MediaWiki  master
IcuCollation.php
Go to the documentation of this file.
1 <?php
22 
26 class IcuCollation extends Collation {
28 
31 
33  private $mainCollator;
34 
36  private $locale;
37 
40 
42  private $useNumericCollation = false;
43 
46 
56  private static $cjkBlocks = [
57  [ 0x2E80, 0x2EFF ], // CJK Radicals Supplement
58  [ 0x2F00, 0x2FDF ], // Kangxi Radicals
59  [ 0x2FF0, 0x2FFF ], // Ideographic Description Characters
60  [ 0x3000, 0x303F ], // CJK Symbols and Punctuation
61  [ 0x31C0, 0x31EF ], // CJK Strokes
62  [ 0x3200, 0x32FF ], // Enclosed CJK Letters and Months
63  [ 0x3300, 0x33FF ], // CJK Compatibility
64  [ 0x3400, 0x4DBF ], // CJK Unified Ideographs Extension A
65  [ 0x4E00, 0x9FFF ], // CJK Unified Ideographs
66  [ 0xF900, 0xFAFF ], // CJK Compatibility Ideographs
67  [ 0xFE30, 0xFE4F ], // CJK Compatibility Forms
68  [ 0x20000, 0x2A6DF ], // CJK Unified Ideographs Extension B
69  [ 0x2A700, 0x2B73F ], // CJK Unified Ideographs Extension C
70  [ 0x2B740, 0x2B81F ], // CJK Unified Ideographs Extension D
71  [ 0x2F800, 0x2FA1F ], // CJK Compatibility Ideographs Supplement
72  ];
73 
95  private static $tailoringFirstLetters = [
96  'af' => [],
97  'am' => [],
98  'ar' => [],
99  'as' => [ "\u{0982}", "\u{0981}", "\u{0983}", "\u{09CE}", "ক্ষ " ],
100  'ast' => [ "Ch", "Ll", "Ñ" ], // not in libicu
101  'az' => [ "Ç", "Ə", "Ğ", "İ", "Ö", "Ş", "Ü" ],
102  'be' => [ "Ё" ],
103  'be-tarask' => [ "Ё" ],
104  'bg' => [],
105  'bn' => [ 'ং', 'ঃ', 'ঁ' ],
106  'bn@collation=traditional' => [
107  'ং', 'ঃ', 'ঁ', 'ক্', 'খ্', 'গ্', 'ঘ্', 'ঙ্', 'চ্', 'ছ্', 'জ্', 'ঝ্',
108  'ঞ্', 'ট্', 'ঠ্', 'ড্', 'ঢ্', 'ণ্', 'ৎ', 'থ্', 'দ্', 'ধ্', 'ন্', 'প্',
109  'ফ্', 'ব্', 'ভ্', 'ম্', 'য্', 'র্', 'ৰ্', 'ল্', 'ৱ্', 'শ্', 'ষ্', 'স্', 'হ্'
110  ],
111  'bo' => [],
112  'br' => [ "Ch", "C'h" ],
113  'bs' => [ "Č", "Ć", "Dž", "Đ", "Lj", "Nj", "Š", "Ž" ],
114  'bs-Cyrl' => [],
115  'ca' => [],
116  'chr' => [],
117  'co' => [], // not in libicu
118  'cs' => [ "Č", "Ch", "Ř", "Š", "Ž" ],
119  'cy' => [ "Ch", "Dd", "Ff", "Ng", "Ll", "Ph", "Rh", "Th" ],
120  'da' => [ "Æ", "Ø", "Å" ],
121  'de' => [],
122  'de-AT@collation=phonebook' => [ 'ä', 'ö', 'ü', 'ß' ],
123  'dsb' => [ "Č", "Ć", "Dź", "Ě", "Ch", "Ł", "Ń", "Ŕ", "Š", "Ś", "Ž", "Ź" ],
124  'ee' => [ "Dz", "Ɖ", "Ɛ", "Ƒ", "Gb", "Ɣ", "Kp", "Ny", "Ŋ", "Ɔ", "Ts", "Ʋ" ],
125  'el' => [],
126  'en' => [],
127  'eo' => [ "Ĉ", "Ĝ", "Ĥ", "Ĵ", "Ŝ", "Ŭ" ],
128  'es' => [ "Ñ" ],
129  'et' => [ "Š", "Ž", "Õ", "Ä", "Ö", "Ü" ],
130  'eu' => [ "Ñ" ], // not in libicu
131  'fa' => [
132  // RTL, let's put each letter on a new line
133  "آ",
134  "ء",
135  "ه",
136  "ا",
137  "و"
138  ],
139  'fi' => [ "Å", "Ä", "Ö" ],
140  'fil' => [ "Ñ", "Ng" ],
141  'fo' => [ "Á", "Ð", "Í", "Ó", "Ú", "Ý", "Æ", "Ø", "Å" ],
142  'fr' => [],
143  'fr-CA' => [], // fr-CA sorts accents slightly different from fr.
