include/languages.php

   1 <?php
   2
   3 /**
   4  * SquirrelMail internationalization functions
   5  *
   6  * This file contains variuos functions that are needed to do
   7  * internationalization of SquirrelMail.
   8  *
   9  * Internally the output character set is used. Other characters are
  10  * encoded using Unicode entities according to HTML 4.0.
  11  *
  12  * @copyright &copy; 1999-2006 The SquirrelMail Project Team
  13  * @license http://opensource.org/licenses/gpl-license.php GNU Public License
  14  * @version $Id$
  15  * @package squirrelmail
  16  * @subpackage i18n
  17  */
  18
  19
  20 /**
  21  * Gettext bindtextdomain wrapper.
  22  *
  23  * Wrapper solves differences between php versions in order to provide
  24  * ngettext support. Should be used if translation uses ngettext
  25  * functions.
  26  * @since 1.5.1
  27  * @param string $domain gettext domain name
  28  * @param string $dir directory that contains all translations
  29  * @return string path to translation directory
  30  */
  31 function sq_bindtextdomain($domain,$dir) {
  32     global $l10n, $gettext_flags, $sm_notAlias;
  33
  34     if ($gettext_flags==7) {
  35         // gettext extension without ngettext
  36         if (substr($dir, -1) != '/') $dir .= '/';
  37         $mofile=$dir . $sm_notAlias . '/LC_MESSAGES/' . $domain . '.mo';
  38         $input = new FileReader($mofile);
  39         $l10n[$domain] = new gettext_reader($input);
  40     }
  41
  42     $dir=bindtextdomain($domain,$dir);
  43
  44     return $dir;
  45 }
  46
  47 /**
  48  * Gettext textdomain wrapper.
  49  * Makes sure that gettext_domain global is modified.
  50  * @since 1.5.1
  51  * @param string $name gettext domain name
  52  * @return string gettext domain name
  53  */
  54 function sq_textdomain($domain) {
  55     global $gettext_domain;
  56     $gettext_domain=textdomain($domain);
  57     return $gettext_domain;
  58 }
  59
  60 /**
  61  * php setlocale function wrapper
  62  *
  63  * From php 4.3.0 it is possible to use arrays in order to set locale.
  64  * php gettext extension works only when locale is set. This wrapper
  65  * function allows to use more than one locale name.
  66  *
  67  * @param int $category locale category name. Use php named constants
  68  *     (LC_ALL, LC_COLLATE, LC_CTYPE, LC_MONETARY, LC_NUMERIC, LC_TIME)
  69  * @param mixed $locale option contains array with possible locales or string with one locale
  70  * @return string name of set locale or false, if all locales fail.
  71  * @since 1.5.1 and 1.4.5
  72  * @see http://www.php.net/setlocale
  73  */
  74 function sq_setlocale($category,$locale) {
  75     // string with only one locale
  76     if (is_string($locale))
  77         return setlocale($category,$locale);
  78
  79     if (! check_php_version(4,3)) {
  80         $ret=false;
  81         $index=0;
  82         while ( ! $ret && $index<count($locale)) {
  83             $ret=setlocale($category,$locale[$index]);
  84             $index++;
  85         }
  86     } else {
  87         // php 4.3.0 or better, use entire array
  88         $ret=setlocale($category,$locale);
  89     }
  90     return $ret;
  91 }
  92
  93 /**
  94  * Converts string from given charset to charset, that can be displayed by user translation.
  95  *
  96  * Function by default returns html encoded strings, if translation uses different encoding.
  97  * If Japanese translation is used - function returns string converted to euc-jp
  98  * If iconv or recode functions are enabled and translation uses utf-8 - function returns utf-8 encoded string.
  99  * If $charset is not supported - function returns unconverted string.
 100  *
 101  * sanitizing of html tags is also done by this function.
 102  *
 103  * @param string $charset
 104  * @param string $string Text to be decoded
 105  * @param boolean $force_decode converts string to html without $charset!=$default_charset check.
 106  * Argument is available since 1.5.1 and 1.4.5.
 107  * @param boolean $save_html disables htmlspecialchars() in order to preserve
 108  *  html formating. Use with care. Available since 1.5.1
 109  * @return string decoded string
 110  */
 111 function charset_decode ($charset, $string, $force_decode=false, $save_html=false) {
 112     global $languages, $squirrelmail_language, $default_charset;
 113     global $use_php_recode, $use_php_iconv, $aggressive_decoding;
 114
 115     if (isset($languages[$squirrelmail_language]['XTRA_CODE']) &&
 116         function_exists($languages[$squirrelmail_language]['XTRA_CODE'] . '_decode')) {
 117         $string = call_user_func($languages[$squirrelmail_language]['XTRA_CODE'] . '_decode', $string);
 118     }
 119
 120     $charset = strtolower($charset);
 121
 122     set_my_charset();
 123
 124     // Variables that allow to use functions without function_exist() calls
 125     if (! isset($use_php_recode) || $use_php_recode=="" ) {
 126         $use_php_recode=false; }
 127     if (! isset($use_php_iconv) || $use_php_iconv=="" ) {
 128         $use_php_iconv=false; }
 129
 130     // Don't do conversion if charset is the same.
 131     if ( ! $force_decode && $charset == strtolower($default_charset) )
 132         return ($save_html ? $string : htmlspecialchars($string));
 133
 134     // catch iso-8859-8-i thing
 135     if ( $charset == "iso-8859-8-i" )
 136         $charset = "iso-8859-8";
 137
 138     /*
 139      * Recode converts html special characters automatically if you use
 140      * 'charset..html' decoding. There is no documented way to put -d option
 141      * into php recode function call.
