| blob | e1a20473ee51c482581f587cd540e5f1b594c853 |
1 <?php
2 /**
3 * Internationalisation code.
4 *
5 * This program is free software; you can redistribute it and/or modify
6 * it under the terms of the GNU General Public License as published by
7 * the Free Software Foundation; either version 2 of the License, or
8 * (at your option) any later version.
9 *
10 * This program is distributed in the hope that it will be useful,
11 * but WITHOUT ANY WARRANTY; without even the implied warranty of
12 * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
13 * GNU General Public License for more details.
14 *
15 * You should have received a copy of the GNU General Public License along
16 * with this program; if not, write to the Free Software Foundation, Inc.,
17 * 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA.
18 * http://www.gnu.org/copyleft/gpl.html
19 *
20 * @file
21 * @ingroup Language
22 */
24 /**
25 * @defgroup Language Language
26 */
31 }
35 }
37 /**
38 * Internationalisation code
39 * @ingroup Language
40 */
42 /**
43 * @var LanguageConverter
44 */
56 /**
57 * ReplacementArray object caches
58 */
61 /**
62 * @var LocalisationCache
63 */
71 );
75 );
80 'december'
81 );
85 'december-gen'
86 );
90 );
97 );
105 );
113 );
120 );
122 /**
123 * @since 1.20
124 * @var array
125 */
136 );
138 /**
139 * Cache for language fallbacks.
140 * @see Language::getFallbacksIncludingSiteLanguage
141 * @since 1.21
142 * @var array
143 */
146 /**
147 * Cache for language names
148 * @var MapCacheLRU|null
149 */
152 /**
153 * Get a cached or new language object for a given language code
154 * @param string $code
155 * @return Language
156 */
162 }
164 // get the language object to process
169 // merge the language object in to get it up front in the cache
171 // get rid of the oldest ones in case we have an overflow
175 }
177 /**
178 * Create a language object for a given language code
179 * @param string $code
180 * @throws MWException
181 * @return Language
182 */
184 // Protect against path traversal below
187 ) {
189 }
192 // It's not possible to customise this code with class files, so
193 // just return a Language object. This is to support uselang= hacks.
197 }
199 // Check if there is a language class for the code
205 }
207 // Keep trying the fallback list until we find an existing class
212 }
220 }
221 }
224 }
226 /**
227 * Checks whether any localisation is available for that language tag
228 * in MediaWiki (MessagesXx.php exists).
229 *
230 * @param string $code Language tag (in lower case)
231 * @return bool Whether language is supported
232 * @since 1.21
233 */
238 );
239 }
241 /**
242 * Returns true if a language code string is a well-formed language tag
243 * according to RFC 5646.
244 * This function only checks well-formedness; it doesn't check that
245 * language, script or variant codes actually exist in the repositories.
246 *
247 * Based on regexes by Mark Davis of the Unicode Consortium:
248 * http://unicode.org/repos/cldr/trunk/tools/java/org/unicode/cldr/util/data/langtagRegex.txt
249 *
250 * @param string $code
251 * @param bool $lenient Whether to allow '_' as separator. The default is only '-'.
252 *
253 * @return bool
254 * @since 1.21
255 */
271 # Define certain grandfathered codes, since otherwise the regex is pretty useless.
272 # Since these are limited, this is safe even later changes to the registry --
273 # the only oddity is that it might change the type of the tag, and thus
274 # the results from the capturing groups.
275 # http://www.iana.org/assignments/language-subtag-registry
293 # The final breakdown, with capturing groups for each of these components
294 # The variants, extensions, grandfathered, and private-use may have interior '-'
299 }
301 /**
302 * Returns true if a language code string is of a valid form, whether or
303 * not it exists. This includes codes which are used solely for
304 * customisation via the MediaWiki namespace.
305 *
306 * @param string $code
307 *
308 * @return bool
309 */
314 }
315 // People think language codes are html safe, so enforce it.
316 // Ideally we should only allow a-zA-Z0-9-
317 // but, .+ and other chars are often used for {{int:}} hacks
318 // see bugs 37564, 37587, 36938
324 }
326 /**
327 * Returns true if a language code is of a valid form for the purposes of
328 * internal customisation of MediaWiki, via Messages*.php or *.json.
329 *
330 * @param string $code
331 *
332 * @throws MWException
333 * @since 1.18
334 * @return bool
335 */
343 }
346 }
349 }
351 /**
352 * Returns true if a language code is an IETF tag known to MediaWiki.
353 *
354 * @param string $tag
355 *
356 * @since 1.21
357 * @return bool
358 */
362 // Quick escape for invalid input to avoid exceptions down the line
363 // when code tries to process tags which are not valid at all.
366 }
371 }
375 ) {
377 }
380 }
382 /**
383 * @param string $code
384 * @return string Name of the language class
385 */
391 }
392 }
394 /**
395 * Includes language class files
396 *
397 * @param string $class Name of the language class
398 */
404 }
408 }
409 }
411 /**
412 * Get the LocalisationCache instance
413 *
414 * @return LocalisationCache
415 */
421 }
423 }
427 // Set the code to the name of the descendant
432 }
434 }
436 /**
437 * Reduce memory usage
438 */
442 }
443 }
445 /**
446 * Hook which will be called if this is the content language.
447 * Descendants can use this to register hook functions or modify globals
448 */
450 }
452 /**
453 * @return array
454 * @since 1.19
455 */
458 }
460 /**
461 * Exports $wgBookstoreListEn
462 * @return array
463 */
466 }
468 /**
469 * Returns an array of localised namespaces indexed by their numbers. If the namespace is not
470 * available in localised form, it will be included in English.
471 *
472 * @return array
473 */
490 }
492 # Sometimes a language will be localised but not actually exist on this wiki.
496 }
497 }
499 # The above mixing may leave namespaces out of canonical order.
500 # Re-order by namespace ID number...
504 }
507 }
509 /**
510 * Arbitrarily set all of the namespace names at once. Mainly used for testing
511 * @param array $namespaces Array of namespaces (id => name)
512 */
516 }
518 /**
519 * Resets all of the namespace caches. Mainly used for testing
520 */
525 }
527 /**
528 * A convenience function that returns the same thing as
529 * getNamespaces() except with the array values changed to ' '
530 * where it found '_', useful for producing output to be displayed
531 * e.g. in <select> forms.
532 *
533 * @return array
534 */
539 }
541 }
543 /**
544 * Get a namespace value by key
545 * <code>
546 * $mw_ns = $wgContLang->getNsText( NS_MEDIAWIKI );
547 * echo $mw_ns; // prints 'MediaWiki'
548 * </code>
549 *
550 * @param int $index The array key of the namespace to return
551 * @return string|bool String if the namespace value exists, otherwise false
552 */
557 }
559 /**
560 * A convenience function that returns the same thing as
561 * getNsText() except with '_' changed to ' ', useful for
562 * producing output.
563 *
564 * <code>
565 * $mw_ns = $wgContLang->getFormattedNsText( NS_MEDIAWIKI_TALK );
566 * echo $mw_ns; // prints 'MediaWiki talk'
567 * </code>
568 *
569 * @param int $index The array key of the namespace to return
570 * @return string Namespace name without underscores (empty string if namespace does not exist)
571 */
576 }
578 /**
579 * Returns gender-dependent namespace alias if available.
580 * See https://www.mediawiki.org/wiki/Manual:$wgExtraGenderNamespaces
581 * @param int $index Namespace index
582 * @param string $gender Gender key (male, female... )
583 * @return string
584 * @since 1.18
585 */
593 }
595 /**
596 * Whether this language uses gender-dependent namespace aliases.
597 * See https://www.mediawiki.org/wiki/Manual:$wgExtraGenderNamespaces
598 * @return bool
599 * @since 1.18
600 */
604 // $wgExtraGenderNamespaces overrides everything
606 } elseif ( isset( $wgExtraNamespaces[NS_USER] ) && isset( $wgExtraNamespaces[NS_USER_TALK] ) ) {
607 /// @todo There may be other gender namespace than NS_USER & NS_USER_TALK in the future
608 // $wgExtraNamespaces overrides any gender aliases specified in i18n files
611 // Check what is in i18n files
614 }
615 }
617 /**
618 * Get a namespace key by value, case insensitive.
