3 require_once( dirname( __FILE__
) . '/../LanguageConverter.php' );
4 require_once( dirname( __FILE__
) . '/LanguageKk_cyrl.php' );
6 define( 'KK_C_UC', 'АӘБВГҒДЕЁЖЗИЙКҚЛМНҢОӨПРСТУҰҮФХҺЦЧШЩЪЫІЬЭЮЯ' ); # Kazakh Cyrillic uppercase
7 define( 'KK_C_LC', 'аәбвгғдеёжзийкқлмнңоөпрстуұүфхһцчшщъыіьэюя' ); # Kazakh Cyrillic lowercase
8 define( 'KK_L_UC', 'AÄBCÇDEÉFGĞHIİÏJKLMNÑOÖPQRSŞTUÜVWXYÝZ' ); # Kazakh Latin uppercase
9 define( 'KK_L_LC', 'aäbcçdeéfgğhıiïjklmnñoöpqrsştuüvwxyýz' ); # Kazakh Latin lowercase
10 // define( 'KK_A', 'ٴابپتجحدرزسشعفقكلمنڭەوۇۋۆىيچھ' ); # Kazakh Arabic
11 define( 'H_HAMZA', 'ٴ' ); # U+0674 ARABIC LETTER HIGH HAMZA
12 // define( 'ZWNJ', '' ); # U+200C ZERO WIDTH NON-JOINER
19 class KkConverter
extends LanguageConverter
{
21 protected $mCyrl2Latn, $mLatn2Cyrl, $mCyLa2Arab;
24 * @param $langobj Language
25 * @param $maincode string
26 * @param $variants array
27 * @param $variantfallbacks array
30 function __construct( $langobj, $maincode,
32 $variantfallbacks = array(),
34 parent
::__construct( $langobj, $maincode,
35 $variants, $variantfallbacks, $flags );
37 // No point delaying this since they're in code.
38 // Waiting until loadDefaultTables() means they never get loaded
39 // when the tables themselves are loaded from cache.
43 function loadDefaultTables() {
44 // require( dirname(__FILE__)."/../../includes/KkConversion.php" );
45 // Placeholder for future implementing. Remove variables declarations
46 // after generating KkConversion.php
54 $this->mTables
= array(
55 'kk-cyrl' => new ReplacementArray( $kk2Cyrl ),
56 'kk-latn' => new ReplacementArray( $kk2Latn ),
57 'kk-arab' => new ReplacementArray( $kk2Arab ),
58 'kk-kz' => new ReplacementArray( array_merge( $kk2Cyrl, $kk2KZ ) ),
59 'kk-tr' => new ReplacementArray( array_merge( $kk2Latn, $kk2TR ) ),
60 'kk-cn' => new ReplacementArray( array_merge( $kk2Arab, $kk2CN ) ),
61 'kk' => new ReplacementArray()
65 function postLoadTables() {
66 $this->mTables
['kk-kz']->merge( $this->mTables
['kk-cyrl'] );
67 $this->mTables
['kk-tr']->merge( $this->mTables
['kk-latn'] );
68 $this->mTables
['kk-cn']->merge( $this->mTables
['kk-arab'] );
73 $this->mCyrl2Latn
= array(
77 '/([АӘЕЁИОӨҰҮЭЮЯЪЬ])Е/u' => '$1YE',
78 '/([АӘЕЁИОӨҰҮЭЮЯЪЬ])е/ui' => '$1ye',
80 '/^Ё([' . KK_C_UC