144  'fur' => [ "À", "Á", "Â", "È", "Ì", "Ò", "Ù" ], // not in libicu
145  'fy' => [], // not in libicu
146  'ga' => [],
147  'gd' => [], // not in libicu
148  'gl' => [ "Ch", "Ll", "Ñ" ],
149  'gu' => [ "\u{0A82}", "\u{0A83}", "\u{0A81}", "\u{0AB3}" ],
150  'ha' => [ 'Ɓ', 'Ɗ', 'Ƙ', 'Sh', 'Ts', 'Ƴ' ],
151  'haw' => [ 'ʻ' ],
152  'he' => [],
153  'hi' => [ "\u{0902}", "\u{0903}" ],
154  'hr' => [ "Č", "Ć", "Dž", "Đ", "Lj", "Nj", "Š", "Ž" ],
155  'hsb' => [ "Č", "Dź", "Ě", "Ch", "Ł", "Ń", "Ř", "Š", "Ć", "Ž" ],
156  'hu' => [ "Cs", "Dz", "Dzs", "Gy", "Ly", "Ny", "Ö", "Sz", "Ty", "Ü", "Zs" ],
157  'hy' => [ "և" ],
158  'id' => [],
159  'ig' => [ "Ch", "Gb", "Gh", "Gw", "Ị", "Kp", "Kw", "Ṅ", "Nw", "Ny", "Ọ", "Sh", "Ụ" ],
160  'is' => [ "Á", "Ð", "É", "Í", "Ó", "Ú", "Ý", "Þ", "Æ", "Ö", "Å" ],
161  'it' => [],
162  'ka' => [],
163  'kk' => [ "Ү", "І" ],
164  'kl' => [ "Æ", "Ø", "Å" ],
165  'km' => [
166  "រ", "ឫ", "ឬ", "ល", "ឭ", "ឮ", "\u{17BB}\u{17C6}",
167  "\u{17C6}", "\u{17B6}\u{17C6}", "\u{17C7}",
168  "\u{17B7}\u{17C7}", "\u{17BB}\u{17C7}",
169  "\u{17C1}\u{17C7}", "\u{17C4}\u{17C7}",
170  ],
171  'kn' => [ "\u{0C81}", "\u{0C83}", "\u{0CF1}", "\u{0CF2}" ],
172  'kok' => [ "\u{0902}", "\u{0903}", "ळ", "क्ष" ],
173  'ku' => [ "Ç", "Ê", "Î", "Ş", "Û" ], // not in libicu
174  'ky' => [ "Ё" ],
175  'la' => [], // not in libicu
176  'lb' => [],
177  'lkt' => [ 'Č', 'Ǧ', 'Ȟ', 'Š', 'Ž' ],
178  'ln' => [ 'Ɛ' ],
179  'lo' => [],
180  'lt' => [ "Č", "Š", "Ž" ],
181  'lv' => [ "Č", "Ģ", "Ķ", "Ļ", "Ņ", "Š", "Ž" ],
182  'mk' => [ "Ѓ", "Ќ" ],
183  'ml' => [],
184  'mn' => [],
185  'mo' => [ "Ă", "Â", "Î", "Ș", "Ț" ], // not in libicu
186  'mr' => [ "\u{0902}", "\u{0903}", "ळ", "क्ष", "ज्ञ" ],
187  'ms' => [],
188  'mt' => [ "Ċ", "Ġ", "Għ", "Ħ", "Ż" ],
189  'nb' => [ "Æ", "Ø", "Å" ],
190  'ne' => [],
191  'nl' => [],
192  'nn' => [ "Æ", "Ø", "Å" ],
193  'no' => [ "Æ", "Ø", "Å" ], // not in libicu. You should probably use nb or nn instead.
