languages/classes/LanguageKk.php

   1 <?php
   2 /**
   3  * Kazakh (Қазақша) 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 define( 'KK_C_UC', 'АӘБВГҒДЕЁЖЗИЙКҚЛМНҢОӨПРСТУҰҮФХҺЦЧШЩЪЫІЬЭЮЯ' ); # Kazakh Cyrillic uppercase
  25 define( 'KK_C_LC', 'аәбвгғдеёжзийкқлмнңоөпрстуұүфхһцчшщъыіьэюя' ); # Kazakh Cyrillic lowercase
  26 define( 'KK_L_UC', 'AÄBCÇDEÉFGĞHIİÏJKLMNÑOÖPQRSŞTUÜVWXYÝZ' ); # Kazakh Latin uppercase
  27 define( 'KK_L_LC', 'aäbcçdeéfgğhıiïjklmnñoöpqrsştuüvwxyýz' ); # Kazakh Latin lowercase
  28 // define( 'KK_A', 'ٴابپتجحدرزسشعفقكلمنڭەوۇۋۆىيچھ' ); # Kazakh Arabic
  29 define( 'H_HAMZA', 'ٴ' ); # U+0674 ARABIC LETTER HIGH HAMZA
  30 // define( 'ZWNJ', '‌' ); # U+200C ZERO WIDTH NON-JOINER
  31
  32 /**
  33  * Kazakh (Қазақша) converter routines
  34  *
  35  * @ingroup Language
  36  */
  37 class KkConverter extends LanguageConverter {
  38         protected $mCyrl2Latn, $mLatn2Cyrl, $mCyLa2Arab;
  39
  40         /**
  41          * @param Language $langobj
  42          * @param string $maincode
  43          * @param array $variants
  44          * @param array $variantfallbacks
  45          * @param array $flags
  46          */
  47         function __construct( Language $langobj, $maincode,
  48                 $variants = [],
  49                 $variantfallbacks = [],
  50                 $flags = []
  51         ) {
  52                 parent::__construct( $langobj, $maincode,
  53                         $variants, $variantfallbacks, $flags );
  54
  55                 // No point delaying this since they're in code.
  56                 // Waiting until loadDefaultTables() means they never get loaded
  57                 // when the tables themselves are loaded from cache.
  58                 $this->loadRegs();
  59         }
  60
  61         function loadDefaultTables() {
  62                 // require __DIR__."/../../includes/KkConversion.php";
  63                 // Placeholder for future implementing. Remove variables declarations
  64                 // after generating KkConversion.php
  65                 $kk2Cyrl = [];
  66                 $kk2Latn = [];
  67                 $kk2Arab = [];
  68                 $kk2KZ = [];
  69                 $kk2TR = [];
  70                 $kk2CN = [];
  71
  72                 $this->mTables = [
  73                         'kk-cyrl' => new ReplacementArray( $kk2Cyrl ),
  74                         'kk-latn' => new ReplacementArray( $kk2Latn ),
  75                         'kk-arab' => new ReplacementArray( $kk2Arab ),
  76                         'kk-kz' => new ReplacementArray( array_merge( $kk2Cyrl, $kk2KZ ) ),
  77                         'kk-tr' => new ReplacementArray( array_merge( $kk2Latn, $kk2TR ) ),
  78                         'kk-cn' => new ReplacementArray( array_merge( $kk2Arab, $kk2CN ) ),
  79                         'kk' => new ReplacementArray()
  80                 ];
  81         }
  82
  83         function postLoadTables() {
  84                 $this->mTables['kk-kz']->merge( $this->mTables['kk-cyrl'] );
  85                 $this->mTables['kk-tr']->merge( $this->mTables['kk-latn'] );
  86                 $this->mTables['kk-cn']->merge( $this->mTables['kk-arab'] );
  87         }
  88
  89         function loadRegs() {
  90                 $this->mCyrl2Latn = [
  91                         # # Punctuation
  92                         '/№/u' => 'No.',
  93                         # # Е after vowels