619 * Only matches namespace names for the current language, not the
620 * canonical ones defined in Namespace.php.
621 *
622 * @param string $text
623 * @return int|bool An integer if $text is a valid value otherwise false
624 */
629 }
631 /**
632 * @return array
633 */
645 }
646 }
647 }
655 }
656 }
658 # Also add converted namespace names as aliases, to avoid confusion.
663 }
666 }
667 }
670 }
673 }
675 /**
676 * @return array
677 */
681 # Put namespace names and aliases into a hashtable.
682 # If this is too slow, then we should arrange it so that it is done
683 # before caching. The catch is that at pre-cache time, the above
684 # class-specific fixup hasn't been done.
688 }
691 }
695 }
696 }
697 }
699 }
701 /**
702 * Get a namespace key by value, case insensitive. Canonical namespace
703 * names override custom ones defined for the current language.
704 *
705 * @param string $text
706 * @return int|bool An integer if $text is a valid value otherwise false
707 */
713 }
716 }
718 /**
719 * short names for language variants used for language conversion links.
720 *
721 * @param string $code
722 * @param bool $usemsg Use the "variantname-xyz" message if it exists
723 * @return string
724 */
729 }
734 # otherwise, output the language code
736 }
737 }
739 /**
740 * @deprecated since 1.24, doesn't handle conflicting aliases. Use
741 * SpecialPageFactory::getLocalNameFor instead.
742 * @param string $name
743 * @return string
744 */
749 }
751 }
753 /**
754 * @return array
755 */
758 }
760 /**
761 * @return array
762 */
765 }
767 /**
768 * @return array|string
769 */
777 }
778 }
780 /**
781 * @return array
782 */
785 }
787 /**
788 * @param string $image
789 * @return array|null
790 */
793 }
795 /**
796 * @return array
797 * @since 1.24
798 */
801 }
803 /**
804 * @return array
805 */
808 }
810 /**
811 * @param string $tog
812 * @return string
813 */
816 }
818 /**
819 * Get native language names, indexed by code.
820 * Only those defined in MediaWiki, no other data like CLDR.
821 * If $customisedOnly is true, only returns codes with a messages file
822 *
823 * @param bool $customisedOnly
824 *
825 * @return array
826 * @deprecated since 1.20, use fetchLanguageNames()
827 */
830 }
832 /**
833 * Get translated language names. This is done on best effort and
834 * by default this is exactly the same as Language::getLanguageNames.
835 * The CLDR extension provides translated names.
836 * @param string $code Language code.
837 * @return array Language code => language name
838 * @since 1.18.0
839 * @deprecated since 1.20, use fetchLanguageNames()
840 */
843 }
845 /**
846 * Get an array of language names, indexed by code.
847 * @param null|string $inLanguage Code of language in which to return the names
848 * Use null for autonyms (native names)
849 * @param string $include One of:
850 * 'all' all available languages
851 * 'mw' only if the language is defined in MediaWiki or wgExtraLanguageNames (default)
852 * 'mwfile' only if the language is in 'mw' *and* has a message file
853 * @return array Language code => language name
854 * @since 1.20
855 */
861 }
867 }
869 }
871 /**
872 * Uncached helper for fetchLanguageNames
873 * @param null|string $inLanguage Code of language in which to return the names
874 * Use null for autonyms (native names)
875 * @param string $include One of:
876 * 'all' all available languages
877 * 'mw' only if the language is defined in MediaWiki or wgExtraLanguageNames (default)
878 * 'mwfile' only if the language is in 'mw' *and* has a message file
879 * @return array Language code => language name
880 */
888 }
890 // If passed an invalid language code to use, fallback to en
893 }
898 # TODO: also include when $inLanguage is null, when this code is more efficient
900 }
904 # - Prefer own MediaWiki native name when not using the hook
905 # - For other names just add if not added through the hook
908 }
909 }
914 }
920 }
924 # We do this using a foreach over the codes instead of a directory
925 # loop so that messages files in extensions will work correctly.
929 ) {
931 }
932 }
936 }
939 # 'mw' option; default if it's not one of the other two options (all/mwfile)
941 }
943 /**
944 * @param string $code The code of the language for which to get the name
945 * @param null|string $inLanguage Code of language in which to return the name (null for autonyms)
946 * @param string $include 'all', 'mw' or 'mwfile'; see fetchLanguageNames()
947 * @return string Language name or empty
948 * @since 1.20
949 */
954 }
956 /**
957 * Get a message from the MediaWiki namespace.
958 *
959 * @param string $msg Message name
960 * @return string
961 */
964 }
966 /**
967 * Get message object in this language. Only for use inside this class.
968 *
969 * @param string $msg Message name
970 * @return Message
971 */
974 }
976 /**
977 * Get the native language name of $code.
978 * Only if defined in MediaWiki, no other data like CLDR.
979 * @param string $code
980 * @return string
981 * @deprecated since 1.20, use fetchLanguageName()
982 */
985 }
987 /**
988 * @param string $key
989 * @return string
990 */
993 }
995 /**
996 * @return array
997 */
1002 }
1004 }
1006 /**
1007 * @param string $key
1008 * @return string
1009 */
1012 }
1014 /**
1015 * @param string $key
1016 * @return string
1017 */
1020 }
1022 /**
1023 * @return array
1024 */
1029 }
1031 }
1033 /**
1034 * @param string $key
1035 * @return string
1036 */
1039 }
1041 /**
1042 * @param string $key
1043 * @return string
1044 */
1047 }
1049 /**
1050 * @param string $key
1051 * @return string
1052 */
1055 }
1057 /**
1058 * @param string $key
1059 * @return string
1060 */
1063 }
1065 /**
1066 * @param string $key
1067 * @return string
1068 */
1071 }
1073 /**
1074 * @param string $key
1075 * @return string
1076 */
1079 }
1081 /**
1082 * Pass through result from $dateTimeObj->format()
1083 * @param DateTime|bool|null &$dateTimeObj
1084 * @param string $ts
1085 * @param DateTimeZone|bool|null $zone
1086 * @param string $code
1087 * @return string
1088 */
1093 );
1094 }
1096 }
1098 /**
1099 * This is a workalike of PHP's date() function, but with better
1100 * internationalisation, a reduced set of format characters, and a better
1101 * escaping format.
1102 *
1103 * Supported format characters are dDjlNwzWFmMntLoYyaAgGhHiscrUeIOPTZ. See
1104 * the PHP manual for definitions. There are a number of extensions, which
1105 * start with "x":
1106 *
1107 * xn Do not translate digits of the next numeric format character
1108 * xN Toggle raw digit (xn) flag, stays set until explicitly unset
1109 * xr Use roman numerals for the next numeric format character
1110 * xh Use hebrew numerals for the next numeric format character
1111 * xx Literal x
1112 * xg Genitive month name
1113 *
1114 * xij j (day number) in Iranian calendar
1115 * xiF F (month name) in Iranian calendar
1116 * xin n (month number) in Iranian calendar
1117 * xiy y (two digit year) in Iranian calendar
1118 * xiY Y (full year) in Iranian calendar
1119 *
1120 * xjj j (day number) in Hebrew calendar
1121 * xjF F (month name) in Hebrew calendar
1122 * xjt t (days in month) in Hebrew calendar
1123 * xjx xg (genitive month name) in Hebrew calendar
1124 * xjn n (month number) in Hebrew calendar
1125 * xjY Y (full year) in Hebrew calendar
1126 *
1127 * xmj j (day number) in Hijri calendar
1128 * xmF F (month name) in Hijri calendar
1129 * xmn n (month number) in Hijri calendar
1130 * xmY Y (full year) in Hijri calendar
1131 *
1132 * xkY Y (full year) in Thai solar calendar. Months and days are
1133 * identical to the Gregorian calendar
1134 * xoY Y (full year) in Minguo calendar or Juche year.