194  'oc' => [], // not in libicu
195  'om' => [ 'Ch', 'Dh', 'Kh', 'Ny', 'Ph', 'Sh' ],
196  'or' => [ "\u{0B01}", "\u{0B02}", "\u{0B03}", "କ୍ଷ" ],
197  'pa' => [ "\u{0A4D}" ],
198  'pl' => [ "Ą", "Ć", "Ę", "Ł", "Ń", "Ó", "Ś", "Ź", "Ż" ],
199  'pt' => [],
200  'rm' => [], // not in libicu
201  'ro' => [ "Ă", "Â", "Î", "Ș", "Ț" ],
202  'ru' => [],
203  'rup' => [ "Ă", "Â", "Î", "Ľ", "Ń", "Ș", "Ț" ], // not in libicu
204  'sco' => [],
205  'se' => [
206  'Á', 'Č', 'Ʒ', 'Ǯ', 'Đ', 'Ǧ', 'Ǥ', 'Ǩ', 'Ŋ',
207  'Š', 'Ŧ', 'Ž', 'Ø', 'Æ', 'Ȧ', 'Ä', 'Ö'
208  ],
209  'si' => [ "\u{0D82}", "\u{0D83}", "\u{0DA4}" ],
210  'sk' => [ "Ä", "Č", "Ch", "Ô", "Š", "Ž" ],
211  'sl' => [ "Č", "Š", "Ž" ],
212  'smn' => [ "Á", "Č", "Đ", "Ŋ", "Š", "Ŧ", "Ž", "Æ", "Ø", "Å", "Ä", "Ö" ],
213  'sq' => [ "Ç", "Dh", "Ë", "Gj", "Ll", "Nj", "Rr", "Sh", "Th", "Xh", "Zh" ],
214  'sr' => [],
215  'sr-Latn' => [ "Č", "Ć", "Dž", "Đ", "Lj", "Nj", "Š", "Ž" ],
216  'sv' => [ "Å", "Ä", "Ö" ],
217  'sv@collation=standard' => [ "Å", "Ä", "Ö" ],
218  'sw' => [],
219  'ta' => [
220  "\u{0B82}", "ஃ", "க்ஷ", "க்", "ங்", "ச்", "ஞ்", "ட்", "ண்", "த்", "ந்",
221  "ப்", "ம்", "ய்", "ர்", "ல்", "வ்", "ழ்", "ள்", "ற்", "ன்", "ஜ்", "ஶ்", "ஷ்",
222  "ஸ்", "ஹ்", "க்ஷ்"
223  ],
224  'te' => [ "\u{0C01}", "\u{0C02}", "\u{0C03}" ],
225  'th' => [ "ฯ", "\u{0E46}", "\u{0E4D}", "\u{0E3A}" ],
226  'tk' => [ "Ç", "Ä", "Ž", "Ň", "Ö", "Ş", "Ü", "Ý" ],
227  'tl' => [ "Ñ", "Ng" ], // not in libicu
228  'to' => [ "Ng", "ʻ" ],
229  'tr' => [ "Ç", "Ğ", "İ", "Ö", "Ş", "Ü" ],
230  '-tr' => [ "ı" ],
231  'tt' => [ "Ә", "Ө", "Ү", "Җ", "Ң", "Һ" ], // not in libicu
232  'uk' => [ "Ґ", "Ь" ],
233  'uz' => [ "Ch", "G'", "Ng", "O'", "Sh" ], // not in libicu
234  'vi' => [ "Ă", "Â", "Đ", "Ê", "Ô", "Ơ", "Ư" ],
235  'vo' => [ "Ä", "Ö", "Ü" ],
236  'yi' => [
237  "\u{05D1}\u{05BF}", "\u{05DB}\u{05BC}", "\u{05E4}\u{05BC}",
238  "\u{05E9}\u{05C2}", "\u{05EA}\u{05BC}"
239  ],
240  'yo' => [ "Ẹ", "Gb", "Ọ", "Ṣ" ],
241  'zu' => [],
242  ];
243 
247  const RECORD_LENGTH = 14;
248 
249  public function __construct( $locale ) {
250  if ( !extension_loaded( 'intl' ) ) {
251  throw new MWException( 'An ICU collation was requested, ' .