  94                         '/([АӘЕЁИОӨҰҮЭЮЯЪЬ])Е/u' => '$1YE',
  95                         '/([АӘЕЁИОӨҰҮЭЮЯЪЬ])е/ui' => '$1ye',
  96                         # # leading ЁЮЯЩ
  97                         '/^Ё([' . KK_C_UC . ']|$)/u' => 'YO$1', '/^Ё([' . KK_C_LC . ']|$)/u' => 'Yo$1',
  98                         '/^Ю([' . KK_C_UC . ']|$)/u' => 'YU$1', '/^Ю([' . KK_C_LC . ']|$)/u' => 'Yu$1',
  99                         '/^Я([' . KK_C_UC . ']|$)/u' => 'YA$1', '/^Я([' . KK_C_LC . ']|$)/u' => 'Ya$1',
 100                         '/^Щ([' . KK_C_UC . ']|$)/u' => 'ŞÇ$1', '/^Щ([' . KK_C_LC . ']|$)/u' => 'Şç$1',
 101                         # # other ЁЮЯ
 102                         '/Ё/u' => 'YO', '/ё/u' => 'yo',
 103                         '/Ю/u' => 'YU', '/ю/u' => 'yu',
 104                         '/Я/u' => 'YA', '/я/u' => 'ya',
 105                         '/Щ/u' => 'ŞÇ', '/щ/u' => 'şç',
 106                         # # soft and hard signs
 107                         '/[ъЪ]/u' => 'ʺ', '/[ьЬ]/u' => 'ʹ',
 108                         # # other characters
 109                         '/А/u' => 'A', '/а/u' => 'a', '/Ә/u' => 'Ä', '/ә/u' => 'ä',
 110                         '/Б/u' => 'B', '/б/u' => 'b', '/В/u' => 'V', '/в/u' => 'v',
 111                         '/Г/u' => 'G', '/г/u' => 'g', '/Ғ/u' => 'Ğ', '/ғ/u' => 'ğ',
 112                         '/Д/u' => 'D', '/д/u' => 'd', '/Е/u' => 'E', '/е/u' => 'e',
 113                         '/Ж/u' => 'J', '/ж/u' => 'j', '/З/u' => 'Z', '/з/u' => 'z',
 114                         '/И/u' => 'Ï', '/и/u' => 'ï', '/Й/u' => 'Ý', '/й/u' => 'ý',
 115                         '/К/u' => 'K', '/к/u' => 'k', '/Қ/u' => 'Q', '/қ/u' => 'q',
 116                         '/Л/u' => 'L', '/л/u' => 'l', '/М/u' => 'M', '/м/u' => 'm',
 117                         '/Н/u' => 'N', '/н/u' => 'n', '/Ң/u' => 'Ñ', '/ң/u' => 'ñ',
 118                         '/О/u' => 'O', '/о/u' => 'o', '/Ө/u' => 'Ö', '/ө/u' => 'ö',
 119                         '/П/u' => 'P', '/п/u' => 'p', '/Р/u' => 'R', '/р/u' => 'r',
 120                         '/С/u' => 'S', '/с/u' => 's', '/Т/u' => 'T', '/т/u' => 't',
 121                         '/У/u' => 'W', '/у/u' => 'w', '/Ұ/u' => 'U', '/ұ/u' => 'u',
 122                         '/Ү/u' => 'Ü', '/ү/u' => 'ü', '/Ф/u' => 'F', '/ф/u' => 'f',
 123                         '/Х/u' => 'X', '/х/u' => 'x', '/Һ/u' => 'H', '/һ/u' => 'h',
 124                         '/Ц/u' => 'C', '/ц/u' => 'c', '/Ч/u' => 'Ç', '/ч/u' => 'ç',
 125                         '/Ш/u' => 'Ş', '/ш/u' => 'ş', '/Ы/u' => 'I', '/ы/u' => 'ı',
 126                         '/І/u' => 'İ', '/і/u' => 'i', '/Э/u' => 'É', '/э/u' => 'é',
 127                 ];
 128
 129                 $this->mLatn2Cyrl = [
 130                         # # Punctuation
 131                         '/#|No\./' => '№',
 132                         # # Şç
 133                         '/ŞÇʹ/u' => 'ЩЬ', '/Şçʹ/u' => 'Щь',
 134                         '/Ş[Çç]/u' => 'Щ', '/şç/u' => 'щ',
 135                         # # soft and hard signs
 136                         '/([' . KK_L_UC . '])ʺ([' . KK_L_UC . '])/u' => '$1Ъ$2',
 137                         '/ʺ([' . KK_L_LC . '])/u' => 'ъ$1',
 138                         '/([' . KK_L_UC . '])ʹ([' . KK_L_UC . '])/u' => '$1Ь$2',
 139                         '/ʹ([' . KK_L_LC . '])/u' => 'ь$1',
 140                         '/ʺ/u' => 'ъ',
 141                         '/ʹ/u' => 'ь',
 142                         # # Ye Yo Yu Ya.
