $languages['uk']['ALIAS'] = 'uk_UA';
 550 */
 551 /*
 552 if ( file_exists( SM_PATH . 'locale/vi_VN') ) {
 553     $languages['vi_VN']['NAME']    = 'Vietnamese';
 554     $languages['vi_VN']['CHARSET'] = 'utf-8';
 555     $languages['vi']['ALIAS'] = 'vi_VN';
 556 }
 557 */
 558
 559 // Right to left languages
 560 $languages['ar']['NAME']    = 'Arabic';
 561 $languages['ar']['CHARSET'] = 'windows-1256';
 562 $languages['ar']['DIR']     = 'rtl';
 563
 564 $languages['he_IL']['NAME']    = 'Hebrew';
 565 $languages['he_IL']['CHARSET'] = 'windows-1255';
 566 $languages['he_IL']['DIR']     = 'rtl';
 567 $languages['he']['ALIAS']      = 'he_IL';
 568
 569 /* Detect whether gettext is installed. */
 570 $gettext_flags = 0;
 571 if (function_exists('_')) {
 572     $gettext_flags += 1;
 573 }
 574 if (function_exists('bindtextdomain')) {
 575     $gettext_flags += 2;
 576 }
 577 if (function_exists('textdomain')) {
 578     $gettext_flags += 4;
 579 }
 580
 581 /* If gettext is fully loaded, cool */
 582 if ($gettext_flags == 7) {
 583     $use_gettext = true;
 584 }
 585 /* If we can fake gettext, try that */
 586 elseif ($gettext_flags == 0) {
 587     $use_gettext = true;
 588     include_once(SM_PATH . 'functions/gettext.php');
 589 } else {
 590     /* Uh-ho.  A weird install */
 591     if (! $gettext_flags & 1) {
 592         function _($str) {
 593             return $str;
 594         }
 595     }
 596     if (! $gettext_flags & 2) {
 597         function bindtextdomain() {
 598             return;
 599         }
 600     }
 601     if (! $gettext_flags & 4) {
 602         function textdomain() {
 603             return;
 604         }
 605     }
 606 }
 607
 608
 609 /**
 610  * Japanese charset extra function
 611  *
 612  * Action performed by function is defined by first argument.
 613  * Default return value is defined by second argument.
 614  * Use of third argument depends on action.
 615  *
 616  * @param string action performed by this function.
 617  *    possible values:
 618  *      decode - convert returned string to euc-jp. third argument unused
 619  *      encode - convert returned string to jis. third argument unused
 620  *      strimwidth - third argument=$width. trims string to $width symbols.
 621  *      encodeheader - create base64 encoded header in iso-2022-jp. third argument unused
 622  *      decodeheader - return human readable string from mime header. string is returned in euc-jp. third argument unused
 623  *      downloadfilename - third argument $useragent. Arguments provide browser info. Returns shift-jis or euc-jp encoded file name
 624  *      wordwrap - third argument=$wrap. wraps text at $wrap symbols
 625  *      utf7-imap_encode - returns string converted from euc-jp to utf7-imap. third argument unused
 626  *      utf7-imap_decode - returns string converted from utf7-imap to euc-jp. third argument unused
 627  * @param string default return value
 628  */
 629 function japanese_charset_xtra() {
 630     $ret = func_get_arg(1);  /* default return value */
 631     if (function_exists('mb_detect_encoding')) {
 632         switch (func_get_arg(0)) { /* action */
 633         case 'decode':
 634             $detect_encoding = @mb_detect_encoding($ret);
 635             if ($detect_encoding == 'JIS' ||
 636                 $detect_encoding == 'EUC-JP' ||
 637                 $detect_encoding == 'SJIS' ||
 638                 $detect_encoding == 'UTF-8') {
 639