252  'but the intl extension is not available.' );
253  }
254 
255  $this->locale = $locale;
256  // Drop everything after the '@' in locale's name
257  $localeParts = explode( '@', $locale );
258  $this->digitTransformLanguage = MediaWikiServices::getInstance()->getLanguageFactory()
259  ->getLanguage( $locale === 'root' ? 'en' : $localeParts[0] );
260 
261  $this->mainCollator = Collator::create( $locale );
262  if ( !$this->mainCollator ) {
263  throw new MWException( "Invalid ICU locale specified for collation: $locale" );
264  }
265 
266  $this->primaryCollator = Collator::create( $locale );
267  $this->primaryCollator->setStrength( Collator::PRIMARY );
268 
269  // If the special suffix for numeric collation is present, turn on numeric collation.
270  if ( substr( $locale, -5, 5 ) === '-u-kn' ) {
271  $this->useNumericCollation = true;
272  // Strip off the special suffix so it doesn't trip up fetchFirstLetterData().
273  $this->locale = substr( $this->locale, 0, -5 );
274  $this->mainCollator->setAttribute( Collator::NUMERIC_COLLATION, Collator::ON );
275  $this->primaryCollator->setAttribute( Collator::NUMERIC_COLLATION, Collator::ON );
276  }
277  }
278 
279  public function getSortKey( $string ) {
280  return $this->mainCollator->getSortKey( $string );
281  }
282 
283  public function getPrimarySortKey( $string ) {
284  return $this->primaryCollator->getSortKey( $string );
285  }
286 
287  public function getFirstLetter( $string ) {
288  $string = strval( $string );
289  if ( $string === '' ) {
290  return '';
291  }
292 
293  $firstChar = mb_substr( $string, 0, 1, 'UTF-8' );
294 
295  // If the first character is a CJK character, just return that character.
296  if ( ord( $firstChar ) > 0x7f && self::isCjk( UtfNormal\Utils::utf8ToCodepoint( $firstChar ) ) ) {
297  return $firstChar;
298  }
299 
300  $sortKey = $this->getPrimarySortKey( $string );
301 
302  // Do a binary search to find the correct letter to sort under
304  [ $this, 'getSortKeyByLetterIndex' ],
305  $this->getFirstLetterCount(),
306  'strcmp',
307  $sortKey );
308 
309  if ( $min === false ) {
310  // Before the first letter
311  return '';
312  }
313 
314  $sortLetter = $this->getLetterByIndex( $min );
315 
316  if ( $this->useNumericCollation ) {
317  // If the sort letter is a number, return '0–9' (or localized equivalent).
318  // ASCII value of 0 is 48. ASCII value of 9 is 57.
319  // Note that this also applies to non-Arabic numerals since they are
320  // mapped to Arabic numeral sort letters. For example, ২ sorts as 2.