1135 * Months and days are identical to the
1136 * Gregorian calendar
1137 * xtY Y (full year) in Japanese nengo. Months and days are
1138 * identical to the Gregorian calendar
1139 *
1140 * Characters enclosed in double quotes will be considered literal (with
1141 * the quotes themselves removed). Unmatched quotes will be considered
1142 * literal quotes. Example:
1143 *
1144 * "The month is" F => The month is January
1145 * i's" => 20'11"
1146 *
1147 * Backslash escaping is also supported.
1148 *
1149 * Input timestamp is assumed to be pre-normalized to the desired local
1150 * time zone, if any. Note that the format characters crUeIOPTZ will assume
1151 * $ts is UTC if $zone is not given.
1152 *
1153 * @param string $format
1154 * @param string $ts 14-character timestamp
1155 * YYYYMMDDHHMMSS
1156 * 01234567890123
1157 * @param DateTimeZone $zone Timezone of $ts
1158 * @param[out] int $ttl The amount of time (in seconds) the output may be cached for.
1159 * Only makes sense if $ts is the current time.
1160 * @todo handling of "o" format character for Iranian, Hebrew, Hijri & Thai?
1161 *
1162 * @throws MWException
1163 * @return string
1164 */
1200 }
1204 }
1212 }
1222 }
1248 }
1259 );
1269 }
1276 }
1283 }
1290 );
1300 }
1307 }
1314 }
1333 }
1340 }
1347 }
1354 }
1365 }
1372 }
1379 }
1386 }
1393 }
1400 }
1411 }
1451 // fall through
1482 // fall through
1488 # Backslash escaping
1493 }
1496 # Quoted literal
1500 # No terminating quote, assume literal "
1505 }
1507 # Quote at end of string, assume literal "
1509 }
1513 }
1526 }
1527 }
1528 }
1540 $usedDay ||
1541 $usedHebrewMonth ||
1542 $usedIranianMonth ||
1543 $usedHijriMonth ||
1544 $usedHebrewYear ||
1545 $usedIranianYear ||
1546 $usedHijriYear ||
1547 $usedTennoYear
1548 ) {
1549 // @todo Someone who understands the non-Gregorian calendars
1550 // should write proper logic for them so that they don't need purged every day.
1562 // December 28th falls on the last ISO week of the year, every year.
1563 // The last ISO week of a year can be 52 or 53.
1575 }
1595 // this isn't a leap year. see when the next one starts
1603 }
1605 // this is a leap year, so the next year isn't
1607 }
1608 }
1612 }
1613 }
1616 }
1621 /**
1622 * Algorithm by Roozbeh Pournader and Mohammad Toossi to convert
1623 * Gregorian dates to Iranian dates. Originally written in C, it
1624 * is released under the terms of GNU Lesser General Public
1625 * License. Conversion to PHP was performed by Niklas Laxström.
1626 *
1627 * Link: http://www.farsiweb.info/jalali/jalali.c
1628 *
1629 * @param string $ts
1630 *
1631 * @return string
1632 */
1638 # Days passed from the beginning (including leap years)
1644 // Add days of the past months of this year
1647 }
1649 // Leap years
1652 }
1654 // Days passed in current month
1668 }
1672 }
1678 }
1680 /**
1681 * Converting Gregorian dates to Hijri dates.
1682 *
1683 * Based on a PHP-Nuke block by Sharjeel which is released under GNU/GPL license
1684 *
1685 * @see http://phpnuke.org/modules.php?name=News&file=article&sid=8234&mode=thread&order=0&thold=0
1686 *
1687 * @param string $ts
1688 *
1689 * @return string
1690 */
1704 ) {
1712 }
1726 }
1728 /**
1729 * Converting Gregorian dates to Hebrew dates.
1730 *
1731 * Based on a JavaScript code by Abu Mami and Yisrael Hersch
1732 * (abu-mami@kaluach.net, http://www.kaluach.net), who permitted
1733 * to translate the relevant functions into PHP and release them under
1734 * GNU GPL.
1735 *
1736 * The months are counted from Tishrei = 1. In a leap year, Adar I is 13
1737 * and Adar II is 14. In a non-leap year, Adar is 6.
1738 *
1739 * @param string $ts
1740 *
1741 * @return string
1742 */
1744 # Parse date
1749 # Calculate Hebrew year
1752 # Month number when September = 1, August = 12
1755 # Next year
1759 }
1761 # Calculate day of year from 1 September
1765 # February
1767 # Check if the year is leap
1770 }
1775 }
1776 }
1778 # Calculate the start of the Hebrew year
1781 # Calculate next year's start
1783 # Day is before the start of the year - it is the previous year
1784 # Next year's start
1786 # Previous year
1789 # Add days since previous year's 1 September
1792 # Leap year
1794 }
1795 # Start of the new (previous) year
1798 # Next year's start
1800 }
1802 # Calculate Hebrew day of year
1805 # Difference between year's days
1807 # Add 12 (or 13 for leap years) days to ignore the difference between
1808 # Hebrew and Gregorian year (353 at least vs. 365/6) - now the
1809 # difference is only about the year type
1814 }
1816 # Check the year pattern, and is leap year
1817 # 0 means an incomplete year, 1 means a regular year, 2 means a complete year
1818 # This is mod 30, to work on both leap years (which add 30 days of Adar I)
1819 # and non-leap years
1821 # Check if leap year
1824 # Calculate day in the month from number of day in the Hebrew year
1825 # Don't check Adar - if the day is not in Adar, we will stop before;
1826 # if it is in Adar, we will use it to check if it is Adar I or Adar II
1831 # Calculate days in this month
1833 # Adar in a leap year
1835 # Leap year - has Adar I, with 30 days, and Adar II, with 29 days
1838 # Day in Adar I
1841 # Subtract the days of Adar I
1843 # Try Adar II
1846 # Day in Adar II
1848 }
1849 }
1850 }
1852 # Cheshvan in a complete year (otherwise as the rule below)
1855 # Kislev in an incomplete year (otherwise as the rule below)
1858 # Odd months have 30 days, even have 29
1860 }
1862 # In the current month
1865 # Subtract the days of the current month
1867 # Try in the next month
1869 }
1870 }
1873 }
1875 /**
1876 * This calculates the Hebrew year start, as days since 1 September.
1877 * Based on Carl Friedrich Gauss algorithm for finding Easter date.
1878 * Used for Hebrew date.
1879 *
1880 * @param int $year
1881 *
1882 * @return string
1883 */
1890 }
1894 }
1904 }
1908 }
1910 /**
1911 * Algorithm to convert Gregorian dates to Thai solar dates,
1912 * Minguo dates or Minguo dates.
1913 *
1914 * Link: http://en.wikipedia.org/wiki/Thai_solar_calendar
1915 * http://en.wikipedia.org/wiki/Minguo_calendar
1916 * http://en.wikipedia.org/wiki/Japanese_era_name
1917 *
1918 * @param string $ts 14-character timestamp
1919 * @param string $cName Calender name
1920 * @return array Converted year, month, day
1921 */
1928 # Thai solar dates
1929 # Add 543 years to the Gregorian calendar
1930 # Months and days are identical
1933 # Minguo dates
1934 # Deduct 1911 years from the Gregorian calendar
1935 # Months and days are identical
1938 # Nengō dates up to Meiji period
1939 # Deduct years from the Gregorian calendar
1940 # depending on the nengo periods
1941 # Months and days are identical
1945 ) {
1946 # Meiji period
1951 }
1959 ) {
1960 # Taishō period
1965 }
1971 ) {
1972 # Shōwa period
1977 }
1980 # Heisei period
1985 }
1987 }
1990 }
1993 }
1995 /**
1996 * Roman number formatting up to 10000
1997 *
1998 * @param int $num
1999 *
2000 * @return string
2001 */
2009 );
2014 }
2020 }
2022 }
2024 }
2026 /**
2027 * Hebrew Gematria number formatting up to 9999
2028 *
2029 * @param int $num
2030 *
2031 * @return string
2032 */
2039 );
2044 }
2056 }
2057 }
2058 }
2060 }
2066 }
2085 }
2087 }
2089 /**
2090 * Used by date() and time() to adjust the time output.
2091 *
2092 * @param string $ts The time in date('YmdHis') format
2093 * @param mixed $tz Adjust the time by this amount (default false, mean we
2094 * get user timecorrection setting)
2095 * @return int
2096 */
2102 }
2115 }
2116 # Unrecognized timezone, default to 'Offset' with the stored offset.