 142      */
 143     if ( $use_php_recode ) {
 144         if ( $default_charset == "utf-8" ) {
 145             // other charsets can be converted to utf-8 without loss.
 146             // and output string is smaller
 147             $string = recode_string($charset . "..utf-8",$string);
 148             return ($save_html ? $string : htmlspecialchars($string));
 149         } else {
 150             $string = recode_string($charset . "..html",$string);
 151             // recode does not convert single quote, htmlspecialchars does.
 152             $string = str_replace("'", '&#039;', $string);
 153             // undo html specialchars
 154             if ($save_html)
 155                 $string=str_replace(array('&amp;','&quot;','&lt;','&gt;'),
 156                                     array('&','"','<','>'),$string);
 157             return $string;
 158         }
 159     }
 160
 161     // iconv functions does not have html target and can be used only with utf-8
 162     if ( $use_php_iconv && $default_charset=='utf-8') {
 163         $string = iconv($charset,$default_charset,$string);
 164         return ($save_html ? $string : htmlspecialchars($string));
 165     }
 166
 167     // If we don't use recode and iconv, we'll do it old way.
 168
 169     /* All HTML special characters are 7 bit and can be replaced first */
 170     if (! $save_html) $string = htmlspecialchars ($string);
 171
 172     /* controls cpu and memory intensive decoding cycles */
 173     if (! isset($aggressive_decoding) || $aggressive_decoding=="" ) {
 174         $aggressive_decoding=false; }
 175
 176     $decode=fixcharset($charset);
 177     $decodefile=SM_PATH . 'functions/decode/' . $decode . '.php';
 178     if (file_exists($decodefile)) {
 179         include_once($decodefile);
 180         // send $save_html argument to decoding function. needed for iso-2022-xx decoding.
 181         $ret = call_user_func('charset_decode_'.$decode, $string, $save_html);
 182     } else {
 183         $ret = $string;
 184     }
 185     return( $ret );
 186 }
 187
 188 /**
 189  * Converts html string to given charset
 190  * @since 1.5.1 and 1.4.4
 191  * @param string $string
 192  * @param string $charset
 193  * @param boolean $htmlencode keep htmlspecialchars encoding
 194  * @param string
 195  */
 196 function charset_encode($string,$charset,$htmlencode=true) {
 197     global $default_charset;
 198
 199     $encode=fixcharset($charset);
 200     $encodefile=SM_PATH . 'functions/encode/' . $encode . '.php';
 201     if (file_exists($encodefile)) {
 202         include_once($encodefile);
 203         $ret = call_user_func('charset_encode_'.$encode, $string);
 204     } elseif(file_exists(SM_PATH . 'functions/encode/us_ascii.php')) {
 205         // function replaces all 8bit html entities with question marks.
 206         // it is used when other encoding functions are unavailable
 207         include_once(SM_PATH . 'functions/encode/us_ascii.php');
 208         $ret = charset_encode_us_ascii($string);
 209     } else {
 210         /**
 211          * fix for yahoo users that remove all us-ascii related things
 212          */
 213         $ret = $string;
 214     }
 215
 216     /**
 217      * Undo html special chars, some places (like compose form) have
 218      * own sanitizing functions and don't need html symbols.
 219      * Undo chars only after encoding in order to prevent conversion of
 220      * html entities in plain text emails.
 221      */
 222     if (! $htmlencode ) {
 223         $ret = str_replace(array('&amp;','&gt;','&lt;','&quot;'),array('&','>','<','"'),$ret);
 224     }
 225     return( $ret );
 226 }
 227
 228 /**
 229  * Combined decoding and encoding functions
 230  *
 231  * If conversion is done to charset different that utf-8, unsupported symbols
 232  * will be replaced with question marks.
 233  * @since 1.5.1 and 1.4.4
 234  * @param string $in_charset initial charset
 235  * @param string $string string that has to be converted
 236  * @param string $out_charset final charset
 237  * @param boolean $htmlencode keep htmlspecialchars encoding
 238  * @return string converted string
 239  */
 240 function charset_convert($in_charset,$string,$out_charset,$htmlencode=true) {
 241     $string=charset_decode($in_charset,$string,true);
 242     $string=sqi18n_convert_entities($string);
 243     $string=charset_encode($string,$out_charset,$htmlencode);
 244     return $string;
 245 }
 246
 247 /**
 248  * Makes charset name suitable for decoding cycles
 249  *
 250  * @since 1.5.0 and 1.4.4
 251  * @param string $charset Name of charset
 252  * @return string $charset Adjusted name of charset
 253  */
 254 function fixcharset($charset) {
 255     /* remove minus and characters that might be used in paths from charset
 256      * name in order to be able to use it in function names and include calls.
 257      */
 258     $charset=preg_replace("/[-:.\/\\\]/",'_',$charset);
 259
 260     // OE ks_c_5601_1987 > cp949
 261     $charset=str_replace('ks_c_5601_1987','cp949',$charset);
 262     // Moz x-euc-tw > euc-tw
 263     $charset=str_replace('x_euc','euc',$charset);
 264     // Moz x-windows-949 > cp949
 265     $charset=str_replace('x_windows_','cp',$charset);
 266
 267     // windows-125x and cp125x charsets
 268     $charset=str_replace('windows_','cp',$charset);
 269
 270     // ibm > cp
 271     $charset=str_replace('ibm','cp',$charset);
 272
 273     // iso-8859-8-i -> iso-8859-8
 274     // use same cycle until I'll find differences
 275     $charset=str_replace('iso_8859_8_i','iso_8859_8',$charset);
 276
 277     return $charset;
 278 }
 279
 280 /**
 281  * Set up the language to be output
 282  * if $do_search is true, then scan the browser information
 283  * for a possible language that we know
 284  *
 285  * Function sets system locale environment (LC_ALL, LANG, LANGUAGE),
 286  * gettext translation bindings and html header information.