 640                 $ret = mb_convert_kana(mb_convert_encoding($ret, 'EUC-JP', 'AUTO'), "KV");
 641             }
 642             break;
 643         case 'encode':
 644             $detect_encoding = @mb_detect_encoding($ret);
 645             if ($detect_encoding == 'JIS' ||
 646                 $detect_encoding == 'EUC-JP' ||
 647                 $detect_encoding == 'SJIS' ||
 648                 $detect_encoding == 'UTF-8') {
 649
 650                 $ret = mb_convert_encoding(mb_convert_kana($ret, "KV"), 'JIS', 'AUTO');
 651             }
 652             break;
 653         case 'strimwidth':
 654             $width = func_get_arg(2);
 655             $ret = mb_strimwidth($ret, 0, $width, '...');
 656             break;
 657         case 'encodeheader':
 658             $result = '';
 659             if (strlen($ret) > 0) {
 660                 $tmpstr = mb_substr($ret, 0, 1);
 661                 $prevcsize = strlen($tmpstr);
 662                 for ($i = 1; $i < mb_strlen($ret); $i++) {
 663                     $tmp = mb_substr($ret, $i, 1);
 664                     if (strlen($tmp) == $prevcsize) {
 665                         $tmpstr .= $tmp;
 666                     } else {
 667                         if ($prevcsize == 1) {
 668                             $result .= $tmpstr;
 669                         } else {
 670                             $result .= str_replace(' ', '',
 671                                                    mb_encode_mimeheader($tmpstr,'iso-2022-jp','B',''));
 672                         }
 673                         $tmpstr = $tmp;
 674                         $prevcsize = strlen($tmp);
 675                     }
 676                 }
 677                 if (strlen($tmpstr)) {
 678                     if (strlen(mb_substr($tmpstr, 0, 1)) == 1)
 679                         $result .= $tmpstr;
 680                     else
 681                         $result .= str_replace(' ', '',
 682                                                mb_encode_mimeheader($tmpstr,'iso-2022-jp','B',''));
 683                 }
 684             }
 685             $ret = $result;
 686             break;
 687         case 'decodeheader':
 688             $ret = str_replace("\t", "", $ret);
 689             if (eregi('=\\?([^?]+)\\?(q|b)\\?([^?]+)\\?=', $ret))
 690                 $ret = @mb_decode_mimeheader($ret);
 691             $ret = @mb_convert_encoding($ret, 'EUC-JP', 'AUTO');
 692             break;
 693         case 'downloadfilename':
 694             $useragent = func_get_arg(2);
 695             if (strstr($useragent, 'Windows') !== false ||
 696                 strstr($useragent, 'Mac_') !== false) {
 697                 $ret = mb_convert_encoding($ret, 'SJIS', 'AUTO');
 698             } else {
 699                 $ret = mb_convert_encoding($ret, 'EUC-JP', 'AUTO');
 700 }
 701             break;
 702         case 'wordwrap':
 703             $no_begin = "\x21\x25\x29\x2c\x2e\x3a\x3b\x3f\x5d\x7d\xa1\xf1\xa1\xeb\xa1" .
 704                 "\xc7\xa1\xc9\xa2\xf3\xa1\xec\xa1\xed\xa1\xee\xa1\xa2\xa1\xa3\xa1\xb9" .
 705                 "\xa1\xd3\xa1\xd5\xa1\xd7\xa1\xd9\xa1\xdb\xa1\xcd\xa4\xa1\xa4\xa3\xa4" .
 706                 "\xa5\xa4\xa7\xa4\xa9\xa4\xc3\xa4\xe3\xa4\xe5\xa4\xe7\xa4\xee\xa1\xab" .
 707                 "\xa1\xac\xa1\xb5\xa1\xb6\xa5\xa1\xa5\xa3\xa5\xa5\xa5\xa7\xa5\xa9\xa5" .
 708                 "\xc3\xa5\xe3\xa5\xe5\xa5\xe7\xa5\xee\xa5\xf5\xa5\xf6\xa1\xa6\xa1\xbc" .