 143                         '/Y[Ee]/u' => 'Е', '/ye/u' => 'е',
 144                         '/Y[Oo]/u' => 'Ё', '/yo/u' => 'ё',
 145                         '/Y[UWuw]/u' => 'Ю', '/y[uw]/u' => 'ю',
 146                         '/Y[Aa]/u' => 'Я', '/ya/u' => 'я',
 147                         # # other characters
 148                         '/A/u' => 'А', '/a/u' => 'а', '/Ä/u' => 'Ә', '/ä/u' => 'ә',
 149                         '/B/u' => 'Б', '/b/u' => 'б', '/C/u' => 'Ц', '/c/u' => 'ц',
 150                         '/Ç/u' => 'Ч', '/ç/u' => 'ч', '/D/u' => 'Д', '/d/u' => 'д',
 151                         '/E/u' => 'Е', '/e/u' => 'е', '/É/u' => 'Э', '/é/u' => 'э',
 152                         '/F/u' => 'Ф', '/f/u' => 'ф', '/G/u' => 'Г', '/g/u' => 'г',
 153                         '/Ğ/u' => 'Ғ', '/ğ/u' => 'ғ', '/H/u' => 'Һ', '/h/u' => 'һ',
 154                         '/I/u' => 'Ы', '/ı/u' => 'ы', '/İ/u' => 'І', '/i/u' => 'і',
 155                         '/Ï/u' => 'И', '/ï/u' => 'и', '/J/u' => 'Ж', '/j/u' => 'ж',
 156                         '/K/u' => 'К', '/k/u' => 'к', '/L/u' => 'Л', '/l/u' => 'л',
 157                         '/M/u' => 'М', '/m/u' => 'м', '/N/u' => 'Н', '/n/u' => 'н',
 158                         '/Ñ/u' => 'Ң', '/ñ/u' => 'ң', '/O/u' => 'О', '/o/u' => 'о',
 159                         '/Ö/u' => 'Ө', '/ö/u' => 'ө', '/P/u' => 'П', '/p/u' => 'п',
 160                         '/Q/u' => 'Қ', '/q/u' => 'қ', '/R/u' => 'Р', '/r/u' => 'р',
 161                         '/S/u' => 'С', '/s/u' => 'с', '/Ş/u' => 'Ш', '/ş/u' => 'ш',
 162                         '/T/u' => 'Т', '/t/u' => 'т', '/U/u' => 'Ұ', '/u/u' => 'ұ',
 163                         '/Ü/u' => 'Ү', '/ü/u' => 'ү', '/V/u' => 'В', '/v/u' => 'в',
 164                         '/W/u' => 'У', '/w/u' => 'у', '/Ý/u' => 'Й', '/ý/u' => 'й',
 165                         '/X/u' => 'Х', '/x/u' => 'х', '/Z/u' => 'З', '/z/u' => 'з',
 166                 ];
 167
 168                 $this->mCyLa2Arab = [
 169                         # # Punctuation -> Arabic
 170                         '/#|№|No\./u' => '؀', # U+0600
 171                         '/\,/' => '،', # U+060C
 172                         '/;/' => '؛', # U+061B
 173                         '/\?/' => '؟', # U+061F
 174                         '/%/' => '٪', # U+066A
 175                         '/\*/' => '٭', # U+066D
 176                         # # Digits -> Arabic
 177                         '/0/' => '۰', # U+06F0
 178                         '/1/' => '۱', # U+06F1
 179                         '/2/' => '۲', # U+06F2
 180                         '/3/' => '۳', # U+06F3
 181                         '/4/' => '۴', # U+06F4
 182                         '/5/' => '۵', # U+06F5
 183                         '/6/' => '۶', # U+06F6
 184                         '/7/' => '۷', # U+06F7
 185                         '/8/' => '۸', # U+06F8
 186                         '/9/' => '۹', # U+06F9
 187                         # # Cyrillic -> Arabic
 188                         '/Аллаһ/ui' => 'ﷲ',
 189                         '/([АӘЕЁИОӨҰҮЭЮЯЪЬ])е/ui' => '$1يە',
 190                         '/[еэ]/ui' => 'ە', '/[ъь]/ui' => '',