 287  *
 288  * Function returns error codes, if there is some fatal error.
 289  *  0 = no error,
 290  *  1 = mbstring support is not present,
 291  *  2 = mbstring support is not present, user's translation reverted to en_US.
 292  *
 293  * @param string $sm_language translation used by user's interface
 294  * @param bool $do_search use browser's preferred language detection functions. Defaults to false.
 295  * @param bool $default set $sm_language to $squirrelmail_default_language if language detection fails or language is not set. Defaults to false.
 296  * @return int function execution error codes.
 297  */
 298 function set_up_language($sm_language, $do_search = false, $default = false) {
 299
 300     static $SetupAlready = 0;
 301     global $use_gettext, $languages,
 302            $squirrelmail_language, $squirrelmail_default_language, $default_charset,
 303            $sm_notAlias, $username, $data_dir;
 304
 305     if ($SetupAlready) {
 306         return;
 307     }
 308
 309     $SetupAlready = TRUE;
 310     sqgetGlobalVar('HTTP_ACCEPT_LANGUAGE',  $accept_lang, SQ_SERVER);
 311
 312     /**
 313      * If function is asked to detect preferred language
 314      *  OR squirrelmail default language is set to empty string
 315      *    AND
 316      * squirrelmail language ($sm_language) is empty string
 317      * (not set in user's prefs and no cookie with language info)
 318      *    AND
 319      * browser provides list of preferred languages
 320      *  THEN
 321      * get preferred language from HTTP_ACCEPT_LANGUAGE header
 322      */
 323     if (($do_search || empty($squirrelmail_default_language)) &&
 324         ! $sm_language &&
 325         isset($accept_lang)) {
 326         // TODO: use more than one language, if first language is not available
 327         // FIXME: function assumes that string contains two or more characters.
 328         // FIXME: some languages use 5 chars
 329         $sm_language = substr($accept_lang, 0, 2);
 330     }
 331
 332     /**
 333      * If language preference is not set OR script asks to use default language
 334      *  AND
 335      * default squirrelmail language is not set to empty string
 336      *  THEN
 337      * use default squirrelmail language value from configuration.
 338      */
 339     if ((!$sm_language||$default) &&
 340         ! empty($squirrelmail_default_language)) {
 341         $squirrelmail_language = $squirrelmail_default_language;
 342         $sm_language = $squirrelmail_default_language;
 343     }
 344
 345     /** provide failsafe language when detection fails */
 346     if (! $sm_language) $sm_language='en_US';
 347
 348     $sm_notAlias = $sm_language;
 349
 350     // Catching removed translation
 351     // System reverts to English translation if user prefs contain translation
 352     // that is not available in $languages array
 353     if (!isset($languages[$sm_notAlias])) {
 354         $sm_notAlias="en_US";
 355     }
 356
 357     while (isset($languages[$sm_notAlias]['ALIAS'])) {
 358         $sm_notAlias = $languages[$sm_notAlias]['ALIAS'];
 359     }
 360
 361     if ( isset($sm_language) &&
 362          $use_gettext &&
 363          $sm_language != '' &&
 364          isset($languages[$sm_notAlias]['CHARSET']) ) {
 365         sq_bindtextdomain( 'squirrelmail', SM_PATH . 'locale/' );
 366         sq_textdomain( 'squirrelmail' );
 367
 368         // set codeset in order to avoid gettext charset conversions
 369         if (function_exists('bind_textdomain_codeset')) {
 370             // Japanese translation uses different internal charset
 371             if ($sm_notAlias == 'ja_JP') {
 372                 bind_textdomain_codeset ('squirrelmail', 'EUC-JP');
 373             } else {
 374                 bind_textdomain_codeset ('squirrelmail', $languages[$sm_notAlias]['CHARSET'] );
 375             }
 376         }
 377
 378         // Use LOCALE key, if it is set.
 379         if (isset($languages[$sm_notAlias]['LOCALE'])){
 380             $longlocale=$languages[$sm_notAlias]['LOCALE'];
 381         } else {
 382             $longlocale=$sm_notAlias;
 383         }
 384
 385         // try setting locale
 386         $retlocale=sq_setlocale(LC_ALL, $longlocale);
 387
 388         // check if locale is set and assign that locale to $longlocale
 389         // in order to use it in putenv calls.
 390         if (! is_bool($retlocale)) {
 391             $longlocale=$retlocale;
 392         } elseif (is_array($longlocale)) {
 393             // setting of all locales failed.
 394             // we need string instead of array used in LOCALE key.
 395             $longlocale=$sm_notAlias;
 396         }
 397
 398         if ( !((bool)ini_get('safe_mode')) &&
 399              getenv( 'LC_ALL' ) != $longlocale ) {
 400             putenv( "LC_ALL=$longlocale" );
 401             putenv( "LANG=$longlocale" );
 402             putenv( "LANGUAGE=$longlocale" );
 403             putenv( "LC_NUMERIC=C" );
 404             if ($sm_notAlias=='tr_TR') putenv( "LC_CTYPE=C" );
 405         }
 406         // Workaround for plugins that use numbers with floating point
 407         // It might be removed if plugins use correct decimal delimiters
 408         // according to locale settings.
