8 require_once( dirname(__FILE__
).'/../LanguageConverter.php' );
9 require_once( dirname(__FILE__
).'/LanguageKk_cyrl.php' );
11 define( 'KK_C_UC', 'АӘБВГҒДЕЁЖЗИЙКҚЛМНҢОӨПРСТУҰҮФХҺЦЧШЩЪЫІЬЭЮЯ' ); # Kazakh Cyrillic uppercase
12 define( 'KK_C_LC', 'аәбвгғдеёжзийкқлмнңоөпрстуұүфхһцчшщъыіьэюя' ); # Kazakh Cyrillic lowercase
13 define( 'KK_L_UC', 'AÄBCÇDEÉFGĞHIİÏJKLMNÑOÖPQRSŞTUÜVWXYÝZ' ); # Kazakh Latin uppercase
14 define( 'KK_L_LC', 'aäbcçdeéfgğhıiïjklmnñoöpqrsştuüvwxyýz' ); # Kazakh Latin lowercase
15 //define( 'KK_A', 'اٵبۆگعدەجزيكقلمنڭوٶپرستۋۇٷفحھچشىٸ' ); # Kazakh Arabic
17 class KkConverter
extends LanguageConverter
{
19 function __construct($langobj, $maincode,
21 $variantfallbacks=array(),
24 parent
::__construct( $langobj, $maincode,
25 $variants, $variantfallbacks, $markup, $flags );
27 // No point delaying this since they're in code.
28 // Waiting until loadDefaultTables() means they never get loaded
29 // when the tables themselves are loaded from cache.
33 function loadDefaultTables() {
34 // require( dirname(__FILE__)."/../../includes/KkConversion.php" );
35 // Placeholder for future implementing. Remove variables declarations
36 // after generating KkConversion.php
44 $this->mTables
= array(
45 'kk-cyrl' => new ReplacementArray( $kk2Cyrl ),
46 'kk-latn' => new ReplacementArray( $kk2Latn ),
47 'kk-arab' => new ReplacementArray( $kk2Arab ),
48 'kk-kz' => new ReplacementArray( array_merge($kk2Cyrl, $kk2KZ) ),
49 'kk-tr' => new ReplacementArray( array_merge($kk2Latn, $kk2TR) ),
50 'kk-cn' => new ReplacementArray( array_merge($kk2Arab, $kk2CN) ),
51 'kk' => new ReplacementArray()
55 function postLoadTables() {
56 $this->mTables
['kk-kz']->merge( $this->mTables
['kk-cyrl'] );
57 $this->mTables
['kk-tr']->merge( $this->mTables
['kk-latn'] );
58 $this->mTables
['kk-cn']->merge( $this->mTables
['kk-arab'] );
63 $this->mCyrl2Latn
= array(
66 '/([АӘЕЁИОӨҰҮЭЮЯЪЬ])Е/u' => '$1YE',
67 '/([АӘЕЁИОӨҰҮЭЮЯЪЬ])е/ui' => '$1ye',
69 '/^Ё(['.KK_C_UC
.']|$)/u' => 'YO$1', '/^Ё(['.KK_C_LC
.']|$)/u' => 'Yo$1',
