languages/classes/LanguageSr.php

   1 <?php
   2 /**
   3  * Serbian (Српски / Srpski) specific code.
   4  *
   5  * This program is free software; you can redistribute it and/or modify
   6  * it under the terms of the GNU General Public License as published by
   7  * the Free Software Foundation; either version 2 of the License, or
   8  * (at your option) any later version.
   9  *
  10  * This program is distributed in the hope that it will be useful,
  11  * but WITHOUT ANY WARRANTY; without even the implied warranty of
  12  * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
  13  * GNU General Public License for more details.
  14  *
  15  * You should have received a copy of the GNU General Public License along
  16  * with this program; if not, write to the Free Software Foundation, Inc.,
  17  * 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA.
  18  * http://www.gnu.org/copyleft/gpl.html
  19  *
  20  * @file
  21  * @ingroup Language
  22  */
  23
  24 require_once( dirname( __FILE__ ) . '/../LanguageConverter.php' );
  25 require_once( dirname( __FILE__ ) . '/LanguageSr_ec.php' );
  26 require_once( dirname( __FILE__ ) . '/LanguageSr_el.php' );
  27
  28 /**
  29  * There are two levels of conversion for Serbian: the script level
  30  * (Cyrillics <-> Latin), and the variant level (ekavian
  31  * <->iyekavian). The two are orthogonal. So we really only need two
  32  * dictionaries: one for Cyrillics and Latin, and one for ekavian and
  33  * iyekavian.
  34  *
  35  * @ingroup Language
  36  */
  37 class SrConverter extends LanguageConverter {
  38         var $mToLatin = array(
  39                 'а' => 'a', 'б' => 'b',  'в' => 'v', 'г' => 'g',  'д' => 'd',
  40                 'ђ' => 'đ', 'е' => 'e',  'ж' => 'ž', 'з' => 'z',  'и' => 'i',
  41                 'ј' => 'j', 'к' => 'k',  'л' => 'l', 'љ' => 'lj', 'м' => 'm',
  42                 'н' => 'n', 'њ' => 'nj', 'о' => 'o', 'п' => 'p',  'р' => 'r',
  43                 'с' => 's', 'т' => 't',  'ћ' => 'ć', 'у' => 'u',  'ф' => 'f',
  44                 'х' => 'h', 'ц' => 'c',  'ч' => 'č', 'џ' => 'dž', 'ш' => 'š',
  45
  46                 'А' => 'A', 'Б' => 'B',  'В' => 'V', 'Г' => 'G',  'Д' => 'D',
  47                 'Ђ' => 'Đ', 'Е' => 'E',  'Ж' => 'Ž', 'З' => 'Z',  'И' => 'I',
  48                 'Ј' => 'J', 'К' => 'K',  'Л' => 'L', 'Љ' => 'Lj', 'М' => 'M',
  49                 'Н' => 'N', 'Њ' => 'Nj', 'О' => 'O', 'П' => 'P',  'Р' => 'R',
  50                 'С' => 'S', 'Т' => 'T',  'Ћ' => 'Ć', 'У' => 'U',  'Ф' => 'F',
  51                 'Х' => 'H', 'Ц' => 'C',  'Ч' => 'Č', 'Џ' => 'Dž', 'Ш' => 'Š',
  52         );
  53
  54         var $mToCyrillics = array(
  55                 'a' => 'а', 'b'  => 'б', 'c' => 'ц', 'č' => 'ч', 'ć'  => 'ћ',
  56                 'd' => 'д', 'dž' => 'џ', 'đ' => 'ђ', 'e' => 'е', 'f'  => 'ф',
  57                 'g' => 'г', 'h'  => 'х', 'i' => 'и', 'j' => 'ј', 'k'  => 'к',
  58                 'l' => 'л', 'lj' => 'љ', 'm' => 'м', 'n' => 'н', 'nj' => 'њ',
  59                 'o' => 'о', 'p'  => 'п', 'r' => 'р', 's' => 'с', 'š'  => 'ш',
  60                 't' => 'т', 'u'  => 'у', 'v' => 'в', 'z' => 'з', 'ž'  => 'ж',
  61
  62                 'A' => 'А', 'B'  => 'Б', 'C' => 'Ц', 'Č' => 'Ч', 'Ć'  => 'Ћ',
  63                 'D' => 'Д', 'Dž' => 'Џ', 'Đ' => 'Ђ', 'E' => 'Е', 'F'  => 'Ф',
  64                 'G' => 'Г', 'H'  => 'Х', 'I' => 'И', 'J' => 'Ј', 'K'  => 'К',
  65                 'L' => 'Л', 'LJ' => 'Љ', 'M' => 'М', 'N' => 'Н', 'NJ' => 'Њ',
  66                 'O' => 'О', 'P'  => 'П', 'R' => 'Р', 'S' => 'С', 'Š'  => 'Ш',
  67                 'T' => 'Т', 'U'  => 'У', 'V' => 'В', 'Z' => 'З', 'Ž'  => 'Ж',
  68
  69                 'DŽ' => 'Џ', 'd!ž' => 'дж', 'D!ž' => 'Дж', 'D!Ž' => 'ДЖ',
  70                 'Lj' => 'Љ', 'l!j' => 'лј', 'L!j' => 'Лј', 'L!J' => 'ЛЈ',
  71                 'Nj' => 'Њ', 'n!j' => 'нј', 'N!j' => 'Нј', 'N!J' => 'НЈ'
  72         );
  73
  74         function loadDefaultTables() {
  75                 $this->mTables = array(
  76                         'sr-ec' => new ReplacementArray( $this->mToCyrillics ),
  77                         'sr-el' => new ReplacementArray( $this->mToLatin ),
  78                         'sr'    => new ReplacementArray()
  79                 );
  80         }
  81
  82         /**
  83          * rules should be defined as -{ekavian | iyekavian-} -or-
  84          * -{code:text | code:text | ...}-
  85          *
  86          * update: delete all rule parsing because it's not used
  87          * currently, and just produces a couple of bugs
  88          *