. ']|$)/u' => 'YO$1', '/^Ё([' . KK_C_LC
. ']|$)/u' => 'Yo$1',
81 '/^Ю([' . KK_C_UC
. ']|$)/u' => 'YU$1', '/^Ю([' . KK_C_LC
. ']|$)/u' => 'Yu$1',
82 '/^Я([' . KK_C_UC
. ']|$)/u' => 'YA$1', '/^Я([' . KK_C_LC
. ']|$)/u' => 'Ya$1',
83 '/^Щ([' . KK_C_UC
. ']|$)/u' => 'ŞÇ$1', '/^Щ([' . KK_C_LC
. ']|$)/u' => 'Şç$1',
85 '/Ё/u' => 'YO', '/ё/u' => 'yo',
86 '/Ю/u' => 'YU', '/ю/u' => 'yu',
87 '/Я/u' => 'YA', '/я/u' => 'ya',
88 '/Щ/u' => 'ŞÇ', '/щ/u' => 'şç',
89 # # soft and hard signs
90 '/[ъЪ]/u' => 'ʺ', '/[ьЬ]/u' => 'ʹ',
92 '/А/u' => 'A', '/а/u' => 'a', '/Ә/u' => 'Ä', '/ә/u' => 'ä',
93 '/Б/u' => 'B', '/б/u' => 'b', '/В/u' => 'V', '/в/u' => 'v',
94 '/Г/u' => 'G', '/г/u' => 'g', '/Ғ/u' => 'Ğ', '/ғ/u' => 'ğ',
95 '/Д/u' => 'D', '/д/u' => 'd', '/Е/u' => 'E', '/е/u' => 'e',
96 '/Ж/u' => 'J', '/ж/u' => 'j', '/З/u' => 'Z', '/з/u' => 'z',
97 '/И/u' => 'Ï', '/и/u' => 'ï', '/Й/u' => 'Ý', '/й/u' => 'ý',
98 '/К/u' => 'K', '/к/u' => 'k', '/Қ/u' => 'Q', '/қ/u' => 'q',
99 '/Л/u' => 'L', '/л/u' => 'l', '/М/u' => 'M', '/м/u' => 'm',
100 '/Н/u' => 'N', '/н/u' => 'n', '/Ң/u' => 'Ñ', '/ң/u' => 'ñ',
101 '/О/u' => 'O', '/о/u' => 'o', '/Ө/u' => 'Ö', '/ө/u' => 'ö',
102 '/П/u' => 'P', '/п/u' => 'p', '/Р/u' => 'R', '/р/u' => 'r',
103 '/С/u' => 'S', '/с/u' => 's', '/Т/u' => 'T', '/т/u' => 't',
104 '/У/u' => 'W', '/у/u' => 'w', '/Ұ/u' => 'U', '/ұ/u' => 'u',
105 '/Ү/u' => 'Ü', '/ү/u' => 'ü', '/Ф/u' => 'F', '/ф/u' => 'f',
106 '/Х/u' => 'X', '/х/u' => 'x', '/Һ/u' => 'H', '/һ/u' => 'h',
107 '/Ц/u' => 'C', '/ц/u' => 'c', '/Ч/u' => 'Ç', '/ч/u' => 'ç',
108 '/Ш/u' => 'Ş', '/ш/u' => 'ş', '/Ы/u' => 'I', '/ы/u' => 'ı',
109 '/І/u' => 'İ', '/і/u' => 'i', '/Э/u' => 'É', '/э/u' => 'é',
112 $this->mLatn2Cyrl
= array(
116 '/ŞÇʹ/u' => 'ЩЬ', '/Şçʹ/u' => 'Щь',
117 '/Ş[Çç]/u' => 'Щ', '/şç/u' => 'щ',
118 # # soft and hard signs
119 '/([' . KK_L_UC
. '])ʺ([' . KK_L_UC
. '])/u' => '$1Ъ$2',
120 '/ʺ([' . KK_L_LC
. '])/u' => 'ъ$1',
121 '/([' . KK_L_UC
. '])ʹ([' . KK_L_UC
. '])/u' => '$1Ь$2',
122 '/ʹ([' . KK_L_LC
. '])/u' => 'ь$1',
126 '/Y[Ee]/u' => 'Е', '/ye/u' => 'е',
127 '/Y[Oo]/u' => 'Ё', '/yo/u' => 'ё',
128 '/Y[UWuw]/u' => 'Ю', '/y[uw]/u' => 'ю',
129 '/Y[Aa]/u' => 'Я', '/ya/u' => 'я',
131 '/A/u' => 'А', '/a/u' => 'а', '/Ä/u' => 'Ә', '/ä/u' => 'ә',
