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 
244  public function __construct( $locale ) {
245  $this->locale = $locale;
246  // Drop everything after the '@' in locale's name
247  $localeParts = explode( '@', $locale );
248  $this->digitTransformLanguage = MediaWikiServices::getInstance()->getLanguageFactory()
249  ->getLanguage( $locale === 'root' ? 'en' : $localeParts[0] );
250 
251  $this->mainCollator = Collator::create( $locale );
252  if ( !$this->mainCollator ) {
253  throw new MWException( "Invalid ICU locale specified for collation: $locale" );
254  }
255 
256  $this->primaryCollator = Collator::create( $locale );
257  $this->primaryCollator->setStrength( Collator::PRIMARY );
258 
259  // If the special suffix for numeric collation is present, turn on numeric collation.
260  if ( substr( $locale, -5, 5 ) === '-u-kn' ) {
261  $this->useNumericCollation = true;
262  // Strip off the special suffix so it doesn't trip up fetchFirstLetterData().
263  $this->locale = substr( $this->locale, 0, -5 );
264  $this->mainCollator->setAttribute( Collator::NUMERIC_COLLATION, Collator::ON );
265  $this->primaryCollator->setAttribute( Collator::NUMERIC_COLLATION, Collator::ON );
266  }
267  }
268 
269  public function getSortKey( $string ) {
270  return $this->mainCollator->getSortKey( $string );
271  }
272 
273  public function getPrimarySortKey( $string ) {
274  return $this->primaryCollator->getSortKey( $string );
275  }
276 
277  public function getFirstLetter( $string ) {
278  $string = strval( $string );
279  if ( $string === '' ) {
280  return '';
281  }
282 
283  $firstChar = mb_substr( $string, 0, 1, 'UTF-8' );
284 
285  // If the first character is a CJK character, just return that character.
286  if ( ord( $firstChar ) > 0x7f && self::isCjk( UtfNormal\Utils::utf8ToCodepoint( $firstChar ) ) ) {
287  return $firstChar;
288  }
289 
290  $sortKey = $this->getPrimarySortKey( $string );
291 
292  // Do a binary search to find the correct letter to sort under
294  [ $this, 'getSortKeyByLetterIndex' ],
295  $this->getFirstLetterCount(),
296  'strcmp',
297  $sortKey );
298 
299  if ( $min === false ) {
300  // Before the first letter
301  return '';
302  }
303 
304  $sortLetter = $this->getLetterByIndex( $min );
305 
306  if ( $this->useNumericCollation ) {
307  // If the sort letter is a number, return '0–9' (or localized equivalent).
308  // ASCII value of 0 is 48. ASCII value of 9 is 57.
309  // Note that this also applies to non-Arabic numerals since they are
310  // mapped to Arabic numeral sort letters. For example, ২ sorts as 2.
311  if ( ord( $sortLetter ) >= 48 && ord( $sortLetter ) <= 57 ) {
312  $sortLetter = wfMessage( 'category-header-numerals' )->numParams( 0, 9 )->text();
313  }
314  }
315  return $sortLetter;
316  }
317 
322  public function getFirstLetterData() {
323  if ( $this->firstLetterData === null ) {
325  $cacheKey = $cache->makeKey(
326  'first-letters',
327  static::class,
328  $this->locale,
329  $this->digitTransformLanguage->getCode(),
330  INTL_ICU_VERSION,
331  self::FIRST_LETTER_VERSION
332  );
333  $this->firstLetterData = $cache->getWithSetCallback( $cacheKey, $cache::TTL_WEEK, function () {
334  return $this->fetchFirstLetterData();
335  } );
336  }
337  return $this->firstLetterData;
338  }
339 
344  private function fetchFirstLetterData() {
345  global $IP;
346  // Generate data from serialized data file
347  if ( isset( self::TAILORING_FIRST_LETTERS[$this->locale] ) ) {
348  $letters = require "$IP/includes/collation/data/first-letters-root.php";
349  // Append additional characters
350  $letters = array_merge( $letters, self::TAILORING_FIRST_LETTERS[$this->locale] );
351  // Remove unnecessary ones, if any
352  if ( isset( self::TAILORING_FIRST_LETTERS['-' . $this->locale] ) ) {
353  $letters = array_diff( $letters, self::TAILORING_FIRST_LETTERS['-' . $this->locale] );
354  }
355  // Apply digit transforms
356  $digits = [ '0', '1', '2', '3', '4', '5', '6', '7', '8', '9' ];
357  $letters = array_diff( $letters, $digits );
358  foreach ( $digits as $digit ) {
359  $letters[] = $this->digitTransformLanguage->formatNumNoSeparators( $digit );
360  }
361  } elseif ( $this->locale === 'root' ) {
362  $letters = require "$IP/includes/collation/data/first-letters-root.php";
363  } else {
364  // FIXME: Is this still used?
365  $letters = $this->getPrecompiledData( "first-letters-{$this->locale}.ser" );
366  if ( $letters === false ) {
367  throw new MWException( "MediaWiki does not support ICU locale " .