2118 }
2121 # Global offset in minutes.
2133 }
2136 }
2137 }
2139 # No difference ? Return time unchanged
2142 }
2145 # Generate an adjusted date; take advantage of the fact that mktime
2146 # will normalize out-of-range values so we don't have to split $minDiff
2147 # into hours and minutes.
2160 }
2162 /**
2163 * This is meant to be used by time(), date(), and timeanddate() to get
2164 * the date preference they're supposed to use, it should be used in
2165 * all children.
2166 *
2167 *<code>
2168 * function timeanddate([...], $format = true) {
2169 * $datePreference = $this->dateFormat($format);
2170 * [...]
2171 * }
2172 *</code>
2173 *
2174 * @param int|string|bool $usePrefs If true, the user's preference is used
2175 * if false, the site/language default is used
2176 * if int/string, assumed to be a format.
2177 * @return string
2178 */
2187 }
2190 }
2192 // return int
2195 }
2198 }
2200 /**
2201 * Get a format string for a given type and preference
2202 * @param string $type May be date, time or both
2203 * @param string $pref The format name as it appears in Messages*.php
2204 *
2205 * @since 1.22 New type 'pretty' that provides a more readable timestamp format
2206 *
2207 * @return string
2208 */
2219 }
2224 }
2225 }
2227 }
2229 }
2231 /**
2232 * @param string $ts The time format which needs to be turned into a
2233 * date('YmdHis') format with wfTimestamp(TS_MW,$ts)
2234 * @param bool $adj Whether to adjust the time output according to the
2235 * user configured offset ($timecorrection)
2236 * @param mixed $format True to use user's date format preference
2237 * @param string|bool $timecorrection The time offset as returned by
2238 * validateTimeZone() in Special:Preferences
2239 * @return string
2240 */
2245 }
2248 }
2250 /**
2251 * @param string $ts The time format which needs to be turned into a
2252 * date('YmdHis') format with wfTimestamp(TS_MW,$ts)
2253 * @param bool $adj Whether to adjust the time output according to the
2254 * user configured offset ($timecorrection)
2255 * @param mixed $format True to use user's date format preference
2256 * @param string|bool $timecorrection The time offset as returned by
2257 * validateTimeZone() in Special:Preferences
2258 * @return string
2259 */
2264 }
2267 }
2269 /**
2270 * @param string $ts The time format which needs to be turned into a
2271 * date('YmdHis') format with wfTimestamp(TS_MW,$ts)
2272 * @param bool $adj Whether to adjust the time output according to the
2273 * user configured offset ($timecorrection)
2274 * @param mixed $format What format to return, if it's false output the
2275 * default one (default true)
2276 * @param string|bool $timecorrection The time offset as returned by
2277 * validateTimeZone() in Special:Preferences
2278 * @return string
2279 */
2284 }
2287 }
2289 /**
2290 * Takes a number of seconds and turns it into a text using values such as hours and minutes.
2291 *
2292 * @since 1.20
2293 *
2294 * @param int $seconds The amount of seconds.
2295 * @param array $chosenIntervals The intervals to enable.
2296 *
2297 * @return string
2298 */
2305 // Messages: duration-seconds, duration-minutes, duration-hours, duration-days, duration-weeks,
2306 // duration-years, duration-decades, duration-centuries, duration-millennia
2309 }
2312 }
2314 /**
2315 * Takes a number of seconds and returns an array with a set of corresponding intervals.
2316 * For example 65 will be turned into array( minutes => 1, seconds => 5 ).
2317 *
2318 * @since 1.20
2319 *
2320 * @param int $seconds The amount of seconds.
2321 * @param array $chosenIntervals The intervals to enable.
2322 *
2323 * @return array
2324 */
2335 'seconds'
2336 );
2337 }
2351 }
2352 }
2355 }
2357 /**
2358 * Internal helper function for userDate(), userTime() and userTimeAndDate()
2359 *
2360 * @param string $type Can be 'date', 'time' or 'both'
2361 * @param string $ts The time format which needs to be turned into a
2362 * date('YmdHis') format with wfTimestamp(TS_MW,$ts)
2363 * @param User $user User object used to get preferences for timezone and format
2364 * @param array $options Array, can contain the following keys:
2365 * - 'timecorrection': time correction, can have the following values:
2366 * - true: use user's preference
2367 * - false: don't use time correction
2368 * - int: value of time correction in minutes
2369 * - 'format': format to use, can have the following values:
2370 * - true: use user's preference
2371 * - false: use default preference
2372 * - string: format to use
2373 * @since 1.19
2374 * @return string
2375 */
2384 }
2386 }
2391 }
2394 }
2396 /**
2397 * Get the formatted date for the given timestamp and formatted for
2398 * the given user.
2399 *
2400 * @param mixed $ts Mixed: the time format which needs to be turned into a
2401 * date('YmdHis') format with wfTimestamp(TS_MW,$ts)
2402 * @param User $user User object used to get preferences for timezone and format
2403 * @param array $options Array, can contain the following keys:
2404 * - 'timecorrection': time correction, can have the following values:
2405 * - true: use user's preference
2406 * - false: don't use time correction
2407 * - int: value of time correction in minutes
2408 * - 'format': format to use, can have the following values:
2409 * - true: use user's preference
2410 * - false: use default preference
2411 * - string: format to use
2412 * @since 1.19
2413 * @return string
2414 */
2417 }
2419 /**
2420 * Get the formatted time for the given timestamp and formatted for
2421 * the given user.
2422 *
2423 * @param mixed $ts The time format which needs to be turned into a
2424 * date('YmdHis') format with wfTimestamp(TS_MW,$ts)
2425 * @param User $user User object used to get preferences for timezone and format
2426 * @param array $options Array, can contain the following keys:
2427 * - 'timecorrection': time correction, can have the following values:
2428 * - true: use user's preference
2429 * - false: don't use time correction
2430 * - int: value of time correction in minutes
2431 * - 'format': format to use, can have the following values:
2432 * - true: use user's preference
2433 * - false: use default preference
2434 * - string: format to use
2435 * @since 1.19
2436 * @return string
2437 */
2440 }
2442 /**
2443 * Get the formatted date and time for the given timestamp and formatted for
2444 * the given user.
2445 *
2446 * @param mixed $ts The time format which needs to be turned into a
2447 * date('YmdHis') format with wfTimestamp(TS_MW,$ts)
2448 * @param User $user User object used to get preferences for timezone and format
2449 * @param array $options Array, can contain the following keys:
2450 * - 'timecorrection': time correction, can have the following values:
2451 * - true: use user's preference
2452 * - false: don't use time correction
2453 * - int: value of time correction in minutes
2454 * - 'format': format to use, can have the following values:
2455 * - true: use user's preference
2456 * - false: use default preference
2457 * - string: format to use
2458 * @since 1.19
2459 * @return string
2460 */
2463 }
2465 /**
2466 * Convert an MWTimestamp into a pretty human-readable timestamp using
2467 * the given user preferences and relative base time.
2468 *
2469 * DO NOT USE THIS FUNCTION DIRECTLY. Instead, call MWTimestamp::getHumanTimestamp
2470 * on your timestamp object, which will then call this function. Calling
2471 * this function directly will cause hooks to be skipped over.
2472 *
2473 * @see MWTimestamp::getHumanTimestamp
2474 * @param MWTimestamp $ts Timestamp to prettify
2475 * @param MWTimestamp $relativeTo Base timestamp
2476 * @param User $user User preferences to use
2477 * @return string Human timestamp
2478 * @since 1.22
2479 */
2487 ) {
2488 // Timestamps are in different years: use full timestamp
2489 // Also do full timestamp for future dates
2490 /**
2491 * @todo FIXME: Add better handling of future timestamps.
2492 */
2496 // Timestamps are in same year, but more than 5 days ago: show day and month only.
2500 // Timestamp within the past week: show the day of the week and time
2503 // Messages:
2504 // sunday-at, monday-at, tuesday-at, wednesday-at, thursday-at, friday-at, saturday-at
2510 // Timestamp was yesterday: say 'yesterday' and the time.