 409         setlocale(LC_NUMERIC, 'C');
 410         // Workaround for specific Turkish strtolower/strtoupper rules.
 411         // Many functions expect English conversion rules.
 412         if ($sm_notAlias=='tr_TR') setlocale(LC_CTYPE,'C');
 413
 414         /**
 415          * Set text direction/alignment variables
 416          * When language environment is setup, scripts can use these globals
 417          * without accessing $languages directly and making checks for optional
 418          * array key.
 419          */
 420         global $text_direction, $left_align, $right_align;
 421         if (isset($languages[$sm_notAlias]['DIR']) &&
 422             $languages[$sm_notAlias]['DIR'] == 'rtl') {
 423             /**
 424              * Text direction
 425              * @global string $text_direction
 426              */
 427             $text_direction='rtl';
 428             /**
 429              * Left alignment
 430              * @global string $left_align
 431              */
 432             $left_align='right';
 433             /**
 434              * Right alignment
 435              * @global string $right_align
 436              */
 437             $right_align='left';
 438         } else {
 439             $text_direction='ltr';
 440             $left_align='left';
 441             $right_align='right';
 442         }
 443
 444         $squirrelmail_language = $sm_notAlias;
 445         if ($squirrelmail_language == 'ja_JP') {
 446             header ('Content-Type: text/html; charset=EUC-JP');
 447             if (!function_exists('mb_internal_encoding')) {
 448                 // Error messages can't be displayed here
 449                 $error = 1;
 450                 // Revert to English if possible.
 451                 if (function_exists('setPref')  && $username!='' && $data_dir!="") {
 452                     setPref($data_dir, $username, 'language', "en_US");
 453                     $error = 2;
 454                 }
 455                 // stop further execution in order not to get php errors on mb_internal_encoding().
 456                 return $error;
 457             }
 458             if (function_exists('mb_language')) {
 459                 mb_language('Japanese');
 460             }
 461             mb_internal_encoding('EUC-JP');
 462             mb_http_output('pass');
 463         } elseif ($squirrelmail_language == 'en_US') {
 464             header( 'Content-Type: text/html; charset=' . $default_charset );
 465         } else {
 466             header( 'Content-Type: text/html; charset=' . $languages[$sm_notAlias]['CHARSET'] );
 467         }
 468         /**
 469          * mbstring.func_overload fix (#929644).
 470          *
 471          * php mbstring extension can replace standard string functions with their multibyte
 472          * equivalents. See http://www.php.net/ref.mbstring#mbstring.overload. This feature
 473          * was added in php v.4.2.0
 474          *
 475          * Some SquirrelMail functions work with 8bit strings in bytes. If interface is forced
 476          * to use mbstring functions and mbstring internal encoding is set to multibyte charset,
 477          * interface can't trust regular string functions. Due to mbstring overloading design
 478          * limits php scripts can't control this setting.
 479          *
 480          * This hack should fix some issues related to 8bit strings in passwords. Correct fix is
 481          * to disable mbstring overloading. Japanese translation uses different internal encoding.
 482          */
 483         if ($squirrelmail_language != 'ja_JP' &&
 484             function_exists('mb_internal_encoding') &&
 485             check_php_version(4,2,0) &&
 486             (int)ini_get('mbstring.func_overload')!=0) {
 487             mb_internal_encoding('pass');
 488         }
 489     }
 490     return 0;
 491 }
 492
 493 /**
 494  * Sets default_charset variable according to the one that is used by user's translations.
 495  *
 496  * Function changes global $default_charset variable in order to be sure, that it
 497  * contains charset used by user's translation. Sanity of $squirrelmail_language
 498  * and $default_charset combination is also tested.
 499  *
 500  * There can be a $default_charset setting in the
 501  * config.php file, but the user may have a different language
 502  * selected for a user interface. This function checks the
 503  * language selected by the user and tags the outgoing messages
 504  * with the appropriate charset corresponding to the language
 505  * selection. This is "more right" (tm), than just stamping the
 506  * message blindly with the system-wide $default_charset.
 507  */
 508 function set_my_charset(){
 509     global $data_dir, $username, $default_charset, $languages, $squirrelmail_language;
 510
 511     $my_language = getPref($data_dir, $username, 'language');
 512     if (!$my_language) {
 513         $my_language = $squirrelmail_language ;
 514     }
 515     // Catch removed translation
 516     if (!isset($languages[$my_language])) {
 517         $my_language="en_US";
 518     }
 519     while (isset($languages[$my_language]['ALIAS'])) {
 520         $my_language = $languages[$my_language]['ALIAS'];
 521     }
 522     $my_charset = $languages[$my_language]['CHARSET'];
 523     if ($my_language!='en_US') {
 524         $default_charset = $my_charset;
 525     }
 526 }
 527
 528 /**
 529  * Replaces non-braking spaces inserted by some browsers with regular space
 530  *
 531  * This function can be used to replace non-braking space symbols
 532  * that are inserted in forms by some browsers instead of normal
 533  * space symbol.
