languages/classes/LanguageSr.php

   1 <?php
   2
   3 require_once( dirname( __FILE__ ) . '/../LanguageConverter.php' );
   4 require_once( dirname( __FILE__ ) . '/LanguageSr_ec.php' );
   5 require_once( dirname( __FILE__ ) . '/LanguageSr_el.php' );
   6
   7 /**
   8  * There are two levels of conversion for Serbian: the script level
   9  * (Cyrillics <-> Latin), and the variant level (ekavian
  10  * <->iyekavian). The two are orthogonal. So we really only need two
  11  * dictionaries: one for Cyrillics and Latin, and one for ekavian and
  12  * iyekavian.
  13  *
  14  * @ingroup Language
  15  */
  16 class SrConverter extends LanguageConverter {
  17         var $mToLatin = array(
  18                 'а' => 'a', 'б' => 'b',  'в' => 'v', 'г' => 'g',  'д' => 'd',
  19                 'ђ' => 'đ', 'е' => 'e',  'ж' => 'ž', 'з' => 'z',  'и' => 'i',
  20                 'ј' => 'j', 'к' => 'k',  'л' => 'l', 'љ' => 'lj', 'м' => 'm',
  21                 'н' => 'n', 'њ' => 'nj', 'о' => 'o', 'п' => 'p',  'р' => 'r',
  22                 'с' => 's', 'т' => 't',  'ћ' => 'ć', 'у' => 'u',  'ф' => 'f',
  23                 'х' => 'h', 'ц' => 'c',  'ч' => 'č', 'џ' => 'dž', 'ш' => 'š',
  24
  25                 'А' => 'A', 'Б' => 'B',  'В' => 'V', 'Г' => 'G',  'Д' => 'D',
  26                 'Ђ' => 'Đ', 'Е' => 'E',  'Ж' => 'Ž', 'З' => 'Z',  'И' => 'I',
  27                 'Ј' => 'J', 'К' => 'K',  'Л' => 'L', 'Љ' => 'Lj', 'М' => 'M',
  28                 'Н' => 'N', 'Њ' => 'Nj', 'О' => 'O', 'П' => 'P',  'Р' => 'R',
  29                 'С' => 'S', 'Т' => 'T',  'Ћ' => 'Ć', 'У' => 'U',  'Ф' => 'F',
  30                 'Х' => 'H', 'Ц' => 'C',  'Ч' => 'Č', 'Џ' => 'Dž', 'Ш' => 'Š',
  31         );
  32
  33         var $mToCyrillics = array(
  34                 'a' => 'а', 'b'  => 'б', 'c' => 'ц', 'č' => 'ч', 'ć'  => 'ћ',
  35                 'd' => 'д', 'dž' => 'џ', 'đ' => 'ђ', 'e' => 'е', 'f'  => 'ф',
  36                 'g' => 'г', 'h'  => 'х', 'i' => 'и', 'j' => 'ј', 'k'  => 'к',
  37                 'l' => 'л', 'lj' => 'љ', 'm' => 'м', 'n' => 'н', 'nj' => 'њ',
  38                 'o' => 'о', 'p'  => 'п', 'r' => 'р', 's' => 'с', 'š'  => 'ш',
  39                 't' => 'т', 'u'  => 'у', 'v' => 'в', 'z' => 'з', 'ž'  => 'ж',
  40
  41                 'A' => 'А', 'B'  => 'Б', 'C' => 'Ц', 'Č' => 'Ч', 'Ć'  => 'Ћ',
  42                 'D' => 'Д', 'Dž' => 'Џ', 'Đ' => 'Ђ', 'E' => 'Е', 'F'  => 'Ф',
  43                 'G' => 'Г', 'H'  => 'Х', 'I' => 'И', 'J' => 'Ј', 'K'  => 'К',
  44                 'L' => 'Л', 'LJ' => 'Љ', 'M' => 'М', 'N' => 'Н', 'NJ' => 'Њ',
  45                 'O' => 'О', 'P'  => 'П', 'R' => 'Р', 'S' => 'С', 'Š'  => 'Ш',
  46                 'T' => 'Т', 'U'  => 'У', 'V' => 'В', 'Z' => 'З', 'Ž'  => 'Ж',
  47
  48                 'DŽ' => 'Џ', 'd!ž' => 'дж', 'D!ž' => 'Дж', 'D!Ž' => 'ДЖ',
  49                 'Lj' => 'Љ', 'l!j' => 'лј', 'L!j' => 'Лј', 'L!J' => 'ЛЈ',
  50                 'Nj' => 'Њ', 'n!j' => 'нј', 'N!j' => 'Нј', 'N!J' => 'НЈ'
  51         );
  52
  53         function loadDefaultTables() {
  54                 $this->mTables = array(
  55                         'sr-ec' => new ReplacementArray( $this->mToCyrillics ),
  56                         'sr-el' => new ReplacementArray( $this->mToLatin ),
  57                         'sr'    => new ReplacementArray()
  58                 );
  59         }
  60
  61         /* rules should be defined as -{ekavian | iyekavian-} -or-
  62                 -{code:text | code:text | ...}-
  63                 update: delete all rule parsing because it's not used
  64                         currently, and just produces a couple of bugs
  65         */
  66         function parseManualRule( $rule, $flags = array() ) {
  67                 if ( in_array( 'T', $flags ) ) {
  68                         return parent::parseManualRule( $rule, $flags );
  69                 }