2517 // Timestamp was today, but more than 90 minutes ago: say 'today' and the time.
2524 // From here on in, the timestamp was soon enough ago so that we can simply say
2525 // XX units ago, e.g., "2 hours ago" or "5 minutes ago"
2527 // Less than 90 minutes, but more than an hour ago.
2530 // A few minutes ago.
2533 // Less than a minute, but more than 30 sec ago.
2536 // Less than 30 seconds ago.
2538 }
2541 }
2543 /**
2544 * @param string $key
2545 * @return array|null
2546 */
2549 }
2551 /**
2552 * @return array
2553 */
2556 }
2558 /**
2559 * @param string $in
2560 * @param string $out
2561 * @param string $string
2562 * @return string
2563 */
2565 # This is a wrapper for iconv in all languages except esperanto,
2566 # which does some nasty x-conversions beforehand
2568 # Even with //IGNORE iconv can whine about illegal characters in
2569 # *input* string. We just ignore those too.
2570 # REF: http://bugs.php.net/bug.php?id=37166
2571 # REF: https://bugzilla.wikimedia.org/show_bug.cgi?id=16885
2576 }
2578 // callback functions for uc(), lc(), ucwords(), ucwordbreaks()
2580 /**
2581 * @param array $matches
2582 * @return mixed|string
2583 */
2586 }
2588 /**
2589 * @param array $matches
2590 * @return string
2591 */
2594 }
2596 /**
2597 * @param array $matches
2598 * @return string
2599 */
2603 }
2605 /**
2606 * @param array $matches
2607 * @return string
2608 */
2612 }
2614 /**
2615 * @param array $matches
2616 * @return string
2617 */
2620 }
2622 /**
2623 * @param array $matches
2624 * @return string
2625 */
2629 }
2631 /**
2632 * Make a string's first character uppercase
2633 *
2634 * @param string $str
2635 *
2636 * @return string
2637 */
2645 // fall back to more complex logic in case of multibyte strings
2647 }
2648 }
2650 /**
2651 * Convert a string to uppercase
2652 *
2653 * @param string $str
2654 * @param bool $first
2655 *
2656 * @return string
2657 */
2665 }
2668 }
2675 $str
2676 );
2679 }
2680 }
2681 }
2683 /**
2684 * @param string $str
2685 * @return mixed|string
2686 */
2698 }
2699 }
2701 /**
2702 * @param string $str
2703 * @param bool $first
2704 * @return mixed|string
2705 */
2713 }
2716 }
2723 $str
2724 );
2727 }
2728 }
2729 }
2731 /**
2732 * @param string $str
2733 * @return bool
2734 */
2737 }
2739 /**
2740 * @param string $str
2741 * @return mixed|string
2742 */
2747 // regexp to find first letter in each word (i.e. after each space)
2748 $replaceRegexp = "/^([a-z]|[\\xc0-\\xff][\\x80-\\xbf]*)| ([a-z]|[\\xc0-\\xff][\\x80-\\xbf]*)/";
2750 // function to use to capitalize a single char
2755 $str
2756 );
2761 $str
2762 );
2763 }
2766 }
2767 }
2769 /**
2770 * capitalize words at word breaks
2771 *
2772 * @param string $str
2773 * @return mixed
2774 */
2779 // since \b doesn't work for UTF-8, we explicitely define word break chars
2782 // find first letter after word break
2790 $str
2791 );
2796 $str
2797 );
2798 }
2803 $str
2804 );
2805 }
2806 }
2808 /**
2809 * Return a case-folded representation of $s
2810 *
2811 * This is a representation such that caseFold($s1)==caseFold($s2) if $s1
2812 * and $s2 are the same except for the case of their characters. It is not
2813 * necessary for the value returned to make sense when displayed.
2814 *
2815 * Do *not* perform any other normalisation in this function. If a caller
2816 * uses this function when it should be using a more general normalisation
2817 * function, then fix the caller.
2818 *
2819 * @param string $s
2820 *
2821 * @return string
2822 */
2825 }
2827 /**
2828 * @param string $s
2829 * @return string
2830 */
2834 }
2837 }
2840 }
2842 /**
2843 * @return array
2844 */
2847 }
2849 /**
2850 * Most writing systems use whitespace to break up words.
2851 * Some languages such as Chinese don't conventionally do this,
2852 * which requires special handling when breaking up words for
2853 * searching etc.
2854 *
2855 * @return bool
2856 */
2859 }
2861 /**
2862 * Some languages such as Chinese require word segmentation,
2863 * Specify such segmentation when overridden in derived class.
2864 *
2865 * @param string $string
2866 * @return string
2867 */
2870 }
2872 /**
2873 * Some languages have special punctuation need to be normalized.
2874 * Make such changes here.
2875 *
2876 * @param string $string
2877 * @return string
2878 */
2881 }
2883 /**
2884 * convert double-width roman characters to single-width.
2885 * range: ff00-ff5f ~= 0020-007f
2886 *
2887 * @param string $string
2888 *
2889 * @return string
2890 */
2900 }
2904 }
2906 /**
2907 * @param string $string
2908 * @param string $pattern
2909 * @return string
2910 */
2915 }
2917 /**
2918 * @param array $termsArray
2919 * @return array
2920 */
2922 # some languages, e.g. Chinese, need to do a conversion
2923 # in order for search results to be displayed correctly
2925 }
2927 /**
2928 * Get the first character of a string.
2929 *
2930 * @param string $s
2931 * @return string
2932 */
2939 $matches
2940 );
2945 }
2947 // Break down Hangul syllables to grab the first jamo
2979 }
2982 }
2983 }
2986 # Some languages may have an alternate char encoding option
2987 # (Esperanto X-coding, Japanese furigana conversion, etc)
2988 # If this language is used as the primary content language,
2989 # an override to the defaults can be set here on startup.
2990 }
2992 /**
2993 * @param string $s
2994 * @return string
2995 */
2997 # For some languages we'll want to explicitly specify
2998 # which characters make it into the edit box raw
2999 # or are converted in some way or another.
3005 }
3006 }
3008 /**
3009 * @param string $s
3010 * @return string
3011 */
3013 # Take the previous into account.
3019 }
3024 }
3025 }
3027 /**
3028 * Convert a UTF-8 string to normal form C. In Malayalam and Arabic, this
3029 * also cleans up certain backwards-compatible sequences, converting them
3030 * to the modern Unicode equivalent.
3031 *
3032 * This is language-specific for performance reasons only.
3033 *
3034 * @param string $s
3035 *
3036 * @return string
3037 */
3044 }
3047 }
3049 /**
3050 * Transform a string using serialized data stored in the given file (which
3051 * must be in the serialized subdirectory of $IP). The file contains pairs
3052 * mapping source characters to destination characters.
3053 *
3054 * The data is cached in process memory. This will go faster if you have the
3055 * FastStringSearch extension.
3056 *
3057 * @param string $file
3058 * @param string $string
3059 *
3060 * @throws MWException
3061 * @return string
3062 */
3068 }
3070 }
3072 }
3074 /**
3075 * For right-to-left language support
3076 *
3077 * @return bool
3078 */
3081 }
3083 /**
3084 * Return the correct HTML 'dir' attribute value for this language.
3085 * @return string
3086 */
3089 }
3091 /**
3092 * Return 'left' or 'right' as appropriate alignment for line-start
3093 * for this language's text direction.
3094 *
3095 * Should be equivalent to CSS3 'start' text-align value....
3096 *
3097 * @return string
3098 */
3101 }
3103 /**
3104 * Return 'right' or 'left' as appropriate alignment for line-end
3105 * for this language's text direction.
3106 *
3107 * Should be equivalent to CSS3 'end' text-align value....
3108 *
3109 * @return string
3110 */
3113 }
3115 /**
3116 * A hidden direction mark (LRM or RLM), depending on the language direction.
3117 * Unlike getDirMark(), this function returns the character as an HTML entity.
3118 * This function should be used when the output is guaranteed to be HTML,
3119 * because it makes the output HTML source code more readable. When
3120 * the output is plain text or can be escaped, getDirMark() should be used.