 534  *
 535  * @param string $string Text that needs to be cleaned
 536  * @param string $charset Charset used in text
 537  * @return string Cleaned text
 538  */
 539 function cleanup_nbsp($string,$charset) {
 540
 541   // reduce number of case statements
 542   if (stristr('iso-8859-',substr($charset,0,9))){
 543     $output_charset="iso-8859-x";
 544   }
 545   if (stristr('windows-125',substr($charset,0,11))){
 546     $output_charset="cp125x";
 547   }
 548   if (stristr('koi8',substr($charset,0,4))){
 549     $output_charset="koi8-x";
 550   }
 551   if (! isset($output_charset)){
 552     $output_charset=strtolower($charset);
 553   }
 554
 555 // where is non-braking space symbol
 556 switch($output_charset):
 557  case "iso-8859-x":
 558  case "cp125x":
 559  case "iso-2022-jp":
 560   $nbsp="\xA0";
 561   break;
 562  case "koi8-x":
 563    $nbsp="\x9A";
 564    break;
 565  case "utf-8":
 566    $nbsp="\xC2\xA0";
 567    break;
 568  default:
 569    // don't change string if charset is unmatched
 570    return $string;
 571 endswitch;
 572
 573 // return space instead of non-braking space.
 574  return str_replace($nbsp,' ',$string);
 575 }
 576
 577 /**
 578  * Function informs if it is safe to convert given charset to the one that is used by user.
 579  *
 580  * It is safe to use conversion only if user uses utf-8 encoding and when
 581  * converted charset is similar to the one that is used by user.
 582  *
 583  * @param string $input_charset Charset of text that needs to be converted
 584  * @return bool is it possible to convert to user's charset
 585  */
 586 function is_conversion_safe($input_charset) {
 587     global $languages, $sm_notAlias, $default_charset, $lossy_encoding;
 588
 589     if (isset($lossy_encoding) && $lossy_encoding )
 590         return true;
 591
 592     // convert to lower case
 593     $input_charset = strtolower($input_charset);
 594
 595     // Is user's locale Unicode based ?
 596     if ( $default_charset == "utf-8" ) {
 597         return true;
 598     }
 599
 600     // Charsets that are similar
 601     switch ($default_charset) {
 602     case "windows-1251":
 603         if ( $input_charset == "iso-8859-5" ||
 604              $input_charset == "koi8-r" ||
 605              $input_charset == "koi8-u" ) {
 606             return true;
 607         } else {
 608             return false;
 609         }
 610     case "windows-1257":
 611         if ( $input_charset == "iso-8859-13" ||
 612              $input_charset == "iso-8859-4" ) {
 613             return true;
 614         } else {
 615             return false;
 616         }
 617     case "iso-8859-4":
 618         if ( $input_charset == "iso-8859-13" ||
 619              $input_charset == "windows-1257" ) {
 620             return true;
 621         } else {
 622             return false;
 623         }
 624     case "iso-8859-5":
 625         if ( $input_charset == "windows-1251" ||
 626              $input_charset == "koi8-r" ||
 627              $input_charset == "koi8-u" ) {
 628             return true;
 629         } else {
 630             return false;
 631         }
 632     case "iso-8859-13":
 633         if ( $input_charset == "iso-8859-4" ||
 634              $input_charset == "windows-1257" ) {
 635             return true;
 636         } else {
 637             return false;
 638         }
 639     case "koi8-r":
 640         if ( $input_charset == "windows-1251" ||
 641              $input_charset == "iso-8859-5" ||
 642              $input_charset == "koi8-u" ) {
 643             return true;
 644         } else {
 645             return false;
 646         }
 647     case "koi8-u":
 648         if ( $input_charset == "windows-1251" ||
 649              $input_charset == "iso-8859-5" ||
 650              $input_charset == "koi8-r" ) {
 651             return true;
 652         } else {
 653             return false;
 654         }
 655     default:
 656         return false;
 657     }
 658 }
 659
 660 /**
 661  * Converts html character entities to numeric entities
 662  *
 663  * SquirrelMail encoding functions work only with numeric entities.
 664  * This function fixes issues with decoding functions that might convert
 665  * some symbols to character entities. Issue is specific to PHP recode
 666  * extension decoding. Function is used internally in charset_convert()
 667  * function.
 668  * @param string $str string that might contain html character entities
 669  * @return string string with character entities converted to decimals.