368  "\"{$this->locale}\"" );
369  }
370  }
371 
372  /* Sort the letters.
373  *
374  * It's impossible to have the precompiled data file properly sorted,
375  * because the sort order changes depending on ICU version. If the
376  * array is not properly sorted, the binary search will return random
377  * results.
378  *
379  * We also take this opportunity to remove primary collisions.
380  */
381  $letterMap = [];
382  foreach ( $letters as $letter ) {
383  $key = $this->getPrimarySortKey( $letter );
384  if ( isset( $letterMap[$key] ) ) {
385  // Primary collision (two characters with the same sort position).
386  // Keep whichever one sorts first in the main collator.
387  $comp = $this->mainCollator->compare( $letter, $letterMap[$key] );
388  wfDebug( "Primary collision '$letter' '{$letterMap[$key]}' (comparison: $comp)" );
389  // If that also has a collision, use codepoint as a tiebreaker.
390  if ( $comp === 0 ) {
391  $comp = UtfNormal\Utils::utf8ToCodepoint( $letter ) <=>
392  UtfNormal\Utils::utf8ToCodepoint( $letterMap[$key] );
393  }
394  if ( $comp < 0 ) {
395  $letterMap[$key] = $letter;
396  }
397  } else {
398  $letterMap[$key] = $letter;
399  }
400  }
401  ksort( $letterMap, SORT_STRING );
402 
403  /* Remove duplicate prefixes. Basically if something has a sortkey
404  * which is a prefix of some other sortkey, then it is an
405  * expansion and probably should not be considered a section
406  * header.
407  *
408  * For example 'þ' is sometimes sorted as if it is the letters
409  * 'th'. Other times it is its own primary element. Another
410  * example is '₨'. Sometimes its a currency symbol. Sometimes it
411  * is an 'R' followed by an 's'.
412  *
413  * Additionally an expanded element should always sort directly
414  * after its first element due to they way sortkeys work.
415  *
416  * UCA sortkey elements are of variable length but no collation
417  * element should be a prefix of some other element, so I think
418  * this is safe. See:
419  * - https://ssl.icu-project.org/repos/icu/icuhtml/trunk/design/collation/ICU_collation_design.htm
420  * - http://site.icu-project.org/design/collation/uca-weight-allocation
421  *
422  * Additionally, there is something called primary compression to
423  * worry about. Basically, if you have two primary elements that
424  * are more than one byte and both start with the same byte then
425  * the first byte is dropped on the second primary. Additionally
426  * either \x03 or \xFF may be added to mean that the next primary
427  * does not start with the first byte of the first primary.
428  *
429  * This shouldn't matter much, as the first primary is not
430  * changed, and that is what we are comparing against.
431  *
432  * tl;dr: This makes some assumptions about how icu implements
433  * collations. It seems incredibly unlikely these assumptions
434  * will change, but nonetheless they are assumptions.
435  */
436 
437  $prev = false;
438  $duplicatePrefixes = [];
439  foreach ( $letterMap as $key => $value ) {
440  // Remove terminator byte. Otherwise the prefix
441  // comparison will get hung up on that.
442  $trimmedKey = rtrim( $key, "\0" );
443  if ( $prev === false || $prev === '' ) {
444  $prev = $trimmedKey;
445  // We don't yet have a collation element
446  // to compare against, so continue.
447  continue;
448  }
449 
450  // Due to the fact the array is sorted, we only have
451  // to compare with the element directly previous
452  // to the current element (skipping expansions).
453  // An element "X" will always sort directly
454  // before "XZ" (Unless we have "XY", but we
455  // do not update $prev in that case).
456  if ( substr( $trimmedKey, 0, strlen( $prev ) ) === $prev ) {
457  $duplicatePrefixes[] = $key;
458  // If this is an expansion, we don't want to
459  // compare the next element to this element,
460  // but to what is currently $prev
461  continue;
462  }
463  $prev = $trimmedKey;
464  }
465  foreach ( $duplicatePrefixes as $badKey ) {
466  wfDebug( "Removing '{$letterMap[$badKey]}' from first letters." );
467  unset( $letterMap[$badKey] );
468  // This code assumes that unsetting does not change sort order.