321  if ( ord( $sortLetter ) >= 48 && ord( $sortLetter ) <= 57 ) {
322  $sortLetter = wfMessage( 'category-header-numerals' )->numParams( 0, 9 )->text();
323  }
324  }
325  return $sortLetter;
326  }
327 
332  public function getFirstLetterData() {
333  if ( $this->firstLetterData === null ) {
335  $cacheKey = $cache->makeKey(
336  'first-letters',
337  static::class,
338  $this->locale,
339  $this->digitTransformLanguage->getCode(),
340  INTL_ICU_VERSION,
342  );
343  $this->firstLetterData = $cache->getWithSetCallback( $cacheKey, $cache::TTL_WEEK, function () {
344  return $this->fetchFirstLetterData();
345  } );
346  }
347  return $this->firstLetterData;
348  }
349 
354  private function fetchFirstLetterData() {
355  global $IP;
356  // Generate data from serialized data file
357  if ( isset( self::$tailoringFirstLetters[$this->locale] ) ) {
358  $letters = require "$IP/includes/collation/data/first-letters-root.php";
359  // Append additional characters
360  $letters = array_merge( $letters, self::$tailoringFirstLetters[$this->locale] );
361  // Remove unnecessary ones, if any
362  if ( isset( self::$tailoringFirstLetters['-' . $this->locale] ) ) {
363  $letters = array_diff( $letters, self::$tailoringFirstLetters['-' . $this->locale] );
364  }
365  // Apply digit transforms
366  $digits = [ '0', '1', '2', '3', '4', '5', '6', '7', '8', '9' ];
367  $letters = array_diff( $letters, $digits );
368  foreach ( $digits as $digit ) {
369  $letters[] = $this->digitTransformLanguage->formatNum( $digit, true );
370  }
371  } elseif ( $this->locale === 'root' ) {
372  $letters = require "$IP/includes/collation/data/first-letters-root.php";
373  } else {
374  // FIXME: Is this still used?
375  $letters = wfGetPrecompiledData( "first-letters-{$this->locale}.ser" );
376  if ( $letters === false ) {
377  throw new MWException( "MediaWiki does not support ICU locale " .
378  "\"{$this->locale}\"" );
379  }
380  }
381 
382  /* Sort the letters.
383  *
384  * It's impossible to have the precompiled data file properly sorted,
385  * because the sort order changes depending on ICU version. If the
386  * array is not properly sorted, the binary search will return random
387  * results.
388  *
389  * We also take this opportunity to remove primary collisions.
390  */
391  $letterMap = [];
392  foreach ( $letters as $letter ) {
393  $key = $this->getPrimarySortKey( $letter );
394  if ( isset( $letterMap[$key] ) ) {
395  // Primary collision (two characters with the same sort position).
396  // Keep whichever one sorts first in the main collator.
397  $comp = $this->mainCollator->compare( $letter, $letterMap[$key] );
398  wfDebug( "Primary collision '$letter' '{$letterMap[$key]}' (comparison: $comp)\n" );
399  // If that also has a collision, use codepoint as a tiebreaker.
400  if ( $comp === 0 ) {
401  $comp = UtfNormal\Utils::utf8ToCodepoint( $letter ) <=>
402  UtfNormal\Utils::utf8ToCodepoint( $letterMap[$key] );
403  }
404  if ( $comp < 0 ) {
405  $letterMap[$key] = $letter;
406  }
407  } else {
408  $letterMap[$key] = $letter;
409  }
410  }
411  ksort( $letterMap, SORT_STRING );
412 
413  /* Remove duplicate prefixes. Basically if something has a sortkey
414  * which is a prefix of some other sortkey, then it is an
415  * expansion and probably should not be considered a section
416  * header.
417  *
418  * For example 'þ' is sometimes sorted as if it is the letters
419  * 'th'. Other times it is its own primary element. Another
420  * example is '₨'. Sometimes its a currency symbol. Sometimes it
421  * is an 'R' followed by an 's'.
422  *
423  * Additionally an expanded element should always sort directly
424  * after its first element due to they way sortkeys work.
425  *
426  * UCA sortkey elements are of variable length but no collation
427  * element should be a prefix of some other element, so I think
428  * this is safe. See:
429  * - https://ssl.icu-project.org/repos/icu/icuhtml/trunk/design/collation/ICU_collation_design.htm
430  * - http://site.icu-project.org/design/collation/uca-weight-allocation
431  *
432  * Additionally, there is something called primary compression to
433  * worry about. Basically, if you have two primary elements that
434  * are more than one byte and both start with the same byte then
435  * the first byte is dropped on the second primary. Additionally
436  * either \x03 or \xFF may be added to mean that the next primary
437  * does not start with the first byte of the first primary.
