includes/language/LanguageCode.php

   1 <?php
   2 /**
   3  * This program is free software; you can redistribute it and/or modify
   4  * it under the terms of the GNU General Public License as published by
   5  * the Free Software Foundation; either version 2 of the License, or
   6  * (at your option) any later version.
   7  *
   8  * This program is distributed in the hope that it will be useful,
   9  * but WITHOUT ANY WARRANTY; without even the implied warranty of
  10  * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
  11  * GNU General Public License for more details.
  12  *
  13  * You should have received a copy of the GNU General Public License along
  14  * with this program; if not, write to the Free Software Foundation, Inc.,
  15  * 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA.
  16  * http://www.gnu.org/copyleft/gpl.html
  17  *
  18  * @file
  19  * @ingroup Language
  20  */
  21
  22 /**
  23  * Methods for dealing with language codes.
  24  * @todo Move some of the code-related static methods out of Language into this class
  25  *
  26  * @since 1.29
  27  * @ingroup Language
  28  */
  29 class LanguageCode {
  30         /**
  31          * Mapping of deprecated language codes that were used in previous
  32          * versions of MediaWiki to up-to-date, current language codes.
  33          * These may or may not be valid BCP 47 codes; they are included here
  34          * because MediaWiki renamed these particular codes at some point.
  35          *
  36          * @var array Mapping from deprecated MediaWiki-internal language code
  37          *   to replacement MediaWiki-internal language code.
  38          *
  39          * @since 1.30
  40          * @see https://meta.wikimedia.org/wiki/Special_language_codes
  41          */
  42         private static $deprecatedLanguageCodeMapping = [
  43                 // Note that als is actually a valid ISO 639 code (Tosk Albanian), but it
  44                 // was previously used in MediaWiki for Alsatian, which comes under gsw
  45                 'als' => 'gsw', // T25215
  46                 'bat-smg' => 'sgs', // T27522
  47                 'be-x-old' => 'be-tarask', // T11823
  48                 'fiu-vro' => 'vro', // T31186
  49                 'roa-rup' => 'rup', // T17988
  50                 'zh-classical' => 'lzh', // T30443
  51                 'zh-min-nan' => 'nan', // T30442
  52                 'zh-yue' => 'yue', // T30441
  53         ];
  54
  55         /**
  56          * Mapping of non-standard language codes used in MediaWiki to
  57          * standardized BCP 47 codes.  These are not deprecated (yet?):
  58          * IANA may eventually recognize the subtag, in which case the `-x-`
  59          * infix could be removed, or else we could rename the code in
  60          * MediaWiki, in which case they'd move up to the above mapping
  61          * of deprecated codes.
  62          *
  63          * As a rule, we preserve all distinctions made by MediaWiki
  64          * internally.  For example, `de-formal` becomes `de-x-formal`
  65          * instead of just `de` because MediaWiki distinguishes `de-formal`
  66          * from `de` (for example, for interface translations).  Similarly,
  67          * BCP 47 indicates that `kk-Cyrl` SHOULD not be used because it
  68          * "typically does not add information", but in our case MediaWiki
  69          * LanguageConverter distinguishes `kk` (render content in a mix of
  70          * Kurdish variants) from `kk-Cyrl` (convert content to be uniformly
  71          * Cyrillic).  As the BCP 47 requirement is a SHOULD not a MUST,
  72          * `kk-Cyrl` is a valid code, although some validators may emit
  73          * a warning note.
  74          *
  75          * @var array Mapping from nonstandard MediaWiki-internal codes to
  76          *   BCP 47 codes
  77          *
  78          * @since 1.32
  79          * @see https://meta.wikimedia.org/wiki/Special_language_codes
  80          * @see https://phabricator.wikimedia.org/T125073
  81          */
  82         private static $nonstandardLanguageCodeMapping = [
  83                 // All codes returned by Language::fetchLanguageNames() validated
  84                 // against IANA registry at
  85                 //   https://www.iana.org/assignments/language-subtag-registry/language-subtag-registry
  86                 // with help of validator at
  87                 //   http://schneegans.de/lv/
  88                 'cbk-zam' => 'cbk', // T124657
  89                 'de-formal' => 'de-x-formal',
  90                 'eml' => 'egl', // T36217
  91                 'en-rtl' => 'en-x-rtl',
  92                 'es-formal' => 'es-x-formal',
  93                 'hu-formal' => 'hu-x-formal',
  94                 'map-bms' => 'jv-x-bms', // [[en:Banyumasan_dialect]] T125073
  95                 'mo' => 'ro-Cyrl-MD', // T125073
  96                 'nrm' => 'nrf', // [[en:Norman_language]] T25216
  97                 'nl-informal' => 'nl-x-informal',
  98                 'roa-tara' => 'nap-x-tara', // [[en:Tarantino_dialect]]
  99                 'simple' => 'en-simple',
 100                 'sr-ec' => 'sr-Cyrl', // T117845
 101                 'sr-el' => 'sr-Latn', // T117845
 102
 103                 // Although these next codes aren't *wrong* per se, including
 104                 // both the script and the country code helps compatibility with
 105                 // other BCP 47 users. Note that MW also uses `zh-Hans`/`zh-Hant`,
 106                 // without a country code, and those should be left alone.