3121 *
3122 * @param bool $opposite Get the direction mark opposite to your language
3123 * @return string
3124 * @since 1.20
3125 */
3129 }
3131 }
3133 /**
3134 * A hidden direction mark (LRM or RLM), depending on the language direction.
3135 * This function produces them as invisible Unicode characters and
3136 * the output may be hard to read and debug, so it should only be used
3137 * when the output is plain text or can be escaped. When the output is
3138 * HTML, use getDirMarkEntity() instead.
3139 *
3140 * @param bool $opposite Get the direction mark opposite to your language
3141 * @return string
3142 */
3148 }
3150 }
3152 /**
3153 * @return array
3154 */
3157 }
3159 /**
3160 * An arrow, depending on the language direction.
3161 *
3162 * @param string $direction The direction of the arrow: forwards (default),
3163 * backwards, left, right, up, down.
3164 * @return string
3165 */
3180 }
3181 }
3183 /**
3184 * To allow "foo[[bar]]" to extend the link over the whole word "foobar"
3185 *
3186 * @return bool
3187 */
3190 }
3192 /**
3193 * Get all magic words from cache.
3194 * @return array
3195 */
3198 }
3200 /**
3201 * Run the LanguageGetMagic hook once.
3202 */
3206 }
3209 }
3211 /**
3212 * Fill a MagicWord object with data from here
3213 *
3214 * @param MagicWord $mw
3215 */
3217 // Saves a function call
3220 }
3227 }
3234 }
3235 }
3237 /**
3238 * Add magic words to the extension array
3239 *
3240 * @param array $newWords
3241 */
3248 }
3249 }
3250 }
3252 /**
3253 * Get special page names, as an associative array
3254 * canonical name => array of valid names, including aliases
3255 * @return array
3256 */
3258 // Cache aliases because it may be slow to load them
3260 // Initialise array
3265 }
3268 }
3270 /**
3271 * Italic is unsuitable for some languages
3272 *
3273 * @param string $text The text to be emphasized.
3274 * @return string
3275 */
3278 }
3280 /**
3281 * Normally we output all numbers in plain en_US style, that is
3282 * 293,291.235 for twohundredninetythreethousand-twohundredninetyone
3283 * point twohundredthirtyfive. However this is not suitable for all
3284 * languages, some such as Punjabi want ੨੯੩,੨੯੫.੨੩੫ and others such as
3285 * Icelandic just want to use commas instead of dots, and dots instead
3286 * of commas like "293.291,235".
3287 *
3288 * An example of this function being called:
3289 * <code>
3290 * wfMessage( 'message' )->numParams( $num )->text()
3291 * </code>
3292 *
3293 * See $separatorTransformTable on MessageIs.php for
3294 * the , => . and . => , implementation.
3295 *
3296 * @todo check if it's viable to use localeconv() for the decimal separator thing.
3297 * @param int|float $number The string to be formatted, should be an integer
3298 * or a floating point number.
3299 * @param bool $nocommafy Set to true for special numbers like dates
3300 * @return string
3301 */
3309 }
3310 }
3316 }
3317 }
3320 }
3322 /**
3323 * Front-end for non-commafied formatNum
3324 *
3325 * @param int|float $number The string to be formatted, should be an integer
3326 * or a floating point number.
3327 * @since 1.21
3328 * @return string
3329 */
3332 }
3334 /**
3335 * @param string $number
3336 * @return string
3337 */
3341 // eliminate empty array values such as ''. (bug 64347)
3344 }
3348 // eliminate empty array values such as ''. (bug 64347)
3351 }
3355 }
3357 /**
3358 * Adds commas to a given number
3359 * @since 1.19
3360 * @param mixed $number
3361 * @return string
3362 */
3367 }
3370 // default grouping is at thousands, use the same for ###,###,### pattern too.
3373 // Ref: http://cldr.unicode.org/translation/number-patterns
3376 // For negative numbers apply the algorithm like positive number and add sign.
3379 }
3387 // the string does not have any number part. Eg: .12345
3389 }
3397 }
3401 // use the last pattern for the rest of the number
3403 }
3406 }
3407 }
3409 }
3410 }
3412 /**
3413 * @return string
3414 */
3417 }
3419 /**
3420 * @return array
3421 */
3424 }
3426 /**
3427 * @return array
3428 */
3431 }
3433 /**
3434 * Take a list of strings and build a locale-friendly comma-separated
3435 * list, using the local comma-separator message.
3436 * The last two strings are chained with an "and".
3437 * NOTE: This function will only work with standard numeric array keys (0, 1, 2…)
3438 *
3439 * @param string[] $l
3440 * @return string
3441 */
3446 }
3452 }
3453 }
3460 }
3461 }
3463 }
3465 /**
3466 * Take a list of strings and build a locale-friendly comma-separated
3467 * list, using the local comma-separator message.
3468 * @param string[] $list Array of strings to put in a comma list
3469 * @return string
3470 */
3474 $list
3475 );
3476 }
3478 /**
3479 * Take a list of strings and build a locale-friendly semicolon-separated
3480 * list, using the local semicolon-separator message.
3481 * @param string[] $list Array of strings to put in a semicolon list
3482 * @return string
3483 */
3487 $list
3488 );
3489 }
3491 /**
3492 * Same as commaList, but separate it with the pipe instead.
3493 * @param string[] $list Array of strings to put in a pipe list
3494 * @return string
3495 */
3499 $list
3500 );
3501 }
3503 /**
3504 * Truncate a string to a specified length in bytes, appending an optional
3505 * string (e.g. for ellipses)
3506 *
3507 * The database offers limited byte lengths for some columns in the database;
3508 * multi-byte character sets mean we need to ensure that only whole characters
3509 * are included, otherwise broken characters can be passed to the user
3510 *
3511 * If $length is negative, the string will be truncated from the beginning
3512 *
3513 * @param string $string String to truncate
3514 * @param int $length Maximum length (including ellipses)
3515 * @param string $ellipsis String to append to the truncated text
3516 * @param bool $adjustLength Subtract length of ellipsis from $length.
3517 * $adjustLength was introduced in 1.18, before that behaved as if false.
3518 * @return string
3519 */
3521 # Use the localized ellipsis character
3524 }
3525 # Check if there is no need to truncate
3530 }
3532 # If ellipsis length is >= $length then we can't apply $adjustLength
3535 # Otherwise, truncate and add ellipsis...
3550 }
3551 }
3552 # Do not truncate if the ellipsis makes the string longer/equal (bug 22181).
3553 # This check is *not* redundant if $adjustLength, due to the single case where
3554 # LEN($ellipsis) > ABS($limit arg); $stringOriginal could be shorter than $string.
3559 }
3560 }
3562 /**
3563 * Remove bytes that represent an incomplete Unicode character
3564 * at the end of string (e.g. bytes of the char are missing)
3565 *
3566 * @param string $string
3567 * @return string
3568 */
3574 # We got the first byte only of a multibyte char; remove it.
3579 ) {
3580 # We chopped in the middle of a character; remove it
3582 }
3583 }
3585 }
3587 /**
3588 * Remove bytes that represent an incomplete Unicode character
3589 * at the start of string (e.g. bytes of the char are missing)
3590 *
3591 * @param string $string
3592 * @return string
3593 */
3598 # We chopped in the middle of a character; remove the whole thing
3600 }
3601 }
3603 }
3605 /**
3606 * Truncate a string of valid HTML to a specified length in bytes,
3607 * appending an optional string (e.g. for ellipses), and return valid HTML
3608 *
3609 * This is only intended for styled/linked text, such as HTML with
3610 * tags like <span> and <a>, were the tags are self-contained (valid HTML).
3611 * Also, this will not detect things like "display:none" CSS.
3612 *
3613 * Note: since 1.18 you do not need to leave extra room in $length for ellipses.
3614 *
3615 * @param string $text HTML string to truncate
3616 * @param int $length (zero/positive) Maximum length (including ellipses)
3617 * @param string $ellipsis String to append to the truncated text
3618 * @return string
3619 */
3621 # Use the localized ellipsis character
3624 }
3625 # Check if there is clearly no need to truncate
3630 }
3644 # Consider truncation once the display length has reached the maximim.