438  *
439  * This shouldn't matter much, as the first primary is not
440  * changed, and that is what we are comparing against.
441  *
442  * tl;dr: This makes some assumptions about how icu implements
443  * collations. It seems incredibly unlikely these assumptions
444  * will change, but nonetheless they are assumptions.
445  */
446 
447  $prev = false;
448  $duplicatePrefixes = [];
449  foreach ( $letterMap as $key => $value ) {
450  // Remove terminator byte. Otherwise the prefix
451  // comparison will get hung up on that.
452  $trimmedKey = rtrim( $key, "\0" );
453  if ( $prev === false || $prev === '' ) {
454  $prev = $trimmedKey;
455  // We don't yet have a collation element
456  // to compare against, so continue.
457  continue;
458  }
459 
460  // Due to the fact the array is sorted, we only have
461  // to compare with the element directly previous
462  // to the current element (skipping expansions).
463  // An element "X" will always sort directly
464  // before "XZ" (Unless we have "XY", but we
465  // do not update $prev in that case).
466  if ( substr( $trimmedKey, 0, strlen( $prev ) ) === $prev ) {
467  $duplicatePrefixes[] = $key;
468  // If this is an expansion, we don't want to
469  // compare the next element to this element,
470  // but to what is currently $prev
471  continue;
472  }
473  $prev = $trimmedKey;
474  }
475  foreach ( $duplicatePrefixes as $badKey ) {
476  wfDebug( "Removing '{$letterMap[$badKey]}' from first letters.\n" );
477  unset( $letterMap[$badKey] );
478  // This code assumes that unsetting does not change sort order.
479  }
480  $data = [
481  'chars' => array_values( $letterMap ),
482  'keys' => array_keys( $letterMap ),
483  ];
484 
485  // Reduce memory usage before caching
486  unset( $letterMap );
487 
488  return $data;
489  }
490 
496  public function getLetterByIndex( $index ) {
497  return $this->getFirstLetterData()['chars'][$index];
498  }
499 
505  public function getSortKeyByLetterIndex( $index ) {
506  return $this->getFirstLetterData()['keys'][$index];
507  }
508 
513  public function getFirstLetterCount() {
514  return count( $this->getFirstLetterData()['chars'] );
515  }
516 
523  public static function isCjk( $codepoint ) {
524  foreach ( self::$cjkBlocks as $block ) {
525  if ( $codepoint >= $block[0] && $codepoint <= $block[1] ) {
526  return true;
527  }
528  }
529  return false;
530  }
531 
539  static function getUnicodeVersionForICU() {
540  $icuVersion = INTL_ICU_VERSION;
541  if ( !$icuVersion ) {
542  return false;
543  }
544 
545  $versionPrefix = substr( $icuVersion, 0, 3 );
546  // Source: http://site.icu-project.org/download
547  $map = [
548  '63.' => '11.0',
549  '62.' => '11.0',
550  '61.' => '10.0',
551  '60.' => '10.0',
552  '59.' => '9.0',
553  '58.' => '9.0',
554  '57.' => '8.0',
555  '56.' => '8.0',
556  '55.' => '7.0',
557  '54.' => '7.0',
558  '53.' => '6.3',
559  '52.' => '6.3',
560  '51.' => '6.2',
561  '50.' => '6.2',
562  '49.' => '6.1',
563  '4.8' => '6.0',
564  '4.6' => '6.0',
565  '4.4' => '5.2',
566  '4.2' => '5.1',
567  '4.0' => '5.1',
568  '3.8' => '5.0',
569  '3.6' => '5.0',
570  '3.4' => '4.1',
571  ];
572 
573  return $map[$versionPrefix] ?? false;
574  }
575 }
MediaWiki\MediaWikiServices
MediaWikiServices is the service locator for the application scope of MediaWiki.