132 '/B/u' => 'Б', '/b/u' => 'б', '/C/u' => 'Ц', '/c/u' => 'ц',
133 '/Ç/u' => 'Ч', '/ç/u' => 'ч', '/D/u' => 'Д', '/d/u' => 'д',
134 '/E/u' => 'Е', '/e/u' => 'е', '/É/u' => 'Э', '/é/u' => 'э',
135 '/F/u' => 'Ф', '/f/u' => 'ф', '/G/u' => 'Г', '/g/u' => 'г',
136 '/Ğ/u' => 'Ғ', '/ğ/u' => 'ғ', '/H/u' => 'Һ', '/h/u' => 'һ',
137 '/I/u' => 'Ы', '/ı/u' => 'ы', '/İ/u' => 'І', '/i/u' => 'і',
138 '/Ï/u' => 'И', '/ï/u' => 'и', '/J/u' => 'Ж', '/j/u' => 'ж',
139 '/K/u' => 'К', '/k/u' => 'к', '/L/u' => 'Л', '/l/u' => 'л',
140 '/M/u' => 'М', '/m/u' => 'м', '/N/u' => 'Н', '/n/u' => 'н',
141 '/Ñ/u' => 'Ң', '/ñ/u' => 'ң', '/O/u' => 'О', '/o/u' => 'о',
142 '/Ö/u' => 'Ө', '/ö/u' => 'ө', '/P/u' => 'П', '/p/u' => 'п',
143 '/Q/u' => 'Қ', '/q/u' => 'қ', '/R/u' => 'Р', '/r/u' => 'р',
144 '/S/u' => 'С', '/s/u' => 'с', '/Ş/u' => 'Ш', '/ş/u' => 'ш',
145 '/T/u' => 'Т', '/t/u' => 'т', '/U/u' => 'Ұ', '/u/u' => 'ұ',
146 '/Ü/u' => 'Ү', '/ü/u' => 'ү', '/V/u' => 'В', '/v/u' => 'в',
147 '/W/u' => 'У', '/w/u' => 'у', '/Ý/u' => 'Й', '/ý/u' => 'й',
148 '/X/u' => 'Х', '/x/u' => 'х', '/Z/u' => 'З', '/z/u' => 'з',
151 $this->mCyLa2Arab
= array(
152 # # Punctuation -> Arabic
153 '/#|№|No\./u' => '', # ؀
154 '/\,/' => '،', # ،
155 '/;/' => '؛', # ؛
156 '/\?/' => '؟', # ؟
157 '/%/' => '٪', # ٪
158 '/\*/' => '٭', # ٭
160 '/0/' => '۰', # ۰
161 '/1/' => '۱', # ۱
162 '/2/' => '۲', # ۲
163 '/3/' => '۳', # ۳
164 '/4/' => '۴', # ۴
165 '/5/' => '۵', # ۵
166 '/6/' => '۶', # ۶
167 '/7/' => '۷', # ۷
168 '/8/' => '۸', # ۸
169 '/9/' => '۹', # ۹
170 # # Cyrillic -> Arabic
172 '/([АӘЕЁИОӨҰҮЭЮЯЪЬ])е/ui' => '$1يە',
173 '/[еэ]/ui' => 'ە', '/[ъь]/ui' => '',
174 '/[аә]/ui' => 'ا', '/[оө]/ui' => 'و', '/[ұү]/ui' => 'ۇ', '/[ыі]/ui' => 'ى',
175 '/[и]/ui' => 'ىي', '/ё/ui' => 'يو', '/ю/ui' => 'يۋ', '/я/ui' => 'يا', '/[й]/ui' => 'ي',
176 '/ц/ui' => 'تس', '/щ/ui' => 'شش',
177 '/һ/ui' => 'ح', '/ч/ui' => 'تش',
178 # '/һ/ui' => 'ھ', '/ч/ui' => 'چ',
179 '/б/ui' => 'ب', '/в/ui' => 'ۆ', '/г/ui' => 'گ', '/ғ/ui' => 'ع',
180 '/д/ui' => 'د', '/ж/ui' => 'ج', '/з/ui' => 'ز', '/к/ui' => 'ك',
181 '/қ/ui' => 'ق', '/л/ui' => 'ل', '/м/ui' => 'م', '/н/ui' => 'ن',
182 '/ң/ui' => 'ڭ', '/п/ui' => 'پ', '/р/ui' => 'ر', '/с/ui' => 'س',
183 '/т/ui' => 'ت', '/у/ui' => 'ۋ', '/ф/ui' => 'ف', '/х/ui' => 'ح',
185 # # Latin -> Arabic // commented for now...