70 '/^Ю(['.KK_C_UC
.']|$)/u' => 'YU$1', '/^Ю(['.KK_C_LC
.']|$)/u' => 'Yu$1',
71 '/^Я(['.KK_C_UC
.']|$)/u' => 'YA$1', '/^Я(['.KK_C_LC
.']|$)/u' => 'Ya$1',
72 '/^Щ(['.KK_C_UC
.']|$)/u' => 'ŞÇ$1', '/^Щ(['.KK_C_LC
.']|$)/u' => 'Şç$1',
74 '/Ё/u' => 'YO', '/ё/u' => 'yo',
75 '/Ю/u' => 'YU', '/ю/u' => 'yu',
76 '/Я/u' => 'YA', '/я/u' => 'ya',
77 '/Щ/u' => 'ŞÇ', '/щ/u' => 'şç',
78 ## soft and hard signs
79 '/[ъЪ]/u' => 'ʺ', '/[ьЬ]/u' => 'ʹ',
81 '/А/u' => 'A', '/а/u' => 'a', '/Ә/u' => 'Ä', '/ә/u' => 'ä',
82 '/Б/u' => 'B', '/б/u' => 'b', '/В/u' => 'V', '/в/u' => 'v',
83 '/Г/u' => 'G', '/г/u' => 'g', '/Ғ/u' => 'Ğ', '/ғ/u' => 'ğ',
84 '/Д/u' => 'D', '/д/u' => 'd', '/Е/u' => 'E', '/е/u' => 'e',
85 '/Ж/u' => 'J', '/ж/u' => 'j', '/З/u' => 'Z', '/з/u' => 'z',
86 '/И/u' => 'Ï', '/и/u' => 'ï', '/Й/u' => 'Ý', '/й/u' => 'ý',
87 '/К/u' => 'K', '/к/u' => 'k', '/Қ/u' => 'Q', '/қ/u' => 'q',
88 '/Л/u' => 'L', '/л/u' => 'l', '/М/u' => 'M', '/м/u' => 'm',
89 '/Н/u' => 'N', '/н/u' => 'n', '/Ң/u' => 'Ñ', '/ң/u' => 'ñ',
90 '/О/u' => 'O', '/о/u' => 'o', '/Ө/u' => 'Ö', '/ө/u' => 'ö',
91 '/П/u' => 'P', '/п/u' => 'p', '/Р/u' => 'R', '/р/u' => 'r',
92 '/С/u' => 'S', '/с/u' => 's', '/Т/u' => 'T', '/т/u' => 't',
93 '/У/u' => 'W', '/у/u' => 'w', '/Ұ/u' => 'U', '/ұ/u' => 'u',
94 '/Ү/u' => 'Ü', '/ү/u' => 'ü', '/Ф/u' => 'F', '/ф/u' => 'f',
95 '/Х/u' => 'X', '/х/u' => 'x', '/Һ/u' => 'H', '/һ/u' => 'h',
96 '/Ц/u' => 'C', '/ц/u' => 'c', '/Ч/u' => 'Ç', '/ч/u' => 'ç',
97 '/Ш/u' => 'Ş', '/ш/u' => 'ş', '/Ы/u' => 'I', '/ы/u' => 'ı',
98 '/І/u' => 'İ', '/і/u' => 'i', '/Э/u' => 'É', '/э/u' => 'é',
101 $this->mLatn2Cyrl
= array(
104 '/ŞÇʹ/u'=> 'ЩЬ', '/Şçʹ/u'=> 'Щь', '/Şçʹ/u'=> 'Щь',
105 '/Ş[Çç]/u' => 'Щ', '/şç/u' => 'щ',
106 ## soft and hard signs
107 '/(['.KK_L_UC
.'])ʺ(['.KK_L_UC
.'])/u' => '$1Ъ$2',
108 '/ʺ(['.KK_L_LC
.'])/u' => 'ъ$1',
109 '/(['.KK_L_UC
.'])ʹ(['.KK_L_UC
.'])/u' => '$1Ь$2',
110 '/ʹ(['.KK_L_LC
.'])/u' => 'ь$1',
114 '/Y[Ee]/u' => 'Е', '/ye/u' => 'е',
115 '/Y[Oo]/u' => 'Ё', '/yo/u' => 'ё',
116 '/Y[UWuw]/u' => 'Ю', '/y[uw]/u' => 'ю',