 107                 // (See $variantfallbacks in LanguageZh.php for Hans/Hant id.)
 108                 'zh-cn' => 'zh-Hans-CN',
 109                 'zh-sg' => 'zh-Hans-SG',
 110                 'zh-my' => 'zh-Hans-MY',
 111                 'zh-tw' => 'zh-Hant-TW',
 112                 'zh-hk' => 'zh-Hant-HK',
 113                 'zh-mo' => 'zh-Hant-MO',
 114         ];
 115
 116         /**
 117          * Returns a mapping of deprecated language codes that were used in previous
 118          * versions of MediaWiki to up-to-date, current language codes.
 119          *
 120          * This array is merged into $wgDummyLanguageCodes in Setup.php, along with
 121          * the fake language codes 'qqq' and 'qqx', which are used internally by
 122          * MediaWiki's localisation system.
 123          *
 124          * @return string[]
 125          *
 126          * @since 1.29
 127          */
 128         public static function getDeprecatedCodeMapping() {
 129                 return self::$deprecatedLanguageCodeMapping;
 130         }
 131
 132         /**
 133          * Returns a mapping of non-standard language codes used by
 134          * (current and previous version of) MediaWiki, mapped to standard
 135          * BCP 47 names.
 136          *
 137          * This array is exported to JavaScript to ensure
 138          * mediawiki.language.bcp47 stays in sync with LanguageCode::bcp47().
 139          *
 140          * @return string[]
 141          *
 142          * @since 1.32
 143          */
 144         public static function getNonstandardLanguageCodeMapping() {
 145                 $result = [];
 146                 foreach ( self::$deprecatedLanguageCodeMapping as $code => $ignore ) {
 147                         $result[$code] = self::bcp47( $code );
 148                 }
 149                 foreach ( self::$nonstandardLanguageCodeMapping as $code => $ignore ) {
 150                         $result[$code] = self::bcp47( $code );
 151                 }
 152                 return $result;
 153         }
 154
 155         /**
 156          * Replace deprecated language codes that were used in previous
 157          * versions of MediaWiki to up-to-date, current language codes.
 158          * Other values will returned unchanged.
 159          *
 160          * @param string $code Old language code
 161          * @return string New language code
 162          *
 163          * @since 1.30
 164          */
 165         public static function replaceDeprecatedCodes( $code ) {
 166                 return self::$deprecatedLanguageCodeMapping[$code] ?? $code;
 167         }
 168
 169         /**
 170          * Get the normalised IETF language tag
 171          * See unit test for examples.
 172          * See mediawiki.language.bcp47 for the JavaScript implementation.
 173          *
 174          * @param string $code The language code.
 175          * @return string A language code complying with BCP 47 standards.
 176          *
 177          * @since 1.31
 178          */
 179         public static function bcp47( $code ) {
 180                 $code = self::replaceDeprecatedCodes( strtolower( $code ) );
 181                 if ( isset( self::$nonstandardLanguageCodeMapping[$code] ) ) {
 182                         $code = self::$nonstandardLanguageCodeMapping[$code];
 183                 }
 184                 $codeSegment = explode( '-', $code );
 185                 $codeBCP = [];
 186                 foreach ( $codeSegment as $segNo => $seg ) {
 187                         // when previous segment is x, it is a private segment and should be lc
 188                         if ( $segNo > 0 && strtolower( $codeSegment[( $segNo - 1 )] ) == 'x' ) {
 189                                 $codeBCP[$segNo] = strtolower( $seg );
 190                         // ISO 3166 country code
 191                         } elseif ( ( strlen( $seg ) == 2 ) && ( $segNo > 0 ) ) {
 192                                 $codeBCP[$segNo] = strtoupper( $seg );
 193                         // ISO 15924 script code
 194                         } elseif ( ( strlen( $seg ) == 4 ) && ( $segNo > 0 ) ) {
 195                                 $codeBCP[$segNo] = ucfirst( strtolower( $seg ) );
 196                         // Use lowercase for other cases
 197                         } else {
 198                                 $codeBCP[$segNo] = strtolower( $seg );
 199                         }
 200                 }
 201                 $langCode = implode( '-', $codeBCP );
 202                 return $langCode;
 203         }
 204 }