186 /*'/Allah/ui' => 'ﷲ',
187 '/[eé]/ui' => 'ە', '/[yý]/ui' => 'ي', '/[ʺʹ]/ui' => '',
188 '/[aä]/ui' => 'ا', '/[oö]/ui' => 'و', '/[uü]/ui' => 'ۇ',
189 '/[ï]/ui' => 'ىي', '/[ıIiİ]/u' => 'ى',
191 '/ç/ui' => 'تش', '/h/ui' => 'ح',
192 #'/ç/ui' => 'چ', '/h/ui' => 'ھ',
193 '/b/ui' => 'ب','/d/ui' => 'د',
194 '/f/ui' => 'ف', '/g/ui' => 'گ', '/ğ/ui' => 'ع',
195 '/j/ui' => 'ج', '/k/ui' => 'ك', '/l/ui' => 'ل', '/m/ui' => 'م',
196 '/n/ui' => 'ن', '/ñ/ui' => 'ڭ', '/p/ui' => 'پ', '/q/ui' => 'ق',
197 '/r/ui' => 'ر', '/s/ui' => 'س', '/ş/ui' => 'ش', '/t/ui' => 'ت',
198 '/v/ui' => 'ۆ', '/w/ui' => 'ۋ', '/x/ui' => 'ح', '/z/ui' => 'ز',*/
203 * rules should be defined as -{ekavian | iyekavian-} -or-
204 * -{code:text | code:text | ...}-
206 * update: delete all rule parsing because it's not used
207 * currently, and just produces a couple of bugs
209 * @param $rule string
210 * @param $flags array
213 function parseManualRule( $rule, $flags = array() ) {
214 if ( in_array( 'T', $flags ) ) {
215 return parent
::parseManualRule( $rule, $flags );
219 // otherwise ignore all formatting
220 foreach ( $this->mVariants
as $v ) {
228 * A function wrapper:
229 * - if there is no selected variant, leave the link
231 * - do not try to find variants for usernames
233 * @param $link string
235 * @param bool $ignoreOtherCond
237 function findVariantLink( &$link, &$nt, $ignoreOtherCond = false ) {
238 // check for user namespace
239 if ( is_object( $nt ) ) {
240 $ns = $nt->getNamespace();
241 if ( $ns == NS_USER ||
$ns == NS_USER_TALK
)
246 parent
::findVariantLink( $link, $nt, $ignoreOtherCond );
247 if ( $this->getPreferredVariant() == $this->mMainLanguageCode
)
252 * An ugly function wrapper for parsing Image titles
253 * (to prevent image name conversion)
255 * @param $text string
256 * @param $toVariant bool
260 function autoConvert( $text, $toVariant = false ) {
262 if ( is_object( $wgTitle ) && $wgTitle->getNameSpace() == NS_FILE
) {
263 $imagename = $wgTitle->getNsText();
264 if ( preg_match( "/^$imagename:/", $text ) ) return $text;
266 return parent
::autoConvert( $text, $toVariant );
270 * It translates text into variant
272 * @param $text string
273 * @param $toVariant string
277 function translate( $text, $toVariant ) {
278 global $wgLanguageCode;
279 $text = parent
::translate( $text, $toVariant );
281 switch( $toVariant ) {
284 $letters = KK_L_UC
. KK_L_LC
. 'ʺʹ#0123456789';
285 $wgLanguageCode = 'kk';
289 $letters = KK_C_UC
. KK_C_LC
. '№0123456789';
290 $wgLanguageCode = 'kk-Latn';
294 $letters = KK_C_UC
. KK_C_LC
. /*KK_L_UC.KK_L_LC.'ʺʹ'.*/',;\?%\*№0123456789';
295 $wgLanguageCode = 'kk-Arab';
298 $wgLanguageCode = 'kk';
301 // disable conversion variables like $1, $2...
302 $varsfix = '\$[0-9]';
304 $matches = preg_split( '/' . $varsfix . '[^' . $letters . ']+/u', $text, -1, PREG_SPLIT_OFFSET_CAPTURE
);
307 foreach ( $matches as $m ) {
308 $ret .= substr( $text, $mstart, $m[1] -$mstart );
309 $ret .= $this->regsConverter( $m[0], $toVariant );
310 $mstart = $m[1] +
strlen( $m[0] );
316 * @param $text string
317 * @param $toVariant string
318 * @return mixed|string
320 function regsConverter( $text, $toVariant ) {
325 switch( $toVariant ) {
328 $letters = KK_C_LC
. KK_C_UC
/*.KK_L_LC.KK_L_UC*/;
329 $front = 'әөүіӘӨҮІ'/*.'äöüiÄÖÜİ'*/;
330 $excludes = 'еэгғкқЕЭГҒКҚ'/*.'eégğkqEÉGĞKQ'*/;
331 // split text to words
332 $matches = preg_split( '/[\b\s\-\.:]+/', $text, -1, PREG_SPLIT_OFFSET_CAPTURE
);
335 foreach ( $matches as $m ) {
336 $ret .= substr( $text, $mstart, $m[1] - $mstart );
337 // is matched the word to front vowels?
