MediaWiki  master
IcuCollation.php
Go to the documentation of this file.
1 <?php
22 
26 class IcuCollation extends Collation {
27  private const FIRST_LETTER_VERSION = 4;
28 
31 
33  private $mainCollator;
34 
36  private $locale;
37 
40 
42  private $useNumericCollation = false;
43 
46 
56  private const CJK_BLOCKS = [
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 const TAILORING_FIRST_LETTERS = [
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 
248  public function __construct(
249  LanguageFactory $languageFactory,
250  $locale
251  ) {
252  $this->locale = $locale;
253  // Drop everything after the '@' in locale's name
254  $localeParts = explode( '@', $locale );
255  $this->digitTransformLanguage = $languageFactory->getLanguage( $locale === 'root' ? 'en' : $localeParts[0] );
256 
257  $mainCollator = Collator::create( $locale );
258  if ( !$mainCollator ) {
259  throw new MWException( "Invalid ICU locale specified for collation: $locale" );
260  }
261  $this->mainCollator = $mainCollator;
262 
263  $this->primaryCollator = Collator::create( $locale );
264  $this->primaryCollator->setStrength( Collator::PRIMARY );
265 
266  // If the special suffix for numeric collation is present, turn on numeric collation.
267  if ( substr( $locale, -5, 5 ) === '-u-kn' ) {
268  $this->useNumericCollation = true;
269  // Strip off the special suffix so it doesn't trip up fetchFirstLetterData().
270  $this->locale = substr( $this->locale, 0, -5 );
271  $this->mainCollator->setAttribute( Collator::NUMERIC_COLLATION, Collator::ON );
272  $this->primaryCollator->setAttribute( Collator::NUMERIC_COLLATION, Collator::ON );
273  }
274  }
275 
276  public function getSortKey( $string ) {
277  return $this->mainCollator->getSortKey( $string );
278  }
279 
280  public function getFirstLetter( $string ) {
281  $string = strval( $string );
282  if ( $string === '' ) {
283  return '';
284  }
285 
286  $firstChar = mb_substr( $string, 0, 1, 'UTF-8' );
287 
288  // If the first character is a CJK character, just return that character.
289  if ( ord( $firstChar ) > 0x7f && self::isCjk( mb_ord( $firstChar ) ) ) {
290  return $firstChar;
291  }
292 
293  $sortKey = $this->getPrimarySortKey( $string );
294  $data = $this->getFirstLetterData();
295  $keys = $data['keys'];
296  $letters = $data['chars'];
297 
298  // Do a binary search to find the correct letter to sort under
300  static function ( $index ) use ( $keys ) {
301  return $keys[$index];
302  },
303  count( $keys ),
304  'strcmp',
305  $sortKey );
306 
307  if ( $min === false ) {
308  // Before the first letter
309  return '';
310  }
311 
312  $sortLetter = $letters[$min];
313 
314  if ( $this->useNumericCollation ) {
315  // If the sort letter is a number, return '0–9' (or localized equivalent).
316  // ASCII value of 0 is 48. ASCII value of 9 is 57.
317  // Note that this also applies to non-Arabic numerals since they are
318  // mapped to Arabic numeral sort letters. For example, ২ sorts as 2.
319  if ( ord( $sortLetter ) >= 48 && ord( $sortLetter ) <= 57 ) {
320  $sortLetter = wfMessage( 'category-header-numerals' )->numParams( 0, 9 )->text();
321  }
322  }
323  return $sortLetter;
324  }
325 
326  private function getPrimarySortKey( $string ) {
327  return $this->primaryCollator->getSortKey( $string );
328  }
329 
334  private function getFirstLetterData() {
335  if ( $this->firstLetterData === null ) {
337  $cacheKey = $cache->makeKey(
338  'first-letters',
339  static::class,
340  $this->locale,
341  $this->digitTransformLanguage->getCode(),
342  INTL_ICU_VERSION,
343  self::FIRST_LETTER_VERSION
344  );
345  $this->firstLetterData = $cache->getWithSetCallback( $cacheKey, $cache::TTL_WEEK, function () {
346  return $this->fetchFirstLetterData();
347  } );
348  }
349  return $this->firstLetterData;
350  }
351 
356  private function fetchFirstLetterData() {
357  // Generate data from serialized data file
358  if ( isset( self::TAILORING_FIRST_LETTERS[$this->locale] ) ) {
359  $letters = require __DIR__ . "/data/first-letters-root.php";
360  // Append additional characters
361  $letters = array_merge( $letters, self::TAILORING_FIRST_LETTERS[$this->locale] );
362  // Remove unnecessary ones, if any
363  if ( isset( self::TAILORING_FIRST_LETTERS['-' . $this->locale] ) ) {
364  $letters = array_diff( $letters, self::TAILORING_FIRST_LETTERS['-' . $this->locale] );
365  }
366  // Apply digit transforms
367  $digits = [ '0', '1', '2', '3', '4', '5', '6', '7', '8', '9' ];
368  $letters = array_diff( $letters, $digits );
369  foreach ( $digits as $digit ) {
370  $letters[] = $this->digitTransformLanguage->formatNumNoSeparators( $digit );
371  }
372  } elseif ( $this->locale === 'root' ) {
373  $letters = require __DIR__ . "/data/first-letters-root.php";
374  } else {
375  throw new MWException( "MediaWiki does not support ICU locale " .