  70
  71                 $carray = array();
  72                 // otherwise ignore all formatting
  73                 foreach ( $this->mVariants as $v ) {
  74                         $carray[$v] = $rule;
  75                 }
  76
  77                 return $carray;
  78         }
  79
  80         /**
  81          * A function wrapper:
  82          *   - if there is no selected variant, leave the link
  83          *     names as they were
  84          *   - do not try to find variants for usernames
  85          */
  86         function findVariantLink( &$link, &$nt, $ignoreOtherCond = false ) {
  87                 // check for user namespace
  88                 if ( is_object( $nt ) ) {
  89                         $ns = $nt->getNamespace();
  90                         if ( $ns == NS_USER || $ns == NS_USER_TALK )
  91                                 return;
  92                 }
  93
  94                 $oldlink = $link;
  95                 parent::findVariantLink( $link, $nt, $ignoreOtherCond );
  96                 if ( $this->getPreferredVariant() == $this->mMainLanguageCode )
  97                         $link = $oldlink;
  98         }
  99
 100         /**
 101          * We want our external link captions to be converted in variants,
 102          * so we return the original text instead -{$text}-, except for URLs
 103          *
 104          * @return string
 105          */
 106         function markNoConversion( $text, $noParse = false ) {
 107                 if ( $noParse || preg_match( "/^https?:\/\/|ftp:\/\/|irc:\/\//", $text ) )
 108                         return parent::markNoConversion( $text );
 109                 return $text;
 110         }
 111
 112         /**
 113          * An ugly function wrapper for parsing Image titles
 114          * (to prevent image name conversion)
 115          *
 116          * @return string
 117          */
 118         function autoConvert( $text, $toVariant = false ) {
 119                 global $wgTitle;
 120                 if ( is_object( $wgTitle ) && $wgTitle->getNameSpace() == NS_FILE ) {
 121                         $imagename = $wgTitle->getNsText();
 122                         if ( preg_match( "/^$imagename:/", $text ) ) return $text;
 123                 }
 124                 return parent::autoConvert( $text, $toVariant );
 125         }
 126
 127         /**
 128          *  It translates text into variant, specials:
 129          *    - ommiting roman numbers
 130          *
 131          * @return string
 132          */
 133         function translate( $text, $toVariant ) {
 134                 $breaks = '[^\w\x80-\xff]';
 135
 136                 // regexp for roman numbers
 137                 $roman = 'M{0,4}(CM|CD|D?C{0,3})(XC|XL|L?X{0,3})(IX|IV|V?I{0,3})';
 138
 139                 $reg = '/^' . $roman . '$|^' . $roman . $breaks . '|' . $breaks . $roman . '$|' . $breaks . $roman . $breaks . '/';
 140
 141                 $matches = preg_split( $reg, $text, -1, PREG_SPLIT_OFFSET_CAPTURE );
 142
 143                 $m = array_shift( $matches );
 144                 if ( !isset( $this->mTables[$toVariant] ) ) {
 145                         throw new MWException( "Broken variant table: " . implode( ',', array_keys( $this->mTables ) ) );
 146                 }
 147                 $ret = $this->mTables[$toVariant]->replace( $m[0] );
 148                 $mstart = $m[1] + strlen( $m[0] );
 149                 foreach ( $matches as $m ) {
 150                         $ret .= substr( $text, $mstart, $m[1] -$mstart );
 151                         $ret .= parent::translate( $m[0], $toVariant );
 152                         $mstart = $m[1] + strlen( $m[0] );
 153                 }
 154
 155                 return $ret;
 156         }
 157
 158         /**
 159          * Guess if a text is written in Cyrillic or Latin.