338 // exclude a words matched to е, э, г, к, к, қ,
339 // them should be without hamza
340 if ( preg_match( '/[' . $front . ']/u', $m[0] ) && !preg_match( '/[' . $excludes . ']/u', $m[0] ) ) {
341 $ret .= preg_replace( '/[' . $letters . ']+/u', H_HAMZA
. '$0', $m[0] );
345 $mstart = $m[1] +
strlen( $m[0] );
348 foreach ( $this->mCyLa2Arab
as $pat => $rep ) {
349 $text = preg_replace( $pat, $rep, $text );
355 foreach ( $this->mCyrl2Latn
as $pat => $rep ) {
356 $text = preg_replace( $pat, $rep, $text );
362 foreach ( $this->mLatn2Cyrl
as $pat => $rep ) {
363 $text = preg_replace( $pat, $rep, $text );
373 * We want our external link captions to be converted in variants,
374 * so we return the original text instead -{$text}-, except for URLs
376 * @param $text string
377 * @param $noParse string|bool
381 function markNoConversion( $text, $noParse = false ) {
382 if ( $noParse ||
preg_match( "/^https?:\/\/|ftp:\/\/|irc:\/\//", $text ) )
383 return parent
::markNoConversion( $text );
391 function convertCategoryKey( $key ) {
392 return $this->autoConvert( $key, 'kk' );
398 * class that handles Cyrillic, Latin and Arabic scripts for Kazakh
399 * right now it only distinguish kk_cyrl, kk_latn, kk_arab and kk_kz, kk_tr, kk_cn.
403 class LanguageKk
extends LanguageKk_cyrl
{
405 function __construct() {
407 parent
::__construct();
409 $variants = array( 'kk', 'kk-cyrl', 'kk-latn', 'kk-arab', 'kk-kz', 'kk-tr', 'kk-cn' );
410 $variantfallbacks = array(
415 'kk-kz' => 'kk-cyrl',
416 'kk-tr' => 'kk-latn',
420 $this->mConverter
= new KkConverter( $this, 'kk', $variants, $variantfallbacks );
422 $wgHooks['ArticleSaveComplete'][] = $this->mConverter
;
426 * Work around for right-to-left direction support in kk-arab and kk-cn
431 $variant = $this->getPreferredVariant();
432 if ( $variant == 'kk-arab' ||
$variant == 'kk-cn' ) {
435 return parent
::isRTL();
440 * It fixes issue with ucfirst for transforming 'i' to 'İ'
442 * @param $string string
446 function ucfirst ( $string ) {
447 $variant = $this->getPreferredVariant();
448 if ( ( $variant == 'kk-latn' ||
$variant == 'kk-tr' ) && $string[0] == 'i' ) {
449 $string = 'İ' . substr( $string, 1 );
451 $string = parent
::ucfirst( $string );
457 * It fixes issue with lcfirst for transforming 'I' to 'ı'
459 * @param $string string
463 function lcfirst ( $string ) {
464 $variant = $this->getPreferredVariant();
465 if ( ( $variant == 'kk-latn' ||
$variant == 'kk-tr' ) && $string[0] == 'I' ) {
466 $string = 'ı' . substr( $string, 1 );
468 $string = parent
::lcfirst( $string );
474 * @param $word string
475 * @param $case string
478 function convertGrammar( $word, $case ) {
479 wfProfileIn( __METHOD__
);
481 $variant = $this->getPreferredVariant();
482 switch ( $variant ) {
485 $word = parent
::convertGrammarKk_arab( $word, $case );
489 $word = parent
::convertGrammarKk_latn( $word, $case );
495 $word = parent
::convertGrammarKk_cyrl( $word, $case );
498 wfProfileOut( __METHOD__
);