3645 # We check if $dispLen > 0 to grab tags for the $neLength = 0 case.
3646 # Check that we're not in the middle of a bracket/entity...
3650 # Save where we are; we will truncate here unless there turn out to
3651 # be so few remaining characters that truncation is not necessary.
3654 }
3656 # String in fact does need truncation, the truncation point was OK.
3661 }
3662 }
3665 }
3667 # Read the next char...
3685 }
3691 // Name found (e.g. "<a href=..."), add on tag attributes...
3693 }
3699 }
3702 // Save state without $ch. We want to *hit* the first
3703 // display char (to get tags) but not *use* it if truncating.
3705 }
3710 // Add the next $max display text chars after this in one swoop...
3715 }
3716 }
3717 }
3718 }
3719 // Close the last tag if left unclosed by bad HTML
3723 }
3725 }
3727 /**
3728 * truncateHtml() helper function
3729 * like strcspn() but adds the skipped chars to $ret
3730 *
3731 * @param string $ret
3732 * @param string $text
3733 * @param string $search
3734 * @param int $start
3735 * @param null|int $len
3736 * @return int
3737 */
3743 }
3748 }
3750 }
3752 /**
3753 * truncateHtml() helper function
3754 * (a) push or pop $tag from $openTags as needed
3755 * (b) clear $tag value
3756 * @param string &$tag Current HTML tag name we are looking at
3757 * @param int $tagType (0-open tag, 1-close tag)
3758 * @param string $lastCh Character before the '>' that ended this tag
3759 * @param array &$openTags Open tag stack (not accounting for $tag)
3760 */
3769 }
3770 }
3772 }
3773 }
3775 /**
3776 * Grammatical transformations, needed for inflected languages
3777 * Invoked by putting {{grammar:case|word}} in a message
3778 *
3779 * @param string $word
3780 * @param string $case
3781 * @return string
3782 */
3787 }
3790 }
3791 /**
3792 * Get the grammar forms for the content language
3793 * @return array Array of grammar forms
3794 * @since 1.20
3795 */
3800 ) {
3802 }
3805 }
3806 /**
3807 * Provides an alternative text depending on specified gender.
3808 * Usage {{gender:username|masculine|feminine|unknown}}.
3809 * username is optional, in which case the gender of current user is used,
3810 * but only in (some) interface messages; otherwise default gender is used.
3811 *
3812 * If no forms are given, an empty string is returned. If only one form is
3813 * given, it will be returned unconditionally. These details are implied by
3814 * the caller and cannot be overridden in subclasses.
3815 *
3816 * If three forms are given, the default is to use the third (unknown) form.
3817 * If fewer than three forms are given, the default is to use the first (masculine) form.
3818 * These details can be overridden in subclasses.
3819 *
3820 * @param string $gender
3821 * @param array $forms
3822 *
3823 * @return string
3824 */
3828 }
3832 }
3835 }
3837 }
3839 /**
3840 * Plural form transformations, needed for some languages.
3841 * For example, there are 3 form of plural in Russian and Polish,
3842 * depending on "count mod 10". See [[w:Plural]]
3843 * For English it is pretty simple.
3844 *
3845 * Invoked by putting {{plural:count|wordform1|wordform2}}
3846 * or {{plural:count|wordform1|wordform2|wordform3}}
3847 *
3848 * Example: {{plural:{{NUMBEROFARTICLES}}|article|articles}}
3849 *
3850 * @param int $count Non-localized number
3851 * @param array $forms Different plural forms
3852 * @return string Correct form of plural for $count in this language
3853 */
3855 // Handle explicit n=pluralform cases
3859 }
3862 }
3867 }
3869 /**
3870 * Handles explicit plural forms for Language::convertPlural()
3871 *
3872 * In {{PLURAL:$1|0=nothing|one|many}}, 0=nothing will be returned if $1 equals zero.
3873 * If an explicitly defined plural form matches the $count, then
3874 * string value returned, otherwise array returned for further consideration
3875 * by CLDR rules or overridden convertPlural().
3876 *
3877 * @since 1.23
3878 *
3879 * @param int $count Non-localized number
3880 * @param array $forms Different plural forms
3881 *
3882 * @return array|string
3883 */
3890 }
3892 }
3893 }
3895 }
3897 /**
3898 * Checks that convertPlural was given an array and pads it to requested
3899 * amount of forms by copying the last one.
3900 *
3901 * @param array $forms Array of forms given to convertPlural
3902 * @param int $count How many forms should there be at least
3903 * @return array Padded array of forms or an exception if not an array
3904 */
3908 }
3910 }
3912 /**
3913 * @todo Maybe translate block durations. Note that this function is somewhat misnamed: it
3914 * deals with translating the *duration* ("1 week", "4 days", etc), not the expiry time
3915 * (which is an absolute timestamp). Please note: do NOT add this blindly, as it is used
3916 * on old expiry lengths recorded in log entries. You'd need to provide the start date to
3917 * match up with it.
3918 *
3919 * @param string $str The validated block duration in English
3920 * @return string Somehow translated block duration
3921 * @see LanguageFi.php for example implementation
3922 */
3928 }
3929 }
3935 }
3936 }
3937 }
3939 // If all else fails, return a standard duration or timestamp description.
3944 // $time is relative to 0 so it's a duration length.
3948 // wfTimestamp() handles 0 as current time instead of epoch.
3952 }
3953 }
3954 }
3956 /**
3957 * languages like Chinese need to be segmented in order for the diff
3958 * to be of any use
3959 *
3960 * @param string $text
3961 * @return string
3962 */
3965 }
3967 /**
3968 * and unsegment to show the result
3969 *
3970 * @param string $text
3971 * @return string
3972 */
3975 }
3977 /**
3978 * Return the LanguageConverter used in the Language
3979 *
3980 * @since 1.19
3981 * @return LanguageConverter
3982 */
3985 }
3987 /**
3988 * convert text to all supported variants
3989 *
3990 * @param string $text
3991 * @return array
3992 */
3995 }
3997 /**
3998 * convert text to different variants of a language.
3999 *
4000 * @param string $text
4001 * @return string
4002 */
4005 }
4007 /**
4008 * Convert a Title object to a string in the preferred variant
4009 *
4010 * @param Title $title
4011 * @return string
4012 */
4015 }
4017 /**
4018 * Convert a namespace index to a string in the preferred variant
4019 *
4020 * @param int $ns
4021 * @return string
4022 */
4025 }
4027 /**
4028 * Check if this is a language with variants
4029 *
4030 * @return bool
4031 */
4034 }
4036 /**
4037 * Check if the language has the specific variant
4038 *
4039 * @since 1.19
4040 * @param string $variant
4041 * @return bool
4042 */
4045 }
4047 /**
4048 * Put custom tags (e.g. -{ }-) around math to prevent conversion
4049 *
4050 * @param string $text
4051 * @return string
4052 * @deprecated since 1.22 is no longer used
4053 */
4056 }
4058 /**
4059 * Perform output conversion on a string, and encode for safe HTML output.
4060 * @param string $text Text to be converted
4061 * @param bool $isTitle Whether this conversion is for the article title
4062 * @return string
4063 * @todo this should get integrated somewhere sane
4064 */
4067 }
4069 /**
4070 * @param string $key
4071 * @return string
4072 */
4075 }
4077 /**
4078 * Get the list of variants supported by this language
4079 * see sample implementation in LanguageZh.php
4080 *
4081 * @return array An array of language codes
4082 */
4085 }
4087 /**
4088 * @return string
4089 */
4092 }
4094 /**
4095 * @return string
4096 */
4099 }
4101 /**
4102 * @return string
4103 */
4106 }
4108 /**
4109 * If a language supports multiple variants, it is
4110 * possible that non-existing link in one variant
4111 * actually exists in another variant. this function
4112 * tries to find it. See e.g. LanguageZh.php
4113 * The input parameters may be modified upon return
4114 *
4115 * @param string &$link The name of the link
4116 * @param Title &$nt The title object of the link
4117 * @param bool $ignoreOtherCond To disable other conditions when
4118 * we need to transclude a template or update a category's link
4119 */
4122 }
4124 /**
4125 * returns language specific options used by User::getPageRenderHash()
4126 * for example, the preferred language variant
4127 *
4128 * @return string
4129 */
4132 }
4134 /**
4135 * For languages that support multiple variants, the title of an
4136 * article may be displayed differently in different variants. this
4137 * function returns the apporiate title defined in the body of the article.