 160          *
 161          * @author Nikola Smolenski <smolensk@eunet.rs>
 162          * @since 1.18
 163          *
 164          * @return bool
 165          */
 166         public function guessVariant( $text, $variant ) {
 167                 $numCyrillic = preg_match_all("/[шђчћжШЂЧЋЖ]/u", $text, $dummy);
 168                 $numLatin = preg_match_all("/[šđčćžŠĐČĆŽ]/u", $text, $dummy);
 169
 170                 if( $variant == 'sr-ec' ) {
 171                         return $numCyrillic > $numLatin;
 172                 } else if( $variant == 'sr-el' ) {
 173                         return $numLatin > $numCyrillic;
 174                 } else {
 175                         return false;
 176                 }
 177         }
 178 }
 179
 180 /**
 181  * Serbian (Српски / Srpski)
 182  *
 183  * @ingroup Language
 184  */
 185 class LanguageSr extends LanguageSr_ec {
 186         function __construct() {
 187                 global $wgHooks;
 188
 189                 parent::__construct();
 190
 191                 $variants = array( 'sr', 'sr-ec', 'sr-el' );
 192                 $variantfallbacks = array(
 193                         'sr'    => 'sr-ec',
 194                         'sr-ec' => 'sr',
 195                         'sr-el' => 'sr',
 196                 );
 197
 198                 $flags = array(
 199                         'S' => 'S', 'писмо' => 'S', 'pismo' => 'S',
 200                         'W' => 'W', 'реч'   => 'W', 'reč'   => 'W', 'ријеч' => 'W', 'riječ' => 'W'
 201                 );
 202                 $this->mConverter = new SrConverter( $this, 'sr', $variants, $variantfallbacks, $flags );
 203                 $wgHooks['ArticleSaveComplete'][] = $this->mConverter;
 204         }
 205
 206         /**
 207          * @param $count int
 208          * @param $forms array
 209          *
 210          * @return string
 211          */
 212         function convertPlural( $count, $forms ) {
 213                 if ( !count( $forms ) ) {
 214                         return '';
 215                 }
 216
 217                 // if no number with word, then use $form[0] for singular and $form[1] for plural or zero
 218                 if ( count( $forms ) === 2 ) {
 219                         return $count == 1 ? $forms[0] : $forms[1];
 220                 }
 221
 222                 // @todo FIXME: CLDR defines 4 plural forms. Form with decimals missing.
 223                 // See http://unicode.org/repos/cldr-tmp/trunk/diff/supplemental/language_plural_rules.html#ru
 224                 $forms = $this->preConvertPlural( $forms, 3 );
 225
 226                 if ( $count > 10 && floor( ( $count % 100 ) / 10 ) == 1 ) {
 227                         return $forms[2];
 228                 } else {
 229                         switch ( $count % 10 ) {
 230                                 case 1:  return $forms[0];
 231                                 case 2:
 232                                 case 3:
 233                                 case 4:  return $forms[1];
 234                                 default: return $forms[2];
 235                         }
 236                 }
 237         }
 238
 239 }