languages/classes/LanguageKu.php

   1 <?php
   2 require_once( dirname( __FILE__ ) . '/../LanguageConverter.php' );
   3 require_once( dirname( __FILE__ ) . '/LanguageKu_ku.php' );
   4
   5 /**
   6  * Kurdish converter routines
   7  *
   8  * @ingroup Language
   9  */
  10 class KuConverter extends LanguageConverter {
  11         var $mArabicToLatin = array(
  12                 'ب' => 'b', 'ج' => 'c', 'چ' => 'ç', 'د' => 'd', 'ف' => 'f', 'گ' => 'g', 'ھ' => 'h',
  13                 'ہ' => 'h', 'ه' => 'h', 'ح' => 'h', 'ژ' => 'j', 'ك' => 'k', 'ک' => 'k', 'ل' => 'l',
  14                 'م' => 'm', 'ن' => 'n', 'پ' => 'p', 'ق' => 'q', 'ر' => 'r', 'س' => 's', 'ش' => 'ş',
  15                 'ت' => 't', 'ڤ' => 'v', 'خ' => 'x', 'غ' => 'x', 'ز' => 'z',
  16
  17 // ک و => ku -- ist richtig
  18 //  و ك=> ku -- ist auch richtig
  19
  20                 /* Doppel- und Halbvokale */
  21                 'ڵ' => 'll', # ll
  22                 'ڕ'  => 'rr', # rr
  23                 'ا'  => 'a',
  24                 # 'ئێ' => 'ê', # initial e
  25                 'ە'  => 'e',
  26                 'ه‌'  => 'e', # with one non-joiner
  27                 'ه‌‌'  => 'e', # with two non-joiner
  28                 'ة'  => 'e',
  29                 'ێ' => 'ê',
  30                 'ي'  => 'î',
  31                 'ی'  => 'î', # U+06CC  db 8c  ARABIC LETTER FARSI YEH
  32                 'ى'  => 'î', # U+0649  d9 89  ARABIC LETTER ALEF MAKSURA
  33                 'ۆ'  => 'o',
  34                 'و'  => 'w',
  35                 'ئ'  => '', # initial hemze should not be shown
  36                 '،'  => ',',
  37                 'ع'  => '\'', # ayn
  38                 '؟'  => '?',
  39
  40                 # digits
  41                 '٠' => '0', # &#x0660;
  42                 '١' => '1', # &#x0661;
  43                 '٢' => '2', # &#x0662;
  44                 '٣' => '3', # &#x0663;
  45                 '٤' => '4', # &#x0664;
  46                 '٥' => '5', # &#x0665;
  47                 '٦' => '6', # &#x0666;
  48                 '٧' => '7', # &#x0667;
  49                 '٨' => '8', # &#x0668;
  50                 '٩' => '9', # &#x0669;
  51         );
  52
  53         var $mLatinToArabic = array(
  54                 'b' => 'ب', 'c' => 'ج', 'ç' => 'چ', 'd' => 'د', 'f' => 'ف', 'g' => 'گ',
  55                 'h' => 'ه', 'j' => 'ژ', 'k' => 'ک', 'l' => 'ل',
  56                 'm' => 'م', 'n' => 'ن', 'p' => 'پ', 'q' => 'ق', 'r' => 'ر', 's' => 'س', 'ş' => 'ش',
  57                 't' => 'ت', 'v' => 'ڤ',
  58                 'x' => 'خ', 'y' => 'ی', 'z' => 'ز',
  59
  60
  61                 'B' => 'ب', 'C' => 'ج', 'Ç' => 'چ', 'D' => 'د', 'F' => 'ف', 'G' => 'گ', 'H' => 'ھ',
  62                 'H' => 'ہ', 'H' => 'ه', 'H' => 'ح', 'J' => 'ژ', 'K' => 'ك', 'K' => 'ک', 'L' => 'ل',
  63                 'M' => 'م', 'N' => 'ن', 'P' => 'پ', 'Q' => 'ق', 'R' => 'ر', 'S' => 'س', 'Ş' => 'ش',
  64                 'T' => 'ت', 'V' => 'ڤ', 'W' => 'و', 'X' => 'خ',
  65                 'Y' => 'ی', 'Z' => 'ز',
  66
  67                 /* Doppelkonsonanten */
  68                 # 'll' => 'ڵ', # wenn es geht, doppel-l und l getrennt zu behandeln
  69                 # 'rr' => 'ڕ', # selbiges für doppel-r
  70
  71                 /* Einzelne Großbuchstaben */
  72                 // ' C' => 'ج',
  73
  74                 /* Vowels */
  75                 'a' => 'ا',
  76                 'e' => 'ە',
  77                 'ê' => 'ێ',
  78                 'i' => '',
  79                 'î' => 'ی',
  80                 'o' => 'ۆ',
  81                 'u' => 'و',
  82                 'û' => 'وو',
  83                 'w' => 'و',
  84                 ',' => '،',
  85                 '?' => '؟',
  86
  87                 # Try to replace the leading vowel
  88                 ' a' => 'ئا ',
  89                 ' e' => 'ئە ',
  90                 ' ê' => 'ئێ ',
  91                 ' î' => 'ئی ',
  92                 ' o' => 'ئۆ ',
  93                 ' u' => 'ئو ',
  94                 ' û' => 'ئوو ',
  95                 'A'  => 'ئا',
  96                 'E'  => 'ئە',
  97                 'Ê'  => 'ئێ',
  98                 'Î'  => 'ئی',
  99                 'O'  => 'ئۆ',
 100                 'U'  => 'ئو',
 101                 'Û'  => 'ئوو',
 102                 ' A' => 'ئا ',
 103                 ' E' => 'ئە ',
 104                 ' Ê' => 'ئێ ',
 105                 ' Î' => 'ئی ',
 106                 ' O' => 'ئۆ ',
 107                 ' U' => 'ئو ',
 108                 ' Û' => 'ئوو ',
 109                 # eyn erstmal deaktivieren, einfache Anführungsstriche sind einfach zu häufig, um sie als eyn zu interpretieren
 110                 # '\'' => 'ع',
 111
 112 /*              # deactivated for now, breaks links i.e. in header of Special:Recentchanges :-(
 113                 # digits
 114                 '0' => '٠', # &#x0660;
 115                 '1' => '١', # &#x0661;
 116                 '2' => '٢', # &#x0662;
 117                 '3' => '٣', # &#x0663;
 118                 '4' => '٤', # &#x0664;
 119                 '5' => '٥', # &#x0665;