376  "\"{$this->locale}\"" );
377  }
378 
379  /* Sort the letters.
380  *
381  * It's impossible to have the precompiled data file properly sorted,
382  * because the sort order changes depending on ICU version. If the
383  * array is not properly sorted, the binary search will return random
384  * results.
385  *
386  * We also take this opportunity to remove primary collisions.
387  */
388  $letterMap = [];
389  foreach ( $letters as $letter ) {
390  $key = $this->getPrimarySortKey( $letter );
391  if ( isset( $letterMap[$key] ) ) {
392  // Primary collision (two characters with the same sort position).
393  // Keep whichever one sorts first in the main collator.
394  $comp = $this->mainCollator->compare( $letter, $letterMap[$key] );
395  wfDebug( "Primary collision '$letter' '{$letterMap[$key]}' (comparison: $comp)" );
396  // If that also has a collision, use codepoint as a tiebreaker.
397  if ( $comp === 0 ) {
398  $comp = mb_ord( $letter ) <=> mb_ord( $letterMap[$key] );
399  }
400  if ( $comp < 0 ) {
401  $letterMap[$key] = $letter;
402  }
403  } else {
404  $letterMap[$key] = $letter;
405  }
406  }
407  ksort( $letterMap, SORT_STRING );
408 
409  /* Remove duplicate prefixes. Basically if something has a sortkey
410  * which is a prefix of some other sortkey, then it is an
411  * expansion and probably should not be considered a section
412  * header.
413  *
414  * For example 'þ' is sometimes sorted as if it is the letters
415  * 'th'. Other times it is its own primary element. Another
416  * example is '₨'. Sometimes its a currency symbol. Sometimes it
417  * is an 'R' followed by an 's'.
418  *
419  * Additionally an expanded element should always sort directly
420  * after its first element due to the way sortkeys work.
421  *
422  * UCA sortkey elements are of variable length but no collation
423  * element should be a prefix of some other element, so I think
424  * this is safe. See:
425  * - https://ssl.icu-project.org/repos/icu/icuhtml/trunk/design/collation/ICU_collation_design.htm
426  * - https://icu.unicode.org/design/collation/uca-weight-allocation
427  *
428  * Additionally, there is something called primary compression to
429  * worry about. Basically, if you have two primary elements that
430  * are more than one byte and both start with the same byte then
431  * the first byte is dropped on the second primary. Additionally
432  * either \x03 or \xFF may be added to mean that the next primary
433  * does not start with the first byte of the first primary.
434  *
435  * This shouldn't matter much, as the first primary is not
436  * changed, and that is what we are comparing against.
437  *
438  * tl;dr: This makes some assumptions about how icu implements
439  * collations. It seems incredibly unlikely these assumptions
440  * will change, but nonetheless they are assumptions.
441  */
442 
443  $prev = false;
444  $duplicatePrefixes = [];
445  foreach ( $letterMap as $key => $value ) {
446  // Remove terminator byte. Otherwise the prefix
447  // comparison will get hung up on that.
448  $trimmedKey = rtrim( $key, "\0" );
449  if ( $prev === false || $prev === '' ) {
450  $prev = $trimmedKey;
451  // We don't yet have a collation element
452  // to compare against, so continue.
453  continue;
454  }
455 
456  // Due to the fact the array is sorted, we only have
457  // to compare with the element directly previous
458  // to the current element (skipping expansions).
459  // An element "X" will always sort directly
460  // before "XZ" (Unless we have "XY", but we
461  // do not update $prev in that case).
462  if ( substr( $trimmedKey, 0, strlen( $prev ) ) === $prev ) {
463  $duplicatePrefixes[] = $key;
464  // If this is an expansion, we don't want to
465  // compare the next element to this element,
466  // but to what is currently $prev
467  continue;
468  }
469  $prev = $trimmedKey;
470  }
471  foreach ( $duplicatePrefixes as $badKey ) {
472  wfDebug( "Removing '{$letterMap[$badKey]}' from first letters." );
473  unset( $letterMap[$badKey] );
474  // This code assumes that unsetting does not change sort order.