 191                         '/[аә]/ui' => 'ا', '/[оө]/ui' => 'و', '/[ұү]/ui' => 'ۇ', '/[ыі]/ui' => 'ى',
 192                         '/[и]/ui' => 'ىي', '/ё/ui' => 'يو', '/ю/ui' => 'يۋ', '/я/ui' => 'يا', '/[й]/ui' => 'ي',
 193                         '/ц/ui' => 'تس', '/щ/ui' => 'شش',
 194                         '/һ/ui' => 'ح', '/ч/ui' => 'تش',
 195                         # '/һ/ui' => 'ھ', '/ч/ui' => 'چ',
 196                         '/б/ui' => 'ب', '/в/ui' => 'ۆ', '/г/ui' => 'گ', '/ғ/ui' => 'ع',
 197                         '/д/ui' => 'د', '/ж/ui' => 'ج', '/з/ui' => 'ز', '/к/ui' => 'ك',
 198                         '/қ/ui' => 'ق', '/л/ui' => 'ل', '/м/ui' => 'م', '/н/ui' => 'ن',
 199                         '/ң/ui' => 'ڭ', '/п/ui' => 'پ', '/р/ui' => 'ر', '/с/ui' => 'س',
 200                         '/т/ui' => 'ت', '/у/ui' => 'ۋ', '/ф/ui' => 'ف', '/х/ui' => 'ح',
 201                         '/ш/ui' => 'ش',
 202                         # # Latin -> Arabic // commented for now...
 203                         /*'/Allah/ui' => 'ﷲ',
 204                         '/[eé]/ui' => 'ە', '/[yý]/ui' => 'ي', '/[ʺʹ]/ui' => '',
 205                         '/[aä]/ui' => 'ا', '/[oö]/ui' => 'و', '/[uü]/ui' => 'ۇ',
 206                         '/[ï]/ui' => 'ىي', '/[ıIiİ]/u' => 'ى',
 207                         '/c/ui' => 'تس',
 208                         '/ç/ui' => 'تش', '/h/ui' => 'ح',
 209                         #'/ç/ui' => 'چ', '/h/ui' => 'ھ',
 210                         '/b/ui' => 'ب','/d/ui' => 'د',
 211                         '/f/ui' => 'ف', '/g/ui' => 'گ', '/ğ/ui' => 'ع',
 212                         '/j/ui' => 'ج', '/k/ui' => 'ك', '/l/ui' => 'ل', '/m/ui' => 'م',
 213                         '/n/ui' => 'ن', '/ñ/ui' => 'ڭ', '/p/ui' => 'پ', '/q/ui' => 'ق',
 214                         '/r/ui' => 'ر', '/s/ui' => 'س', '/ş/ui' => 'ش', '/t/ui' => 'ت',
 215                         '/v/ui' => 'ۆ', '/w/ui' => 'ۋ', '/x/ui' => 'ح', '/z/ui' => 'ز',*/
 216                 ];
 217         }
 218
 219         /**
 220          * A function wrapper:
 221          *  - if there is no selected variant, leave the link
 222          *    names as they were
 223          *  - do not try to find variants for usernames
 224          *
 225          * @param string &$link
 226          * @param Title &$nt
 227          * @param bool $ignoreOtherCond
 228          */
 229         function findVariantLink( &$link, &$nt, $ignoreOtherCond = false ) {
 230                 // check for user namespace
 231                 if ( is_object( $nt ) ) {
 232                         $ns = $nt->getNamespace();
 233                         if ( $ns == NS_USER || $ns == NS_USER_TALK ) {
 234                                 return;
 235                         }
 236                 }
 237
 238                 $oldlink = $link;
 239                 parent::findVariantLink( $link, $nt, $ignoreOtherCond );
 240                 if ( $this->getPreferredVariant() == $this->mMainLanguageCode ) {
 241                         $link = $oldlink;
 242                 }
 243         }
 244
 245         /**
 246          *  It translates text into variant
