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