 670  * @since 1.5.2
 671  */
 672 function sqi18n_convert_entities($str) {
 673
 674     $entities = array(
 675         // Latin 1
 676         '&nbsp;'   => '&#160;',
 677         '&iexcl;'  => '&#161;',
 678         '&cent;'   => '&#162;',
 679         '&pound;'  => '&#163;',
 680         '&curren;' => '&#164;',
 681         '&yen;'    => '&#165;',
 682         '&brvbar;' => '&#166;',
 683         '&sect;'   => '&#167;',
 684         '&uml;'    => '&#168;',
 685         '&copy;'   => '&#169;',
 686         '&ordf;'   => '&#170;',
 687         '&laquo;'  => '&#171;',
 688         '&not;'    => '&#172;',
 689         '&shy;'    => '&#173;',
 690         '&reg;'    => '&#174;',
 691         '&macr;'   => '&#175;',
 692         '&deg;'    => '&#176;',
 693         '&plusmn;' => '&#177;',
 694         '&sup2;'   => '&#178;',
 695         '&sup3;'   => '&#179;',
 696         '&acute;'  => '&#180;',
 697         '&micro;'  => '&#181;',
 698         '&para;'   => '&#182;',
 699         '&middot;' => '&#183;',
 700         '&cedil;'  => '&#184;',
 701         '&sup1;'   => '&#185;',
 702         '&ordm;'   => '&#186;',
 703         '&raquo;'  => '&#187;',
 704         '&frac14;' => '&#188;',
 705         '&frac12;' => '&#189;',
 706         '&frac34;' => '&#190;',
 707         '&iquest;' => '&#191;',
 708         '&Agrave;' => '&#192;',
 709         '&Aacute;' => '&#193;',
 710         '&Acirc;'  => '&#194;',
 711         '&Atilde;' => '&#195;',
 712         '&Auml;'   => '&#196;',
 713         '&Aring;'  => '&#197;',
 714         '&AElig;'  => '&#198;',
 715         '&Ccedil;' => '&#199;',
 716         '&Egrave;' => '&#200;',
 717         '&Eacute;' => '&#201;',
 718         '&Ecirc;'  => '&#202;',
 719         '&Euml;'   => '&#203;',
 720         '&Igrave;' => '&#204;',
 721         '&Iacute;' => '&#205;',
 722         '&Icirc;'  => '&#206;',
 723         '&Iuml;'   => '&#207;',
 724         '&ETH;'    => '&#208;',
 725         '&Ntilde;' => '&#209;',
 726         '&Ograve;' => '&#210;',
 727         '&Oacute;' => '&#211;',
 728         '&Ocirc;'  => '&#212;',
 729         '&Otilde;' => '&#213;',
 730         '&Ouml;'   => '&#214;',
 731         '&times;'  => '&#215;',
 732         '&Oslash;' => '&#216;',
 733         '&Ugrave;' => '&#217;',
 734         '&Uacute;' => '&#218;',
 735         '&Ucirc;'  => '&#219;',
 736         '&Uuml;'   => '&#220;',
 737         '&Yacute;' => '&#221;',
 738         '&THORN;'  => '&#222;',
 739         '&szlig;'  => '&#223;',
 740         '&agrave;' => '&#224;',
 741         '&aacute;' => '&#225;',
 742         '&acirc;'  => '&#226;',
 743         '&atilde;' => '&#227;',
 744         '&auml;'   => '&#228;',
 745         '&aring;'  => '&#229;',
 746         '&aelig;'  => '&#230;',
 747         '&ccedil;' => '&#231;',
 748         '&egrave;' => '&#232;',
 749         '&eacute;' => '&#233;',
 750         '&ecirc;'  => '&#234;',
 751         '&euml;'   => '&#235;',
 752         '&igrave;' => '&#236;',
 753         '&iacute;' => '&#237;',
 754         '&icirc;'  => '&#238;',
 755         '&iuml;'   => '&#239;',
 756         '&eth;'    => '&#240;',
 757         '&ntilde;' => '&#241;',
 758         '&ograve;' => '&#242;',
 759         '&oacute;' => '&#243;',
 760         '&ocirc;'  => '&#244;',
 761         '&otilde;' => '&#245;',
 762         '&ouml;'   => '&#246;',
 763         '&divide;' => '&#247;',
 764         '&oslash;' => '&#248;',
 765         '&ugrave;' => '&#249;',
 766         '&uacute;' => '&#250;',
 767         '&ucirc;'  => '&#251;',
 768         '&uuml;'   => '&#252;',
 769         '&yacute;' => '&#253;',
 770         '&thorn;'  => '&#254;',
 771         '&yuml;'   => '&#255;',
 772         // Latin Extended-A
 773         '&OElig;'  => '&#338;',
 774         '&oelig;'  => '&#339;',
 775         '&Scaron;' => '&#352;',
 776         '&scaron;' => '&#353;',
 777         '&Yuml;'   => '&#376;',
 778         // Spacing Modifier Letters
 779         '&circ;'   => '&#710;',
 780         '&tilde;'  => '&#732;',
 781         // General Punctuation
 782         '&ensp;'   => '&#8194;',
 783         '&emsp;'   => '&#8195;',
 784         '&thinsp;' => '&#8201;',
 785         '&zwnj;'   => '&#8204;',
 786         '&zwj;'    => '&#8205;',
 787         '&lrm;'    => '&#8206;',
 788         '&rlm;'    => '&#8207;',
 789         '&ndash;'  => '&#8211;',
 790         '&mdash;'  => '&#8212;',
 791         '&lsquo;'  => '&#8216;',
 792         '&rsquo;'  => '&#8217;',
 793         '&sbquo;'  => '&#8218;',
 794         '&ldquo;'  => '&#8220;',
 795         '&rdquo;'  => '&#8221;',
 796         '&bdquo;'  => '&#8222;',
 797         '&dagger;' => '&#8224;',
 798         '&Dagger;' => '&#8225;',
 799         '&permil;' => '&#8240;',
 800         '&lsaquo;' => '&#8249;',
 801         '&rsaquo;' => '&#8250;',
 802         '&euro;'   => '&#8364;',
 803         // Latin Extended-B
 804         '&fnof;' => '&#402;',
 805         // Greek
 806         '&Alpha;'  => '&#913;',
 807         '&Beta;'   => '&#914;',
 808         '&Gamma;'  => '&#915;',
 809         '&Delta;'  => '&#916;',
 810         '&Epsilon;' => '&#917;',