 247          *
 248          * @param string $text
 249          * @param string $toVariant
 250          *
 251          * @return string
 252          */
 253         function translate( $text, $toVariant ) {
 254                 $text = parent::translate( $text, $toVariant );
 255
 256                 switch ( $toVariant ) {
 257                         case 'kk-cyrl':
 258                         case 'kk-kz':
 259                                 $letters = KK_L_UC . KK_L_LC . 'ʺʹ#0123456789';
 260                                 break;
 261                         case 'kk-latn':
 262                         case 'kk-tr':
 263                                 $letters = KK_C_UC . KK_C_LC . '№0123456789';
 264                                 break;
 265                         case 'kk-arab':
 266                         case 'kk-cn':
 267                                 $letters = KK_C_UC . KK_C_LC . /*KK_L_UC.KK_L_LC.'ʺʹ'.*/',;\?%\*№0123456789';
 268                                 break;
 269                         default:
 270                                 return $text;
 271                 }
 272                 // disable conversion variables like $1, $2...
 273                 $varsfix = '\$[0-9]';
 274
 275                 $matches = preg_split(
 276                         '/' . $varsfix . '[^' . $letters . ']+/u',
 277                         $text,
 278                         -1,
 279                         PREG_SPLIT_OFFSET_CAPTURE
 280                 );
 281
 282                 $mstart = 0;
 283                 $ret = '';
 284
 285                 foreach ( $matches as $m ) {
 286                         $ret .= substr( $text, $mstart, $m[1] - $mstart );
 287                         $ret .= $this->regsConverter( $m[0], $toVariant );
 288                         $mstart = $m[1] + strlen( $m[0] );
 289                 }
 290
 291                 return $ret;
 292         }
 293
 294         /**
 295          * @param string $text
 296          * @param string $toVariant
 297          * @return mixed|string
 298          */
 299         function regsConverter( $text, $toVariant ) {
 300                 if ( $text == '' ) {
 301                         return $text;
 302                 }
 303
 304                 switch ( $toVariant ) {
 305                         case 'kk-arab':
 306                         case 'kk-cn':
 307                                 $letters = KK_C_LC . KK_C_UC; /*.KK_L_LC.KK_L_UC*/
 308                                 $front = 'әөүіӘӨҮІ'; /*.'äöüiÄÖÜİ'*/
 309                                 $excludes = 'еэгғкқЕЭГҒКҚ'; /*.'eégğkqEÉGĞKQ'*/
 310                                 // split text to words
 311                                 $matches = preg_split( '/[\b\s\-\.:]+/', $text, -1, PREG_SPLIT_OFFSET_CAPTURE );
 312                                 $mstart = 0;
 313                                 $ret = '';
 314                                 foreach ( $matches as $m ) {
 315                                         $ret .= substr( $text, $mstart, $m[1] - $mstart );
 316                                         // is matched the word to front vowels?