 709                 "\xa1\xb3\xa1\xb4\xa1\xaa\xa1\xf3\xa1\xcb\xa1\xa4\xa1\xa5\xa1\xa7\xa1" .
 710                 "\xa8\xa1\xa9\xa1\xcf\xa1\xd1";
 711             $no_end = "\x5c\x24\x28\x5b\x7b\xa1\xf2\x5c\xa1\xc6\xa1\xc8\xa1\xd2\xa1" .
 712                 "\xd4\xa1\xd6\xa1\xd8\xa1\xda\xa1\xcc\xa1\xf0\xa1\xca\xa1\xce\xa1\xd0\xa1\xef";
 713             $wrap = func_get_arg(2);
 714
 715             if (strlen($ret) >= $wrap &&
 716                 substr($ret, 0, 1) != '>' &&
 717                 strpos($ret, 'http://') === FALSE &&
 718                 strpos($ret, 'https://') === FALSE &&
 719                 strpos($ret, 'ftp://') === FALSE) {
 720
 721                 $ret = mb_convert_kana($ret, "KV");
 722
 723                 $line_new = '';
 724                 $ptr = 0;
 725
 726                 while ($ptr < strlen($ret) - 1) {
 727                     $l = mb_strcut($ret, $ptr, $wrap);
 728                     $ptr += strlen($l);
 729                     $tmp = $l;
 730
 731                     $l = mb_strcut($ret, $ptr, 2);
 732                     while (strlen($l) != 0 && mb_strpos($no_begin, $l) !== FALSE ) {
 733                         $tmp .= $l;
 734                         $ptr += strlen($l);
 735                         $l = mb_strcut($ret, $ptr, 1);
 736                     }
 737                     $line_new .= $tmp;
 738                     if ($ptr < strlen($ret) - 1)
 739                         $line_new .= "\n";
 740                 }
 741                 $ret = $line_new;
 742             }
 743             break;
 744         case 'utf7-imap_encode':
 745             $ret = mb_convert_encoding($ret, 'UTF7-IMAP', 'EUC-JP');
 746             break;
 747         case 'utf7-imap_decode':
 748             $ret = mb_convert_encoding($ret, 'EUC-JP', 'UTF7-IMAP');
 749             break;
 750         }
 751     }
 752     return $ret;
 753 }
 754
 755
 756 /**
 757  * Korean charset extra functions
 758  *
 759  * Action performed by function is defined by first argument.
 760  * Default return value is defined by second argument.
 761  *
 762  * @param string action performed by this function.
 763  *    possible values:
 764  *      downloadfilename - Hangul(Korean Character) Attached File Name Fix.
 765  * @param string default return value
 766  */
 767 function korean_charset_xtra() {
 768
 769     $ret = func_get_arg(1);  /* default return value */
 770     if (func_get_arg(0) == 'downloadfilename') { /* action */
 771         $ret = str_replace("\x0D\x0A", '', $ret);  /* Hanmail's CR/LF Clear */
 772         for ($i=0;$i<strlen($ret);$i++) {
 773             if ($ret[$i] >= "\xA1" && $ret[$i] <= "\xFE") {   /* 0xA1 - 0XFE are Valid */
 774                 $i++;
 775                 continue;
 776             } else if (($ret[$i] >= 'a' && $ret[$i] <= 'z') || /* From Original ereg_replace in download.php */
 777                        ($ret[$i] >= 'A' && $ret[$i] <= 'Z') ||
 778                        ($ret[$i] == '.') || ($ret[$i] == '-')) {
 779                 continue;
 780             } else {
 781                 $ret[$i] = '_';
 782             }
 783         }
 784
 785     }
 786     return $ret;
 787 }
 788
 789 /**
 790  * Replaces non-braking spaces inserted by some browsers with regular space
 791  *
 792  * This function can be used to replace non-braking space symbols
 793  * that are inserted in forms by some browsers instead of normal
 794  * space symbol.