 120                 '6' => '٦', # &#x0666;
 121                 '7' => '٧', # &#x0667;
 122                 '8' => '٨', # &#x0668;
 123                 '9' => '٩', # &#x0669;
 124 */
 125                 );
 126
 127         function loadDefaultTables() {
 128                 $this->mTables = array(
 129                         'ku-latn' => new ReplacementArray( $this->mArabicToLatin ),
 130                         'ku-arab' => new ReplacementArray( $this->mLatinToArabic ),
 131                         'ku'      => new ReplacementArray()
 132                 );
 133         }
 134
 135         /*
 136          * A function wrapper:
 137          *   - if there is no selected variant, leave the link
 138          *     names as they were
 139          *   - do not try to find variants for usernames
 140          */
 141         function findVariantLink( &$link, &$nt, $ignoreOtherCond = false ) {
 142                 // check for user namespace
 143                 if ( is_object( $nt ) ) {
 144                         $ns = $nt->getNamespace();
 145                         if ( $ns == NS_USER || $ns == NS_USER_TALK )
 146                                 return;
 147                 }
 148
 149                 $oldlink = $link;
 150                 parent::findVariantLink( $link, $nt, $ignoreOtherCond );
 151                 if ( $this->getPreferredVariant() == $this->mMainLanguageCode )
 152                         $link = $oldlink;
 153         }
 154
 155         /*
 156          * We want our external link captions to be converted in variants,
 157          * so we return the original text instead -{$text}-, except for URLs
 158          */
 159         function markNoConversion( $text, $noParse = false ) {
 160                 if ( $noParse || preg_match( "/^https?:\/\/|ftp:\/\/|irc:\/\//", $text ) )
 161                     return parent::markNoConversion( $text );
 162                 return $text;
 163         }
 164
 165         /*
 166          * An ugly function wrapper for parsing Image titles
 167          * (to prevent image name conversion)
 168          */
 169         function autoConvert( $text, $toVariant = false ) {
 170                 global $wgTitle;
 171                 if ( is_object( $wgTitle ) && $wgTitle->getNameSpace() == NS_FILE ) {
 172                         $imagename = $wgTitle->getNsText();
 173                         if ( preg_match( "/^$imagename:/", $text ) ) return $text;
 174                 }
 175                 return parent::autoConvert( $text, $toVariant );
 176         }
 177
 178         /**
 179          *  It translates text into variant, specials:
 180          *    - ommiting roman numbers
 181          */
 182         function translate( $text, $toVariant ) {
 183                 /* From Kazakh interface, maybe we need it later
 184                 $breaks = '[^\w\x80-\xff]';
 185                 // regexp for roman numbers
 186                 $roman = 'M{0,4}(CM|CD|D?C{0,3})(XC|XL|L?X{0,3})(IX|IV|V?I{0,3})';
 187                 $roman = '';
 188
 189                 $reg = '/^'.$roman.'$|^'.$roman.$breaks.'|'.$breaks.$roman.'$|'.$breaks.$roman.$breaks.'/';
 190
 191                 $matches = preg_split($reg, $text, -1, PREG_SPLIT_OFFSET_CAPTURE);
 192
 193                 $m = array_shift($matches);
 194                 if( !isset( $this->mTables[$toVariant] ) ) {
 195                         throw new MWException( "Broken variant table: " . implode( ',', array_keys( $this->mTables ) ) );
 196                 }
 197                 $ret = $this->mTables[$toVariant]->replace( $m[0] );
 198                 $mstart = $m[1]+strlen($m[0]);
 199                 foreach($matches as $m) {
 200                         $ret .= substr($text, $mstart, $m[1]-$mstart);
 201                         $ret .= parent::translate($m[0], $toVariant);
 202                         $mstart = $m[1] + strlen($m[0]);
 203                 }
 204
 205                 return $ret;
 206                 */
 207
 208                 if ( !isset( $this->mTables[$toVariant] ) ) {
 209                         throw new MWException( "Broken variant table: " . implode( ',', array_keys( $this->mTables ) ) );
 210                 }
 211
 212                 return parent::translate( $text, $toVariant );
 213         }
 214 }
 215
 216 /**
 217  * Kurdish (Kurdî / كوردی)
 218  *
 219  * @ingroup Language
 220  */
 221 class LanguageKu extends LanguageKu_ku {
 222
 223         function __construct() {
 224                 global $wgHooks;
 225                 parent::__construct();
 226
 227                 $variants = array( 'ku', 'ku-arab', 'ku-latn' );
 228                 $variantfallbacks = array(
 229                         'ku'      => 'ku-latn',
 230                         'ku-arab' => 'ku-latn',
 231                         'ku-latn' => 'ku-arab',
 232                 );
 233
 234                 $this->mConverter = new KuConverter( $this, 'ku', $variants, $variantfallbacks );
 235                 $wgHooks['ArticleSaveComplete'][] = $this->mConverter;
 236         }
 237 }