 317                                         // exclude a words matched to е, э, г, к, к, қ,
 318                                         // them should be without hamza
 319                                         if ( preg_match( '/[' . $front . ']/u', $m[0] ) &&
 320                                                 !preg_match( '/[' . $excludes . ']/u', $m[0] )
 321                                         ) {
 322                                                 $ret .= preg_replace( '/[' . $letters . ']+/u', H_HAMZA . '$0', $m[0] );
 323                                         } else {
 324                                                 $ret .= $m[0];
 325                                         }
 326                                         $mstart = $m[1] + strlen( $m[0] );
 327                                 }
 328                                 $text =& $ret;
 329                                 foreach ( $this->mCyLa2Arab as $pat => $rep ) {
 330                                         $text = preg_replace( $pat, $rep, $text );
 331                                 }
 332                                 return $text;
 333                         case 'kk-latn':
 334                         case 'kk-tr':
 335                                 foreach ( $this->mCyrl2Latn as $pat => $rep ) {
 336                                         $text = preg_replace( $pat, $rep, $text );
 337                                 }
 338                                 return $text;
 339                         case 'kk-cyrl':
 340                         case 'kk-kz':
 341                                 foreach ( $this->mLatn2Cyrl as $pat => $rep ) {
 342                                         $text = preg_replace( $pat, $rep, $text );
 343                                 }
 344                                 return $text;
 345                         default:
 346                                 return $text;
 347                 }
 348         }
 349
 350         /**
 351          * @param string $key
 352          * @return string
 353          */
 354         function convertCategoryKey( $key ) {
 355                 return $this->autoConvert( $key, 'kk' );
 356         }
 357 }
 358
 359 /**
 360  * class that handles Cyrillic, Latin and Arabic scripts for Kazakh
 361  * right now it only distinguish kk_cyrl, kk_latn, kk_arab and kk_kz, kk_tr, kk_cn.
 362  *
 363  * @ingroup Language
 364  */
 365 class LanguageKk extends LanguageKk_cyrl {
 366         function __construct() {
 367                 parent::__construct();
 368
 369                 $variants = [ 'kk', 'kk-cyrl', 'kk-latn', 'kk-arab', 'kk-kz', 'kk-tr', 'kk-cn' ];
 370                 $variantfallbacks = [
 371                         'kk' => 'kk-cyrl',
 372                         'kk-cyrl' => 'kk',
 373                         'kk-latn' => 'kk',
 374                         'kk-arab' => 'kk',
 375                         'kk-kz' => 'kk-cyrl',
 376                         'kk-tr' => 'kk-latn',
 377                         'kk-cn' => 'kk-arab'
 378                 ];
 379
 380                 $this->mConverter = new KkConverter( $this, 'kk', $variants, $variantfallbacks );
 381         }
 382
 383         /**
 384          * It fixes issue with ucfirst for transforming 'i' to 'İ'
 385          *
 386          * @param string $string
 387          *
 388          * @return string
 389          */
 390         public function ucfirst( $string ) {
 391                 if ( substr( $string, 0, 1 ) === 'i' ) {
 392                         $variant = $this->getPreferredVariant();
 393                         if ( $variant == 'kk-latn' || $variant == 'kk-tr' ) {
 394                                 return 'İ' . substr( $string, 1 );
 395                         }
 396                 }
 397                 return parent::ucfirst( $string );
 398         }
 399
 400         /**
 401          * It fixes issue with  lcfirst for transforming 'I' to 'ı'
 402          *
 403          * @param string $string
 404          *
 405          * @return string
 406          */
 407         function lcfirst( $string ) {
 408                 if ( substr( $string, 0, 1 ) === 'I' ) {
 409                         $variant = $this->getPreferredVariant();
 410                         if ( $variant == 'kk-latn' || $variant == 'kk-tr' ) {
 411                                 return 'ı' . substr( $string, 1 );
 412                         }
 413                 }
 414                 return parent::lcfirst( $string );
 415         }
 416
 417         /**
 418          * @param string $word
 419          * @param string $case
 420          * @return string
 421          */
 422         function convertGrammar( $word, $case ) {
 423                 $variant = $this->getPreferredVariant();
 424                 switch ( $variant ) {
 425                         case 'kk-arab':
 426                         case 'kk-cn':
 427                                 $word = parent::convertGrammarKk_arab( $word, $case );
 428                                 break;
 429                         case 'kk-latn':
 430                         case 'kk-tr':
 431                                 $word = parent::convertGrammarKk_latn( $word, $case );
 432                                 break;
 433                         case 'kk-cyrl':
 434                         case 'kk-kz':
 435                         case 'kk':
 436                         default:
 437                                 $word = parent::convertGrammarKk_cyrl( $word, $case );
 438                 }
 439
 440                 return $word;
 441         }
 442 }