 795  *
 796  * @param string $string Text that needs to be cleaned
 797  * @param string $charset Charset used in text
 798  * @return string Cleaned text
 799  */
 800 function cleanup_nbsp($string,$charset) {
 801
 802   // reduce number of case statements
 803   if (stristr('iso-8859-',substr($charset,0,9))){
 804     $output_charset="iso-8859-x";
 805   }
 806   if (stristr('windows-125',substr($charset,0,11))){
 807     $output_charset="cp125x";
 808   }
 809   if (stristr('koi8',substr($charset,0,4))){
 810     $output_charset="koi8-x";
 811   }
 812   if (! isset($output_charset)){
 813     $output_charset=strtolower($charset);
 814   }
 815
 816 // where is non-braking space symbol
 817 switch($output_charset):
 818  case "iso-8859-x":
 819  case "cp125x":
 820  case "iso-2022-jp":
 821   $nbsp="\xA0";
 822   break;
 823  case "koi8-x":
 824    $nbsp="\x9A";
 825    break;
 826  case "utf-8":
 827    $nbsp="\xC2\xA0";
 828    break;
 829  default:
 830    // don't change string if charset is unmatched
 831    return $string;
 832 endswitch;
 833
 834 // return space instead of non-braking space.
 835  return str_replace($nbsp,' ',$string);
 836 }
 837
 838 /**
 839  * Function informs if it is safe to convert given charset to the one that is used by user.
 840  *
 841  * It is safe to use conversion only if user uses utf-8 encoding and when
 842  * converted charset is similar to the one that is used by user.
 843  *
 844  * @param string $input_charset Charset of text that needs to be converted
 845  * @return bool is it possible to convert to user's charset
 846  */
 847 function is_conversion_safe($input_charset) {
 848   global $languages, $sm_notAlias, $default_charset;
 849
 850  // convert to lower case
 851  $input_charset = strtolower($input_charset);
 852
 853  // Is user's locale Unicode based ?
 854  if ( $default_charset == "utf-8" ) {
 855    return true;
 856  }
 857
 858  // Charsets that are similar
 859 switch ($default_charset):
 860 case "windows-1251":
 861       if ( $input_charset == "iso-8859-5" ||
 862            $input_charset == "koi8-r" ||
 863            $input_charset == "koi8-u" ) {
 864         return true;
 865      } else {
 866         return false;
 867      }
 868 case "windows-1257":
 869   if ( $input_charset == "iso-8859-13" ||
 870          $input_charset == "iso-8859-4" ) {
 871     return true;
 872   } else {
 873     return false;
 874   }
 875 case "iso-8859-4":
 876   if ( $input_charset == "iso-8859-13" ||
 877          $input_charset == "windows-1257" ) {
 878      return true;
 879   } else {
 880      return false;
 881   }
 882 case "iso-8859-5":
 883   if ( $input_charset == "windows-1251" ||
 884          $input_charset == "koi8-r" ||
 885          $input_charset == "koi8-u" ) {
 886      return true;
 887   } else {
 888      return false;
 889   }
 890 case "iso-8859-13":
 891   if ( $input_charset == "iso-8859-4" ||
 892        $input_charset == "windows-1257" ) {
 893      return true;
 894   } else {
 895      return false;
 896   }
 897 case "koi8-r":
 898   if ( $input_charset == "windows-1251" ||
 899          $input_charset == "iso-8859-5" ||
 900          $input_charset == "koi8-u" ) {
 901      return true;
 902   } else {
 903      return false;
 904   }
 905 case "koi8-u":
 906   if ( $input_charset == "windows-1251" ||
 907         $input_charset == "iso-8859-5" ||
 908         $input_charset == "koi8-r" ) {
 909      return true;
 910   } else {
 911      return false;
 912   }
 913 default:
 914    return false;
 915 endswitch;
 916 }
 917 ?>