 811         '&Zeta;'   => '&#918;',
 812         '&Eta;'    => '&#919;',
 813         '&Theta;'  => '&#920;',
 814         '&Iota;'   => '&#921;',
 815         '&Kappa;'  => '&#922;',
 816         '&Lambda;' => '&#923;',
 817         '&Mu;'     => '&#924;',
 818         '&Nu;'     => '&#925;',
 819         '&Xi;'     => '&#926;',
 820         '&Omicron;' => '&#927;',
 821         '&Pi;'     => '&#928;',
 822         '&Rho;'    => '&#929;',
 823         '&Sigma;'  => '&#931;',
 824         '&Tau;'    => '&#932;',
 825         '&Upsilon;' => '&#933;',
 826         '&Phi;'    => '&#934;',
 827         '&Chi;'    => '&#935;',
 828         '&Psi;'    => '&#936;',
 829         '&Omega;'  => '&#937;',
 830         '&alpha;'  => '&#945;',
 831         '&beta;'   => '&#946;',
 832         '&gamma;'  => '&#947;',
 833         '&delta;'  => '&#948;',
 834         '&epsilon;' => '&#949;',
 835         '&zeta;'   => '&#950;',
 836         '&eta;'    => '&#951;',
 837         '&theta;'  => '&#952;',
 838         '&iota;'   => '&#953;',
 839         '&kappa;'  => '&#954;',
 840         '&lambda;' => '&#955;',
 841         '&mu;'     => '&#956;',
 842         '&nu;'     => '&#957;',
 843         '&xi;'     => '&#958;',
 844         '&omicron;' => '&#959;',
 845         '&pi;'     => '&#960;',
 846         '&rho;'    => '&#961;',
 847         '&sigmaf;' => '&#962;',
 848         '&sigma;'  => '&#963;',
 849         '&tau;'    => '&#964;',
 850         '&upsilon;' => '&#965;',
 851         '&phi;'    => '&#966;',
 852         '&chi;'    => '&#967;',
 853         '&psi;'    => '&#968;',
 854         '&omega;'  => '&#969;',
 855         '&thetasym;' => '&#977;',
 856         '&upsih;'  => '&#978;',
 857         '&piv;'    => '&#982;',
 858         // General Punctuation
 859         '&bull;'   => '&#8226;',
 860         '&hellip;' => '&#8230;',
 861         '&prime;'  => '&#8242;',
 862         '&Prime;'  => '&#8243;',
 863         '&oline;'  => '&#8254;',
 864         '&frasl;'  => '&#8260;',
 865         // Letterlike Symbols
 866         '&weierp;' => '&#8472;',
 867         '&image;'  => '&#8465;',
 868         '&real;'   => '&#8476;',
 869         '&trade;'  => '&#8482;',
 870         '&alefsym;' => '&#8501;',
 871         // Arrows
 872         '&larr;'   => '&#8592;',
 873         '&uarr;'   => '&#8593;',
 874         '&rarr;'   => '&#8594;',
 875         '&darr;'   => '&#8595;',
 876         '&harr;'   => '&#8596;',
 877         '&crarr;'  => '&#8629;',
 878         '&lArr;'   => '&#8656;',
 879         '&uArr;'   => '&#8657;',
 880         '&rArr;'   => '&#8658;',
 881         '&dArr;'   => '&#8659;',
 882         '&hArr;'   => '&#8660;',
 883         // Mathematical Operators
 884         '&forall;' => '&#8704;',
 885         '&part;'   => '&#8706;',
 886         '&exist;'  => '&#8707;',
 887         '&empty;'  => '&#8709;',
 888         '&nabla;'  => '&#8711;',
 889         '&isin;'   => '&#8712;',
 890         '&notin;'  => '&#8713;',
 891         '&ni;'     => '&#8715;',
 892         '&prod;'   => '&#8719;',
 893         '&sum;'    => '&#8721;',
 894         '&minus;'  => '&#8722;',
 895         '&lowast;' => '&#8727;',
 896         '&radic;'  => '&#8730;',
 897         '&prop;'   => '&#8733;',
 898         '&infin;'  => '&#8734;',
 899         '&ang;'    => '&#8736;',
 900         '&and;'    => '&#8743;',
 901         '&or;'     => '&#8744;',
 902         '&cap;'    => '&#8745;',
 903         '&cup;'    => '&#8746;',
 904         '&int;'    => '&#8747;',
 905         '&there4;' => '&#8756;',
 906         '&sim;'    => '&#8764;',
 907         '&cong;'   => '&#8773;',
 908         '&asymp;'  => '&#8776;',
 909         '&ne;'     => '&#8800;',
 910         '&equiv;'  => '&#8801;',
 911         '&le;'     => '&#8804;',
 912         '&ge;'     => '&#8805;',
 913         '&sub;'    => '&#8834;',
 914         '&sup;'    => '&#8835;',
 915         '&nsub;'   => '&#8836;',
 916         '&sube;'   => '&#8838;',
 917         '&supe;'   => '&#8839;',
 918         '&oplus;'  => '&#8853;',
 919         '&otimes;' => '&#8855;',
 920         '&perp;'   => '&#8869;',
 921         '&sdot;'   => '&#8901;',
 922         // Miscellaneous Technical
 923         '&lceil;'  => '&#8968;',
 924         '&rceil;'  => '&#8969;',
 925         '&lfloor;' => '&#8970;',
 926         '&rfloor;' => '&#8971;',
 927         '&lang;'   => '&#9001;',
 928         '&rang;'   => '&#9002;',
 929         // Geometric Shapes
 930         '&loz;'    => '&#9674;',
 931         // Miscellaneous Symbols
 932         '&spades;' => '&#9824;',
 933         '&clubs;'  => '&#9827;',
 934         '&hearts;' => '&#9829;',
 935         '&diams;'  => '&#9830;');
 936
 937     $str = str_replace(array_keys($entities), array_values($entities), $str);
 938
 939     return $str;
 940 }
 941
 942 /* ------------------------------ main --------------------------- */
 943
 944 global $squirrelmail_language, $languages, $use_gettext;
 945
 946 if (! sqgetGlobalVar('squirrelmail_language',$squirrelmail_language,SQ_COOKIE)) {
 947     $squirrelmail_language = '';
 948 }
 949
 950 /**
 951  * Array specifies the available translations.