117 '/Y[Aa]/u' => 'Я', '/ya/u' => 'я',
119 '/A/u' => 'А', '/a/u' => 'а', '/Ä/u' => 'Ә', '/ä/u' => 'ә',
120 '/B/u' => 'Б', '/b/u' => 'б', '/C/u' => 'Ц', '/c/u' => 'ц',
121 '/Ç/u' => 'Ч', '/ç/u' => 'ч', '/D/u' => 'Д', '/d/u' => 'д',
122 '/E/u' => 'Е', '/e/u' => 'е', '/É/u' => 'Э', '/é/u' => 'э',
123 '/F/u' => 'Ф', '/f/u' => 'ф', '/G/u' => 'Г', '/g/u' => 'г',
124 '/Ğ/u' => 'Ғ', '/ğ/u' => 'ғ', '/H/u' => 'Һ', '/h/u' => 'һ',
125 '/I/u' => 'Ы', '/ı/u' => 'ы', '/İ/u' => 'І', '/i/u' => 'і',
126 '/Ï/u' => 'И', '/ï/u' => 'и', '/J/u' => 'Ж', '/j/u' => 'ж',
127 '/K/u' => 'К', '/k/u' => 'к', '/L/u' => 'Л', '/l/u' => 'л',
128 '/M/u' => 'М', '/m/u' => 'м', '/N/u' => 'Н', '/n/u' => 'н',
129 '/Ñ/u' => 'Ң', '/ñ/u' => 'ң', '/O/u' => 'О', '/o/u' => 'о',
130 '/Ö/u' => 'Ө', '/ö/u' => 'ө', '/P/u' => 'П', '/p/u' => 'п',
131 '/Q/u' => 'Қ', '/q/u' => 'қ', '/R/u' => 'Р', '/r/u' => 'р',
132 '/S/u' => 'С', '/s/u' => 'с', '/Ş/u' => 'Ш', '/ş/u' => 'ш',
133 '/T/u' => 'Т', '/t/u' => 'т', '/U/u' => 'Ұ', '/u/u' => 'ұ',
134 '/Ü/u' => 'Ү', '/ü/u' => 'ү', '/V/u' => 'В', '/v/u' => 'в',
135 '/W/u' => 'У', '/w/u' => 'у', '/Ý/u' => 'Й', '/ý/u' => 'й',
136 '/X/u' => 'Х', '/x/u' => 'х', '/Z/u' => 'З', '/z/u' => 'з',
139 $this->mCyLa2Arab
= array(
140 ## Cyrillic -> Arabic
142 '/([АӘЕЁИОӨҰҮЭЮЯЪЬ])е/ui' => '$1يە',
143 '/[еэ]/ui' => 'ە', '/[ъь]/ui' => '',
144 '/а/ui' => 'ا', '/ә/ui' => 'ٵ', '/б/ui' => 'ب', '/в/ui' => 'ۆ',
145 '/г/ui' => 'گ', '/ғ/ui' => 'ع', '/д/ui' => 'د', '/ё/ui' => 'يو',
146 '/ж/ui' => 'ج', '/з/ui' => 'ز', '/и/ui' => 'ي', '/й/ui' => 'ي',
147 '/к/ui' => 'ك', '/қ/ui' => 'ق', '/л/ui' => 'ل', '/м/ui' => 'م',
148 '/н/ui' => 'ن', '/ң/ui' => 'ڭ', '/о/ui' => 'و', '/ө/ui' => 'ٶ',
149 '/п/ui' => 'پ', '/р/ui' => 'ر', '/с/ui' => 'س', '/т/ui' => 'ت',
150 '/у/ui' => 'ۋ', '/ұ/ui' => 'ۇ', '/ү/ui' => 'ٷ', '/ф/ui' => 'ف',
151 '/х/ui' => 'ح', '/һ/ui' => 'ھ', '/ц/ui' => 'تس', '/ч/ui' => 'چ',
152 '/ш/ui' => 'ش', '/щ/ui' => 'شش', '/ы/ui' => 'ى', '/і/ui' => 'ٸ',
153 '/ю/ui' => 'يۋ', '/я/ui' => 'يا',
154 ## Latin -> Arabic // commented for now...