  89          * @param $rule string
  90          * @param $flags array
  91          * @return array
  92          */
  93         function parseManualRule( $rule, $flags = array() ) {
  94                 if ( in_array( 'T', $flags ) ) {
  95                         return parent::parseManualRule( $rule, $flags );
  96                 }
  97
  98                 $carray = array();
  99                 // otherwise ignore all formatting
 100                 foreach ( $this->mVariants as $v ) {
 101                         $carray[$v] = $rule;
 102                 }
 103
 104                 return $carray;
 105         }
 106
 107         /**
 108          * A function wrapper:
 109          *   - if there is no selected variant, leave the link
 110          *     names as they were
 111          *   - do not try to find variants for usernames
 112          *
 113          * @param $link string
 114          * @param $nt Title
 115          * @param $ignoreOtherCond bool
 116          */
 117         function findVariantLink( &$link, &$nt, $ignoreOtherCond = false ) {
 118                 // check for user namespace
 119                 if ( is_object( $nt ) ) {
 120                         $ns = $nt->getNamespace();
 121                         if ( $ns == NS_USER || $ns == NS_USER_TALK )
 122                                 return;
 123                 }
 124
 125                 $oldlink = $link;
 126                 parent::findVariantLink( $link, $nt, $ignoreOtherCond );
 127                 if ( $this->getPreferredVariant() == $this->mMainLanguageCode )
 128                         $link = $oldlink;
 129         }
 130
 131         /**
 132          * We want our external link captions to be converted in variants,
 133          * so we return the original text instead -{$text}-, except for URLs
 134          *
 135          * @param $text string
 136          * @param $noParse bool
 137          *
 138          * @return string
 139          */
 140         function markNoConversion( $text, $noParse = false ) {
 141                 if ( $noParse || preg_match( "/^https?:\/\/|ftp:\/\/|irc:\/\//", $text ) )
 142                         return parent::markNoConversion( $text );
 143                 return $text;
 144         }
 145
 146         /**
 147          * An ugly function wrapper for parsing Image titles
 148          * (to prevent image name conversion)
 149          *
 150          * @param $text string
 151          * @param $toVariant bool
 152          *
 153          * @return string
 154          */
 155         function autoConvert( $text, $toVariant = false ) {
 156                 global $wgTitle;
 157                 if ( is_object( $wgTitle ) && $wgTitle->getNameSpace() == NS_FILE ) {
 158                         $imagename = $wgTitle->getNsText();
 159                         if ( preg_match( "/^$imagename:/", $text ) ) return $text;
 160                 }
 161                 return parent::autoConvert( $text, $toVariant );
 162         }
 163
 164         /**
 165          *  It translates text into variant, specials:
 166          *    - ommiting roman numbers
 167          *
 168          * @param $text string
 169          * @param $toVariant string
 170          *
 171          * @throws MWException
 172          * @return string
 173          */
 174         function translate( $text, $toVariant ) {
 175                 $breaks = '[^\w\x80-\xff]';
 176
 177                 // regexp for roman numbers
 178                 $roman = 'M{0,4}(CM|CD|D?C{0,3})(XC|XL|L?X{0,3})(IX|IV|V?I{0,3})';
 179
 180                 $reg = '/^' . $roman . '$|^' . $roman . $breaks . '|' . $breaks . $roman . '$|' . $breaks . $roman . $breaks . '/';
 181
 182                 $matches = preg_split( $reg, $text, -1, PREG_SPLIT_OFFSET_CAPTURE );
 183
 184                 $m = array_shift( $matches );
 185                 if ( !isset( $this->mTables[$toVariant] ) ) {
 186                         throw new MWException( "Broken variant table: " . implode( ',', array_keys( $this->mTables ) ) );
 187                 }
 188                 $ret = $this->mTables[$toVariant]->replace( $m[0] );
 189                 $mstart = $m[1] + strlen( $m[0] );
 190                 foreach ( $matches as $m ) {
 191                         $ret .= substr( $text, $mstart, $m[1] -$mstart );
 192                         $ret .= parent::translate( $m[0], $toVariant );
 193                         $mstart = $m[1] + strlen( $m[0] );
 194                 }
 195
 196                 return $ret;
 197         }
 198
 199         /**
 200          * Guess if a text is written in Cyrillic or Latin.