 952  *
 953  * Structure of array:
 954  * $languages['language']['variable'] = 'value'
 955  *
 956  * Possible 'variable' names:
 957  *  NAME      - Translation name in English
 958  *  CHARSET   - Encoding used by translation
 959  *  ALIAS     - used when 'language' is only short name and 'value' should provide long language name
 960  *  ALTNAME   - Native translation name. Any 8bit symbols must be html encoded.
 961  *  LOCALE    - Full locale name (in xx_XX.charset format). It can use array with more than one locale name since 1.4.5 and 1.5.1
 962  *  DIR       - Text direction. Used to define Right-to-Left languages. Possible values 'rtl' or 'ltr'. If undefined - defaults to 'ltr'
 963  *  XTRA_CODE - translation uses special functions. See doc/i18n.txt
 964  *
 965  * Each 'language' definition requires NAME+CHARSET or ALIAS variables.
 966  *
 967  * @name $languages
 968  * @global array $languages
 969  */
 970 $languages['en_US']['NAME']    = 'English';
 971 $languages['en_US']['CHARSET'] = 'iso-8859-1';
 972 $languages['en_US']['LOCALE']  = 'en_US.ISO8859-1';
 973 $languages['en']['ALIAS'] = 'en_US';
 974
 975 /**
 976  * Automatic translation loading from setup.php files.
 977  * Solution for bug. 1240889.
 978  * setup.php file can contain $languages array entries and XTRA_CODE functions.
 979  */
 980 if (is_dir(SM_PATH . 'locale') &&
 981     is_readable(SM_PATH . 'locale')) {
 982     $localedir = dir(SM_PATH . 'locale');
 983     while($lang_dir=$localedir->read()) {
 984         // remove trailing slash, if present
 985         if (substr($lang_dir,-1)=='/') {
 986             $lang_dir = substr($lang_dir,0,-1);
 987         }
 988         if ($lang_dir != '..' && $lang_dir != '.' && $lang_dir != 'CVS' &&
 989             is_dir(SM_PATH.'locale/'.$lang_dir) &&
 990             file_exists(SM_PATH.'locale/'.$lang_dir.'/setup.php')) {
 991             include_once(SM_PATH.'locale/'.$lang_dir.'/setup.php');
 992         }
 993     }
 994     $localedir->close();
 995 }
 996
 997 /* Detect whether gettext is installed. */
 998 $gettext_flags = 0;
 999 if (function_exists('_')) {
1000     $gettext_flags += 1;
1001 }
1002 if (function_exists('bindtextdomain')) {
1003     $gettext_flags += 2;
1004 }
1005 if (function_exists('textdomain')) {
1006     $gettext_flags += 4;
1007 }
1008 if (function_exists('ngettext')) {
1009     $gettext_flags += 8;
1010 }
1011
1012 /* If gettext is fully loaded, cool */
1013 if ($gettext_flags == 15) {
1014     $use_gettext = true;
1015 }
1016
1017 /* If ngettext support is missing, load it */
1018 elseif ($gettext_flags == 7) {
1019     $use_gettext = true;
1020     // load internal ngettext functions
1021     include_once(SM_PATH . 'class/l10n.class.php');
1022     include_once(SM_PATH . 'functions/ngettext.php');
1023 }
1024
1025 /* If we can fake gettext, try that */
1026 elseif ($gettext_flags == 0) {
1027     $use_gettext = true;
1028     include_once(SM_PATH . 'functions/gettext.php');
1029 } else {
1030     /* Uh-ho.  A weird install */
1031     if (! $gettext_flags & 1) {
1032       /**
1033        * Function is used as replacement in broken installs
1034        * @ignore
1035        */
1036         function _($str) {
1037             return $str;
1038         }
1039     }
1040     if (! $gettext_flags & 2) {
1041       /**
1042        * Function is used as replacement in broken installs
1043        * @ignore
1044        */
1045         function bindtextdomain() {
1046             return;
1047         }
1048     }
1049     if (! $gettext_flags & 4) {
1050       /**
1051        * Function is used as replacemet in broken installs
1052        * @ignore
1053        */
1054         function textdomain() {
1055             return;
1056         }
1057     }
1058     if (! $gettext_flags & 8) {
1059         /**
1060          * Function is used as replacemet in broken installs
1061          * @ignore
1062          */
1063         function ngettext($str,$str2,$number) {
1064             if ($number>1) {
1065                 return $str2;
1066             } else {
1067                 return $str;
1068             }
1069         }
1070     }
1071     if (! function_exists('dgettext')) {
1072         /**
1073          * Replacement for broken setups.
1074          * @ignore
1075          */
1076         function dgettext($domain,$str) {
1077             return $str;
1078         }
1079     }
1080     if (! function_exists('dngettext')) {
1081         /**
1082          * Replacement for broken setups
1083          * @ignore
1084          */
1085         function dngettext($domain,$str1,$strn,$number) {
1086             return ($number==1 ? $str1 : $strn);
1087         }
1088     }
1089 }
1090 ?>