155 /*'/No\./u' => 'نٶ.',
156 '/[ıI]/u' => 'ى', '/[iİ]/u' => 'ٸ',
157 '/[eé]/ui' => 'ە', '/[yý]/ui' => 'ي',
159 '/a/ui' => 'ا', '/ä/ui' => 'ٵ', '/b/ui' => 'ب', '/c/ui' => 'تس',
160 '/ç/ui' => 'چ', '/d/ui' => 'د', '/f/ui' => 'ف', '/g/ui' => 'گ',
161 '/ğ/ui' => 'ع', '/h/ui' => 'ھ', '/ï/ui' => 'ي', '/j/ui' => 'ج',
162 '/k/ui' => 'ك', '/l/ui' => 'ل', '/m/ui' => 'م', '/n/ui' => 'ن',
163 '/ñ/ui' => 'ڭ', '/o/ui' => 'و', '/ö/ui' => 'ٶ', '/p/ui' => 'پ',
164 '/q/ui' => 'ق', '/r/ui' => 'ر', '/s/ui' => 'س', '/ş/ui' => 'ش',
165 '/t/ui' => 'ت', '/u/ui' => 'ۇ', '/ü/ui' => 'ٷ', '/v/ui' => 'ۆ',
166 '/w/ui' => 'ۋ', '/x/ui' => 'ح', '/z/ui' => 'ز',*/
167 ## Punctuation -> Arabic
168 '/\,/' => '،', # ،
169 '/;/' => '؛', # ؛
170 '/\?/' => '؟', # ؟
171 '/%/' => '٪', # ٪
172 '/\*/' => '٭', # ٭
174 '/0/' => '۰', # ۰
175 '/1/' => '۱', # ۱
176 '/2/' => '۲', # ۲
177 '/3/' => '۳', # ۳
178 '/4/' => '۴', # ۴
179 '/5/' => '۵', # ۵
180 '/6/' => '۶', # ۶
181 '/7/' => '۷', # ۷
182 '/8/' => '۸', # ۸
183 '/9/' => '۹', # ۹
188 /* rules should be defined as -{ekavian | iyekavian-} -or-
189 -{code:text | code:text | ...}-
190 update: delete all rule parsing because it's not used
191 currently, and just produces a couple of bugs
193 function parseManualRule($rule, $flags=array()) {
194 if(in_array('T',$flags)){
195 return parent
::parseManualRule($rule, $flags);
198 // otherwise ignore all formatting
199 foreach($this->mVariants
as $v) {
206 // Do not convert content on talk pages
207 function parserConvert( $text, &$parser ){
208 if(is_object($parser->getTitle() ) && $parser->getTitle()->isTalkPage())
209 $this->mDoContentConvert
=false;
211 $this->mDoContentConvert
=true;
213 return parent
::parserConvert($text, $parser );
217 * A function wrapper:
218 * - if there is no selected variant, leave the link
220 * - do not try to find variants for usernames
222 function findVariantLink( &$link, &$nt ) {
223 // check for user namespace
225 $ns = $nt->getNamespace();
226 if($ns==NS_USER ||
$ns==NS_USER_TALK
)
231 parent
::findVariantLink($link,$nt);
232 if($this->getPreferredVariant()==$this->mMainLanguageCode
)
237 * An ugly function wrapper for parsing Image titles
238 * (to prevent image name conversion)
240 function autoConvert($text, $toVariant=false) {
242 if(is_object($wgTitle) && $wgTitle->getNameSpace()==NS_IMAGE
){
243 $imagename = $wgTitle->getNsText();
244 if(preg_match("/^$imagename:/",$text)) return $text;
246 return parent
::autoConvert($text,$toVariant);
250 * It translates text into variant
252 function translate( $text, $toVariant ){
253 global $wgContLanguageCode;
254 $text = parent
::translate( $text, $toVariant );
257 switch( $toVariant ) {
260 $letters = KK_L_UC
. KK_L_LC
. 'ʺʹ0123456789';
261 $wgContLanguageCode = 'kk';
265 $letters = KK_C_UC
. KK_C_LC
. '№0123456789';
266 $wgContLanguageCode = 'kk-Latn';
270 // $letters = KK_C_UC.KK_C_LC.KK_L_UC.KK_L_LC.'ʺʹ%№0123456789?,;';
271 $letters = KK_C_UC
. KK_C_LC
. '%№0123456789?,;';
272 $wgContLanguageCode = 'kk-Arab';
275 $wgContLanguageCode = 'kk';
278 // disable conversion variables like $1, $2...