475  }
476  $data = [
477  'chars' => array_values( $letterMap ),
478  'keys' => array_keys( $letterMap ),
479  ];
480 
481  // Reduce memory usage before caching
482  unset( $letterMap );
483 
484  return $data;
485  }
486 
493  public static function isCjk( $codepoint ) {
494  foreach ( self::CJK_BLOCKS as $block ) {
495  if ( $codepoint >= $block[0] && $codepoint <= $block[1] ) {
496  return true;
497  }
498  }
499  return false;
500  }
501 
509  public static function getUnicodeVersionForICU() {
510  $icuVersion = INTL_ICU_VERSION;
511  if ( !$icuVersion ) {
512  return false;
513  }
514 
515  $versionPrefix = substr( $icuVersion, 0, 3 );
516  // Source: https://icu.unicode.org/download
517  $map = [
518  '69.' => '13.0',
519  '68.' => '13.0',
520  '67.' => '13.0',
521  '66.' => '13.0',
522  '65.' => '12.0',
523  '64.' => '12.0',
524  '63.' => '11.0',
525  '62.' => '11.0',
526  '61.' => '10.0',
527  '60.' => '10.0',
528  '59.' => '9.0',
529  '58.' => '9.0',
530  '57.' => '8.0',
531  '56.' => '8.0',
532  '55.' => '7.0',
533  '54.' => '7.0',
534  '53.' => '6.3',
535  '52.' => '6.3',
536  '51.' => '6.2',
537  '50.' => '6.2',
538  '49.' => '6.1',
539  '4.8' => '6.0',
540  '4.6' => '6.0',
541  '4.4' => '5.2',
542  '4.2' => '5.1',
543  '4.0' => '5.1',
544  '3.8' => '5.0',
545  '3.6' => '5.0',
546  '3.4' => '4.1',
547  ];
548 
549  return $map[$versionPrefix] ?? false;
550  }
551 }
CACHE_ANYTHING
const CACHE_ANYTHING
Definition: Defines.php:85
MediaWiki\Languages\LanguageFactory
Internationalisation code See https://www.mediawiki.org/wiki/Special:MyLanguage/Localisation for more...
Definition: LanguageFactory.php:42
IcuCollation\$primaryCollator
Collator $primaryCollator
Definition: IcuCollation.php:30
IcuCollation\getPrimarySortKey
getPrimarySortKey( $string)
Definition: IcuCollation.php:326
wfMessage
wfMessage( $key,... $params)
This is the function for getting translated interface messages.
Definition: GlobalFunctions.php:1167
Collation
Definition: Collation.php:30
IcuCollation\CJK_BLOCKS
const CJK_BLOCKS
Unified CJK blocks.
Definition: IcuCollation.php:56
IcuCollation\__construct
__construct(LanguageFactory $languageFactory, $locale)
Definition: IcuCollation.php:248
IcuCollation\$mainCollator
Collator $mainCollator
Definition: IcuCollation.php:33
MWException
MediaWiki exception.
Definition: MWException.php:29
IcuCollation\$digitTransformLanguage
Language $digitTransformLanguage
Definition: IcuCollation.php:39
IcuCollation\isCjk
static isCjk( $codepoint)
Test if a code point is a CJK (Chinese, Japanese, Korean) character.
Definition: IcuCollation.php:493
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:894
MediaWiki\Languages\LanguageFactory\getLanguage
getLanguage( $code)
Get a cached or new language object for a given language code.
Definition: LanguageFactory.php:111
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:280
IcuCollation\getFirstLetterData
getFirstLetterData()
Definition: IcuCollation.php:334
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:509
$cache
$cache
Definition: mcc.php:33
IcuCollation\TAILORING_FIRST_LETTERS
const TAILORING_FIRST_LETTERS
Additional characters (or character groups) to be considered separate letters for given languages,...
Definition: IcuCollation.php:95
IcuCollation\$useNumericCollation
bool $useNumericCollation
Definition: IcuCollation.php:42
$keys
$keys
Definition: testCompression.php:72
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:276
IcuCollation\$firstLetterData
array $firstLetterData
Definition: IcuCollation.php:45
Language
Internationalisation code See https://www.mediawiki.org/wiki/Special:MyLanguage/Localisation for more...
Definition: Language.php:42
IcuCollation\fetchFirstLetterData
fetchFirstLetterData()
Definition: IcuCollation.php:356
ObjectCache\getLocalServerInstance
static getLocalServerInstance( $fallback=CACHE_NONE)
Factory function for CACHE_ACCEL (referenced from DefaultSettings.php)
Definition: ObjectCache.php:255
IcuCollation
Definition: IcuCollation.php:26