 201          * Overrides LanguageConverter::guessVariant()
 202          *
 203          * @param string  $text The text to be checked
 204          * @param string  $variant Language code of the variant to be checked for
 205          * @return bool  true if $text appears to be written in $variant
 206          *
 207          * @author Nikola Smolenski <smolensk@eunet.rs>
 208          * @since 1.19
 209          */
 210         public function guessVariant( $text, $variant ) {
 211                 $numCyrillic = preg_match_all("/[шђчћжШЂЧЋЖ]/u", $text, $dummy);
 212                 $numLatin = preg_match_all("/[šđčćžŠĐČĆŽ]/u", $text, $dummy);
 213
 214                 if( $variant == 'sr-ec' ) {
 215                         return (boolean) ($numCyrillic > $numLatin);
 216                 } elseif( $variant == 'sr-el' ) {
 217                         return (boolean) ($numLatin > $numCyrillic);
 218                 } else {
 219                         return false;
 220                 }
 221
 222         }
 223
 224 }
 225
 226 /**
 227  * Serbian (Српски / Srpski)
 228  *
 229  * @ingroup Language
 230  */
 231 class LanguageSr extends LanguageSr_ec {
 232         function __construct() {
 233                 global $wgHooks;
 234
 235                 parent::__construct();
 236
 237                 $variants = array( 'sr', 'sr-ec', 'sr-el' );
 238                 $variantfallbacks = array(
 239                         'sr'    => 'sr-ec',
 240                         'sr-ec' => 'sr',
 241                         'sr-el' => 'sr',
 242                 );
 243
 244                 $flags = array(
 245                         'S' => 'S', 'писмо' => 'S', 'pismo' => 'S',
 246                         'W' => 'W', 'реч'   => 'W', 'reč'   => 'W', 'ријеч' => 'W', 'riječ' => 'W'
 247                 );
 248                 $this->mConverter = new SrConverter( $this, 'sr', $variants, $variantfallbacks, $flags );
 249                 $wgHooks['ArticleSaveComplete'][] = $this->mConverter;
 250         }
 251
 252         /**
 253          * @param $count int
 254          * @param $forms array
 255          *
 256          * @return string
 257          */
 258         function convertPlural( $count, $forms ) {
 259                 if ( !count( $forms ) ) {
 260                         return '';
 261                 }
 262
 263                 // If the actual number is not mentioned in the expression, then just two forms are enough:
 264                 // singular for $count == 1
 265                 // plural   for $count != 1
 266                 // For example, "This user belongs to {{PLURAL:$1|one group|several groups}}."
 267                 if ( count( $forms ) === 2 ) {
 268                         return $count == 1 ? $forms[0] : $forms[1];
 269                 }
 270
 271                 // @todo FIXME: CLDR defines 4 plural forms. Form with decimals missing.
 272                 // See http://unicode.org/repos/cldr-tmp/trunk/diff/supplemental/language_plural_rules.html#ru
 273                 $forms = $this->preConvertPlural( $forms, 3 );
 274
 275                 if ( $count > 10 && floor( ( $count % 100 ) / 10 ) == 1 ) {
 276                         return $forms[2];
 277                 } else {
 278                         switch ( $count % 10 ) {
 279                                 case 1:  return $forms[0];
 280                                 case 2:
 281                                 case 3:
 282                                 case 4:  return $forms[1];
 283                                 default: return $forms[2];
 284                         }
 285                 }
 286         }
 287 }