languages/classes/LanguageKu.php

   1 <?php
   2 /** Kurdish
   3   * converter routines
   4   *
   5   * @addtogroup Language
   6   */
   7
   8 require_once( dirname(__FILE__).'/../LanguageConverter.php' );
   9 require_once( dirname(__FILE__).'/LanguageKu_ku.php' );
  10
  11 class KuConverter extends LanguageConverter {
  12         var $mArabicToLatin = array(
  13                 'ب' => 'b', 'ج' => 'c', 'چ' => 'ç', 'د' => 'd', 'ف' => 'f', 'گ' => 'g', 'ھ' => 'h',
  14                 'ہ' => 'h', 'ه' => 'h', 'ح' => 'h', 'ژ' => 'j', 'ك' => 'k', 'ک' => 'k', 'ل' => 'l',
  15                 'م' => 'm', 'ن' => 'n', 'پ' => 'p', 'ق' => 'q', 'ر' => 'r', 'س' => 's', 'ش' => 'ş',
  16                 'ت' => 't', 'ڤ' => 'v', 'خ' => 'x', 'غ' => 'x', 'ز' => 'z',
  17
  18                 /* Doppel- und Halbvokale */
  19                 'ڵ' => 'll', #ll
  20                 'ڕ'  => 'rr', #rr
  21                 'ا'  => 'a',
  22                 # 'ئێ' => 'ê', # initial e
  23                 'ە'  => 'e',
  24                 'ه‌'  => 'e', # with one non-joiner
  25                 'ه‌‌'  => 'e', # with two non-joiner
  26                 'ة'  => 'e',
  27                 'ێ' => 'ê',
  28                 'ي'  => 'î',
  29                 'ی'  => 'î', # U+06CC  db 8c  ARABIC LETTER FARSI YEH
  30                 'ى'  => 'î', # U+0649  d9 89  ARABIC LETTER ALEF MAKSURA
  31                 'ۆ'  => 'o',
  32                 'و'  => 'w',
  33                 'ئ'  => '', # initial hemze should not be shown
  34                 '،'  => ',',
  35                 'ع'  => '\'', # ayn
  36                 '؟'  => '?',
  37         );
  38
  39         var $mLatinToArabic = array(
  40                 'b' => 'ب', 'c' => 'ج', 'ç' => 'چ', 'd' => 'د', 'f' => 'ف', 'g' => 'گ',
  41                 'h' => 'ه', 'j' => 'ژ', 'k' => 'ک', 'l' => 'ل',
  42                 'm' => 'م', 'n' => 'ن', 'p' => 'پ', 'q' => 'ق', 'r' => 'ر', 's' => 'س', 'ş' => 'ش',
  43                 't' => 'ت', 'v' => 'ڤ',
  44                 'x' => 'خ', 'y' => 'ی', 'z' => 'ز',
  45
  46
  47                 'B' => 'ب', 'C' => 'ج', 'Ç' => 'چ', 'D' => 'د', 'F' => 'ف', 'G' => 'گ', 'H' => 'ھ',
  48                 'H' => 'ہ', 'H' => 'ه', 'H' => 'ح', 'J' => 'ژ', 'K' => 'ك', 'K' => 'ک', 'L' => 'ل',
  49                 'M' => 'م', 'N' => 'ن', 'P' => 'پ', 'Q' => 'ق', 'R' => 'ر', 'S' => 'س', 'Ş' => 'ش',
  50                 'T' => 'ت', 'V' => 'ڤ', 'W' => 'و', 'X' => 'خ',
  51                 'Y' => 'ی', 'Z' => 'ز',
  52
  53                 /* Doppelkonsonanten */
  54                 # 'll' => 'ڵ', # wenn es geht, doppel-l und l getrennt zu behandeln
  55                 # 'rr' => 'ڕ', # selbiges für doppel-r
  56
  57                 /* Einzelne Großbuchstaben */
  58                 //' C' => 'ج',
  59
  60                 /* Vowels */
  61                 'a' => 'ا',
  62                 'e' => 'ە',
  63                 'ê' => 'ێ',
  64                 'i' => '',
  65                 'î' => 'ی',
  66                 'o' => 'ۆ',
  67                 'u' => 'و',
  68                 'û' => 'وو',
  69                 'w' => 'و',
  70                 ',' => '،',
  71                 '?' => '؟',
  72
  73                 # Try to replace the leading vowel
  74                 ' a' => 'ئا ',
  75                 ' e' => 'ئە ',
  76                 ' ê' => 'ئێ ',
  77                 ' î' => 'ئی ',
  78                 ' o' => 'ئۆ ',
  79                 ' u' => 'ئو ',
  80                 ' û' => 'ئوو ',
  81                 'A'  => 'ئا',
  82                 'E'  => 'ئە',
  83                 'Ê'  => 'ئێ',
  84                 'Î'  => 'ئی',
  85                 'O'  => 'ئۆ',
  86                 'U'  => 'ئو',
  87                 'Û'  => 'ئوو',
  88                 ' A' => 'ئا ',
  89                 ' E' => 'ئە ',
  90                 ' Ê' => 'ئێ ',
  91                 ' Î' => 'ئی ',
  92                 ' O' => 'ئۆ ',
  93                 ' U' => 'ئو ',
  94                 ' Û' => 'ئوو ',
  95                 # eyn erstmal deaktivieren, einfache Anführungsstriche sind einfach zu häufig, um sie als eyn zu interpretieren
  96                 # '\'' => 'ع',
  97
  98         );
  99
 100         function loadDefaultTables() {
 101                 $this->mTables = array(
 102                         'ku-latn' => new ReplacementArray( $this->mArabicToLatin ),
 103                         'ku-arab' => new ReplacementArray( $this->mLatinToArabic ),
 104                         'ku'      => new ReplacementArray()
 105                 );
 106         }
 107
 108
 109         // Do not convert content on talk pages
 110         function parserConvert( $text, &$parser ){
 111                 if(is_object($parser->getTitle() ) && $parser->getTitle()->isTalkPage())
 112                         $this->mDoContentConvert=false;
 113                 else
 114                         $this->mDoContentConvert=true;
 115
 116                 return parent::parserConvert($text, $parser );
 117         }
 118
 119         /*
 120          * A function wrapper:
 121          *   - if there is no selected variant, leave the link
 122          *     names as they were
 123          *   - do not try to find variants for usernames
 124          */
 125         function findVariantLink( &$link, &$nt ) {
 126                 // check for user namespace
 127                 if(is_object($nt)){
 128                         $ns = $nt->getNamespace();
 129                         if($ns==NS_USER || $ns==NS_USER_TALK)
 130                                 return;
 131                 }
 132
 133                 $oldlink=$link;
 134                 parent::findVariantLink($link,$nt);
 135                 if($this->getPreferredVariant()==$this->mMainLanguageCode)
 136                         $link=$oldlink;
 137         }
 138
 139         /*
 140          * We want our external link captions to be converted in variants,
 141          * so we return the original text instead -{$text}-, except for URLs
 142          */
 143         function markNoConversion($text, $noParse=false) {
 144                 if($noParse || preg_match("/^https?:\/\/|ftp:\/\/|irc:\/\//",$text))
 145                     return parent::markNoConversion($text);
 146                 return $text;
 147         }
 148
 149         /*
 150          * An ugly function wrapper for parsing Image titles
 151          * (to prevent image name conversion)
 152          */
 153         function autoConvert($text, $toVariant=false) {
 154                 global $wgTitle;
 155                 if(is_object($wgTitle) && $wgTitle->getNameSpace()==NS_IMAGE){
 156                         $imagename = $wgTitle->getNsText();
 157                         if(preg_match("/^$imagename:/",$text)) return $text;
 158                 }
 159                 return parent::autoConvert($text,$toVariant);
 160         }
 161
 162         /**
 163          *  It translates text into variant, specials:
 164          *    - ommiting roman numbers
 165          */
 166         function translate($text, $toVariant){
 167                 $breaks = '[^\w\x80-\xff]';
 168
 169                 /* From Kazakh interface, maybe we need it later
 170                  *
 171                 // regexp for roman numbers
 172                 $roman = 'M{0,4}(CM|CD|D?C{0,3})(XC|XL|L?X{0,3})(IX|IV|V?I{0,3})';
 173                 $roman = '';
 174
 175                 $reg = '/^'.$roman.'$|^'.$roman.$breaks.'|'.$breaks.$roman.'$|'.$breaks.$roman.$breaks.'/';
 176
 177                 $matches = preg_split($reg, $text, -1, PREG_SPLIT_OFFSET_CAPTURE);
 178
 179                 $m = array_shift($matches);
 180                 if( !isset( $this->mTables[$toVariant] ) ) {
 181                         throw new MWException( "Broken variant table: " . implode( ',', array_keys( $this->mTables ) ) );
 182                 }
 183                 $ret = $this->mTables[$toVariant]->replace( $m[0] );
 184                 $mstart = $m[1]+strlen($m[0]);
 185                 foreach($matches as $m) {
 186                         $ret .= substr($text, $mstart, $m[1]-$mstart);
 187                         $ret .= parent::translate($m[0], $toVariant);
 188                         $mstart = $m[1] + strlen($m[0]);
 189                 }
 190
 191                 return $ret;
 192                 */
 193
 194                 if( !isset( $this->mTables[$toVariant] ) ) {
 195                         throw new MWException( "Broken variant table: " . implode( ',', array_keys( $this->mTables ) ) );
 196                 }
 197
 198                 return parent::translate( $text, $toVariant );
 199         }
 200 }
 201
 202 class LanguageKu extends LanguageKu_ku {
 203
 204         function __construct() {
 205                 global $wgHooks;
 206                 parent::__construct();
 207
 208                 $variants = array( 'ku', 'ku-arab', 'ku-latn' );
 209                 $variantfallbacks = array(
 210                         'ku'      => 'ku-latn',
 211                         'ku-arab' => 'ku-latn',
 212                         'ku-latn' => 'ku-arab',
 213                 );
 214
 215                 $this->mConverter = new KuConverter( $this, 'ku', $variants, $variantfallbacks );
 216                 $wgHooks['ArticleSaveComplete'][] = $this->mConverter;
 217         }
 218
 219 /*   From Kazakh interface, not needed for the moment
 220
 221         function convertGrammar( $word, $case ) {
 222                 $fname="LanguageKu::convertGrammar";
 223                 wfProfileIn( $fname );
 224
 225                 //always convert to ku-latn before convertGrammar
 226                 $w1 = $word;
 227                 $word = $this->mConverter->autoConvert( $word, 'ku-latn' );
 228                 $w2 = $word;
 229                 $word = parent::convertGrammar( $word, $case );
 230                 //restore encoding
 231                 if( $w1 != $w2 ) {
 232                         $word = $this->mConverter->translate( $word, 'ku-latn' );
 233                 }
 234                 wfProfileOut( $fname );
 235                 return $word;
 236         }
 237 */
 238 }
 239
 240