279 $varsfix = '\$[0-9]';
281 $matches = preg_split( '/' . $varsfix . '[^' . $letters . ']+/u', $text, -1, PREG_SPLIT_OFFSET_CAPTURE
);
284 foreach( $matches as $m ) {
285 $ret .= substr( $text, $mstart, $m[1]-$mstart );
286 $ret .= $this->regsConverter( $m[0], $toVariant );
287 $mstart = $m[1] +
strlen($m[0]);
292 function regsConverter( $text, $toVariant ) {
293 if ($text == '') return $text;
297 switch( $toVariant ) {
300 foreach( $this->mCyLa2Arab
as $pat => $rep ) {
301 $text = preg_replace( $pat, $rep, $text );
307 foreach( $this->mCyrl2Latn
as $pat => $rep ) {
308 $text = preg_replace( $pat, $rep, $text );
314 foreach( $this->mLatn2Cyrl
as $pat => $rep ) {
315 $text = preg_replace( $pat, $rep, $text );
325 * We want our external link captions to be converted in variants,
326 * so we return the original text instead -{$text}-, except for URLs
328 function markNoConversion( $text, $noParse=false ) {
329 if( $noParse ||
preg_match( "/^https?:\/\/|ftp:\/\/|irc:\/\//", $text ) )
330 return parent
::markNoConversion( $text );
334 function convertCategoryKey( $key ) {
335 return $this->autoConvert( $key, 'kk' );
340 /* class that handles Cyrillic, Latin and Arabic scripts for Kazakh
341 right now it only distinguish kk_cyrl, kk_latn, kk_arab and kk_kz, kk_tr, kk_cn.
343 class LanguageKk
extends LanguageKk_cyrl
{
345 function __construct() {
347 parent
::__construct();
349 $variants = array( 'kk', 'kk-cyrl', 'kk-latn', 'kk-arab', 'kk-kz', 'kk-tr', 'kk-cn' );
350 $variantfallbacks = array(
355 'kk-kz' => 'kk-cyrl',
356 'kk-tr' => 'kk-latn',
360 $this->mConverter
= new KkConverter( $this, 'kk', $variants, $variantfallbacks );
362 $wgHooks['ArticleSaveComplete'][] = $this->mConverter
;
366 * Work around for right-to-left direction support in kk-arab and kk-cn
371 $variant = $this->getPreferredVariant();
372 if ( $variant == 'kk-arab' ||
$variant == 'kk-cn' ) {
375 return parent
::isRTL();
380 * It fixes issue with ucfirst for transforming 'i' to 'İ'
383 function ucfirst ( $string ) {
384 $variant = $this->getPreferredVariant();
385 if ( ($variant == 'kk-latn' ||
$variant == 'kk-tr') && $string[0] == 'i' ) {
386 $string = 'İ' . substr( $string, 1 );
388 $string = parent
::ucfirst( $string );
394 * It fixes issue with lcfirst for transforming 'I' to 'ı'
397 function lcfirst ( $string ) {
398 $variant = $this->getPreferredVariant();
399 if ( ($variant == 'kk-latn' ||
$variant == 'kk-tr') && $string[0] == 'I' ) {
400 $string = 'ı' . substr( $string, 1 );
402 $string = parent
::lcfirst( $string );
407 function convertGrammar( $word, $case ) {
408 $fname="LanguageKk::convertGrammar";
409 wfProfileIn( $fname );
411 $variant = $this->getPreferredVariant();
412 switch ( $variant ) {
415 $word = parent
::convertGrammar( $word, $case, $variant = 'kk-arab' );
419 $word = parent
::convertGrammar( $word, $case, $variant = 'kk-latn' );
425 $word = parent
::convertGrammar( $word, $case, $variant = 'kk-cyrl' );
428 wfProfileOut( $fname );