469  }
470  $data = [
471  'chars' => array_values( $letterMap ),
472  'keys' => array_keys( $letterMap ),
473  ];
474 
475  // Reduce memory usage before caching
476  unset( $letterMap );
477 
478  return $data;
479  }
480 
489  private function getPrecompiledData( $name ) {
490  global $IP;
491  $file = "$IP/serialized/$name";
492  if ( file_exists( $file ) ) {
493  $blob = file_get_contents( $file );
494  if ( $blob ) {
495  return unserialize( $blob );
496  }
497  }
498  return false;
499  }
500 
506  public function getLetterByIndex( $index ) {
507  return $this->getFirstLetterData()['chars'][$index];
508  }
509 
515  public function getSortKeyByLetterIndex( $index ) {
516  return $this->getFirstLetterData()['keys'][$index];
517  }
518 
523  public function getFirstLetterCount() {
524  return count( $this->getFirstLetterData()['chars'] );
525  }
526 
533  public static function isCjk( $codepoint ) {
534  foreach ( self::CJK_BLOCKS as $block ) {
535  if ( $codepoint >= $block[0] && $codepoint <= $block[1] ) {
536  return true;
537  }
538  }
539  return false;
540  }
541 
549  public static function getUnicodeVersionForICU() {
550  $icuVersion = INTL_ICU_VERSION;
551  if ( !$icuVersion ) {
552  return false;
553  }
554 
555  $versionPrefix = substr( $icuVersion, 0, 3 );
556  // Source: http://site.icu-project.org/download
557  $map = [
558  '67.' => '13.0',
559  '66.' => '13.0',
560  '65.' => '12.0',
561  '64.' => '12.0',
562  '63.' => '11.0',
563  '62.' => '11.0',
564  '61.' => '10.0',
565  '60.' => '10.0',
566  '59.' => '9.0',
567  '58.' => '9.0',
568  '57.' => '8.0',
569  '56.' => '8.0',
570  '55.' => '7.0',
571  '54.' => '7.0',
572  '53.' => '6.3',
573  '52.' => '6.3',
574  '51.' => '6.2',
575  '50.' => '6.2',
576  '49.' => '6.1',
577  '4.8' => '6.0',
578  '4.6' => '6.0',
579  '4.4' => '5.2',
580  '4.2' => '5.1',
581  '4.0' => '5.1',
582  '3.8' => '5.0',
583  '3.6' => '5.0',
584  '3.4' => '4.1',
585  ];
586 
587  return $map[$versionPrefix] ?? false;
588  }
589 }
CACHE_ANYTHING
const CACHE_ANYTHING
Definition: Defines.php:84
MediaWiki\MediaWikiServices
MediaWikiServices is the service locator for the application scope of MediaWiki.
Definition: MediaWikiServices.php:166
IcuCollation\getFirstLetterCount
getFirstLetterCount()
Definition: IcuCollation.php:523
IcuCollation\$primaryCollator
Collator $primaryCollator
Definition: IcuCollation.php:30
IcuCollation\getPrimarySortKey
getPrimarySortKey( $string)
Definition: IcuCollation.php:273
$file
if(PHP_SAPI !='cli-server') if(!isset( $_SERVER['SCRIPT_FILENAME'])) $file
Item class for a filearchive table row.
Definition: router.php:42
wfMessage
wfMessage( $key,... $params)
This is the function for getting translated interface messages.
Definition: GlobalFunctions.php:1230
Collation
Definition: Collation.php:30
IcuCollation\CJK_BLOCKS
const CJK_BLOCKS
Unified CJK blocks.
Definition: IcuCollation.php:56
IcuCollation\getPrecompiledData
getPrecompiledData( $name)
Get an object from the precompiled serialized directory.
Definition: IcuCollation.php:489
IcuCollation\$mainCollator
Collator $mainCollator
Definition: IcuCollation.php:33
MWException
MediaWiki exception.
Definition: MWException.php:29
IcuCollation\$digitTransformLanguage
Language $digitTransformLanguage
Definition: IcuCollation.php:39
$blob
$blob
Definition: testCompression.php:70
IcuCollation\getSortKeyByLetterIndex
getSortKeyByLetterIndex( $index)
Definition: IcuCollation.php:515
IcuCollation\isCjk
static isCjk( $codepoint)
Test if a code point is a CJK (Chinese, Japanese, Korean) character.
Definition: IcuCollation.php:533
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:914
IcuCollation\getLetterByIndex
getLetterByIndex( $index)
Definition: IcuCollation.php:506
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:277
IcuCollation\getFirstLetterData
getFirstLetterData()
Definition: IcuCollation.php:322
IcuCollation\FIRST_LETTER_VERSION
const FIRST_LETTER_VERSION
Definition: IcuCollation.php:27
IcuCollation\$locale
string $locale
Definition: IcuCollation.php:36
unserialize
unserialize( $serialized)
Definition: ApiMessageTrait.php:146
IcuCollation\getUnicodeVersionForICU
static getUnicodeVersionForICU()
Return the version of Unicode appropriate for the version of ICU library currently in use,...
Definition: IcuCollation.php:549
$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
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:269
IcuCollation\$firstLetterData
array $firstLetterData
Definition: IcuCollation.php:45
IcuCollation\__construct
__construct( $locale)
Definition: IcuCollation.php:244
$IP
$IP
Definition: WebStart.php:49
Language
Internationalisation code See https://www.mediawiki.org/wiki/Special:MyLanguage/Localisation for more...
Definition: Language.php:42
IcuCollation\fetchFirstLetterData
fetchFirstLetterData()
Definition: IcuCollation.php:344
ObjectCache\getLocalServerInstance
static getLocalServerInstance( $fallback=CACHE_NONE)
Factory function for CACHE_ACCEL (referenced from DefaultSettings.php)
Definition: ObjectCache.php:258
IcuCollation
Definition: IcuCollation.php:26