Definition: MediaWikiServices.php:130
IcuCollation\getFirstLetterCount
getFirstLetterCount()
Definition: IcuCollation.php:513
IcuCollation\$primaryCollator
Collator $primaryCollator
Definition: IcuCollation.php:30
IcuCollation\getPrimarySortKey
getPrimarySortKey( $string)
Definition: IcuCollation.php:283
wfMessage
wfMessage( $key,... $params)
This is the function for getting translated interface messages.
Definition: GlobalFunctions.php:1263
IcuCollation\$tailoringFirstLetters
static $tailoringFirstLetters
Additional characters (or character groups) to be considered separate letters for given languages,...
Definition: IcuCollation.php:95
Collation
Definition: Collation.php:29
IcuCollation\$mainCollator
Collator $mainCollator
Definition: IcuCollation.php:33
MWException
MediaWiki exception.
Definition: MWException.php:26
IcuCollation\$digitTransformLanguage
Language $digitTransformLanguage
Definition: IcuCollation.php:39
IcuCollation\getSortKeyByLetterIndex
getSortKeyByLetterIndex( $index)
Definition: IcuCollation.php:505
IcuCollation\isCjk
static isCjk( $codepoint)
Test if a code point is a CJK (Chinese, Japanese, Korean) character.
Definition: IcuCollation.php:523
wfDebug
wfDebug( $text, $dest='all', array $context=[])
Sends a line to the debug log if enabled or, optionally, to a comment in output.
Definition: GlobalFunctions.php:913
IcuCollation\getLetterByIndex
getLetterByIndex( $index)
Definition: IcuCollation.php:496
IcuCollation\getFirstLetter
getFirstLetter( $string)
Given a string, return the logical "first letter" to be used for grouping on category pages and so on...
Definition: IcuCollation.php:287
CACHE_ANYTHING
const CACHE_ANYTHING
Definition: Defines.php:81
IcuCollation\getFirstLetterData
getFirstLetterData()
Definition: IcuCollation.php:332
IcuCollation\RECORD_LENGTH
const RECORD_LENGTH
Definition: IcuCollation.php:247
IcuCollation\FIRST_LETTER_VERSION
const FIRST_LETTER_VERSION
Definition: IcuCollation.php:27
IcuCollation\$locale
string $locale
Definition: IcuCollation.php:36
IcuCollation\getUnicodeVersionForICU
static getUnicodeVersionForICU()
Return the version of Unicode appropriate for the version of ICU library currently in use,...
Definition: IcuCollation.php:539
$cache
$cache
Definition: mcc.php:33
wfGetPrecompiledData
wfGetPrecompiledData( $name)
Get an object from the precompiled serialized directory.
Definition: GlobalFunctions.php:2481
IcuCollation\$useNumericCollation
bool $useNumericCollation
Definition: IcuCollation.php:42
ArrayUtils\findLowerBound
static findLowerBound( $valueCallback, $valueCount, $comparisonCallback, $target)
Do a binary search, and return the index of the largest item that sorts less than or equal to the tar...
Definition: ArrayUtils.php:112
IcuCollation\getSortKey
getSortKey( $string)
Given a string, convert it to a (hopefully short) key that can be used for efficient sorting.
Definition: IcuCollation.php:279
IcuCollation\$cjkBlocks
static $cjkBlocks
Unified CJK blocks.
Definition: IcuCollation.php:56
IcuCollation\$firstLetterData
array $firstLetterData
Definition: IcuCollation.php:45
IcuCollation\__construct
__construct( $locale)
Definition: IcuCollation.php:249
$IP
$IP
Definition: WebStart.php:41
Language
Internationalisation code.
Definition: Language.php:39
IcuCollation\fetchFirstLetterData
fetchFirstLetterData()
Definition: IcuCollation.php:354
ObjectCache\getLocalServerInstance
static getLocalServerInstance( $fallback=CACHE_NONE)
Factory function for CACHE_ACCEL (referenced from DefaultSettings.php)
Definition: ObjectCache.php:249
IcuCollation
Definition: IcuCollation.php:26