4138 *
4139 * @return string
4140 */
4143 }
4145 /**
4146 * Prepare external link text for conversion. When the text is
4147 * a URL, it shouldn't be converted, and it'll be wrapped in
4148 * the "raw" tag (-{R| }-) to prevent conversion.
4149 *
4150 * This function is called "markNoConversion" for historical
4151 * reasons.
4152 *
4153 * @param string $text Text to be used for external link
4154 * @param bool $noParse Wrap it without confirming it's a real URL first
4155 * @return string The tagged text
4156 */
4158 // Excluding protocal-relative URLs may avoid many false positives.
4163 }
4164 }
4166 /**
4167 * A regular expression to match legal word-trailing characters
4168 * which should be merged onto a link of the form [[foo]]bar.
4169 *
4170 * @return string
4171 */
4174 }
4176 /**
4177 * A regular expression character set to match legal word-prefixing
4178 * characters which should be merged onto a link of the form foo[[bar]].
4179 *
4180 * @return string
4181 */
4184 }
4186 /**
4187 * @deprecated since 1.24, will be removed in 1.25
4188 * @return Language
4189 */
4193 }
4195 /**
4196 * Get the "parent" language which has a converter to convert a "compatible" language
4197 * (in another variant) to this language (eg. zh for zh-cn, but not en for en-gb).
4198 *
4199 * @return Language|null
4200 * @since 1.22
4201 */
4205 }
4212 }
4217 }
4221 }
4223 /**
4224 * Get the RFC 3066 code for this language object
4225 *
4226 * NOTE: The return value of this function is NOT HTML-safe and must be escaped with
4227 * htmlspecialchars() or similar
4228 *
4229 * @return string
4230 */
4233 }
4235 /**
4236 * Get the code in Bcp47 format which we can use
4237 * inside of html lang="" tags.
4238 *
4239 * NOTE: The return value of this function is NOT HTML-safe and must be escaped with
4240 * htmlspecialchars() or similar.
4241 *
4242 * @since 1.19
4243 * @return string
4244 */
4248 }
4250 }
4252 /**
4253 * @param string $code
4254 */
4257 // Ensure we don't leave incorrect cached data lying around
4260 }
4262 /**
4263 * Get the name of a file for a certain language code
4264 * @param string $prefix Prepend this to the filename
4265 * @param string $code Language code
4266 * @param string $suffix Append this to the filename
4267 * @throws MWException
4268 * @return string $prefix . $mangledCode . $suffix
4269 */
4273 }
4276 }
4278 /**
4279 * Get the language code from a file name. Inverse of getFileName()
4280 * @param string $filename $prefix . $languageCode . $suffix
4281 * @param string $prefix Prefix before the language code
4282 * @param string $suffix Suffix after the language code
4283 * @return string Language code, or false if $prefix or $suffix isn't found
4284 */
4285 public static function getCodeFromFileName( $filename, $prefix = 'Language', $suffix = '.php' ) {
4291 }
4293 }
4295 /**
4296 * @param string $code
4297 * @return string
4298 */
4304 }
4306 /**
4307 * @param string $code
4308 * @return string
4309 * @since 1.23
4310 */
4316 }
4319 }
4321 /**
4322 * @param string $code
4323 * @return string
4324 */
4328 }
4330 /**
4331 * Get the first fallback for a given language.
4332 *
4333 * @param string $code
4334 *
4335 * @return bool|string
4336 */
4344 }
4345 }
4347 /**
4348 * Get the ordered list of fallback languages.
4349 *
4350 * @since 1.19
4351 * @param string $code Language code
4352 * @return array
4353 */
4362 }
4364 }
4365 }
4367 /**
4368 * Get the ordered list of fallback languages, ending with the fallback
4369 * language chain for the site language.
4370 *
4371 * @since 1.22
4372 * @param string $code Language code
4373 * @return array Array( fallbacks, site fallbacks )
4374 */
4378 // Usually, we will only store a tiny number of fallback chains, so we
4379 // keep them in static memory.
4385 // Append the site's fallback chain, including the site language itself
4389 // Eliminate any languages already included in the chain
4393 }
4395 }
4397 /**
4398 * Get all messages for a given language
4399 * WARNING: this may take a long time. If you just need all message *keys*
4400 * but need the *contents* of only a few messages, consider using getMessageKeysFor().
4401 *
4402 * @param string $code
4403 *
4404 * @return array
4405 */
4408 }
4410 /**
4411 * Get a message for a given language
4412 *
4413 * @param string $key
4414 * @param string $code
4415 *
4416 * @return string
4417 */
4420 }
4422 /**
4423 * Get all message keys for a given language. This is a faster alternative to
4424 * array_keys( Language::getMessagesFor( $code ) )
4425 *
4426 * @since 1.19
4427 * @param string $code Language code
4428 * @return array Array of message keys (strings)
4429 */
4432 }
4434 /**
4435 * @param string $talk
4436 * @return mixed
4437 */
4441 }
4446 # Allow grammar transformations
4447 # Allowing full message-style parsing would make simple requests
4448 # such as action=raw much more expensive than they need to be.
4449 # This will hopefully cover most cases.
4453 }
4455 /**
4456 * @param string $m
4457 * @return string
4458 */
4461 }
4463 /**
4464 * @throws MWException
4465 * @return array
4466 */
4471 }
4477 }
4481 }
4483 /**
4484 * Decode an expiry (block, protection, etc) which has come from the DB
4485 *
4486 * @param string $expiry Database expiry String
4487 * @param bool|int $format True to process using language functions, or TS_ constant
4488 * to return the expiry in a given timestamp
4489 * @param string $inifinity If $format is not true, use this string for infinite expiry
4490 * @return string
4491 * @since 1.18
4492 */
4497 }
4507 }
4508 }
4510 /**
4511 * @todo Document
4512 * @param int|float $seconds
4513 * @param array $format Optional
4514 * If $format['avoid'] === 'avoidseconds': don't mention seconds if $seconds >= 1 hour.
4515 * If $format['avoid'] === 'avoidminutes': don't mention seconds/minutes if $seconds > 48 hours.
4516 * If $format['noabbrevs'] is true: use 'seconds' and friends instead of 'seconds-abbrev'
4517 * and friends.
4518 * For backwards compatibility, $format may also be one of the strings 'avoidseconds'
4519 * or 'avoidminutes'.
4520 * @return string
4521 */
4525 }
4528 }
4531 }
4553 }
4564 }
4568 }
4574 }
4582 }
4592 }
4596 }
4606 }
4607 }
4609 }
4611 /**
4612 * Format a bitrate for output, using an appropriate
4613 * unit (bps, kbps, Mbps, Gbps, Tbps, Pbps, Ebps, Zbps or Ybps) according to
4614 * the magnitude in question.
4615 *
4616 * This use base 1000. For base 1024 use formatSize(), for another base
4617 * see formatComputingNumbers().
4618 *
4619 * @param int $bps
4620 * @return string
4621 */
4624 }
4626 /**
4627 * @param int $size Size of the unit
4628 * @param int $boundary Size boundary (1000, or 1024 in most cases)
4629 * @param string $messageKey Message key to be uesd
4630 * @return string
4631 */
4636 );
4637 }
4645 }
4647 // For small sizes no decimal places necessary
4650 // For MB and bigger two decimal places are smarter
4652 }
4658 }
4660 /**
4661 * Format a size in bytes for output, using an appropriate
4662 * unit (B, KB, MB, GB, TB, PB, EB, ZB or YB) according to the magnitude in question
4663 *
4664 * This method use base 1024. For base 1000 use formatBitrate(), for
4665 * another base see formatComputingNumbers()
4666 *
4667 * @param int $size Size to format
4668 * @return string Plain text (not HTML)
4669 */
4672 }
4674 /**
4675 * Make a list item, used by various special pages
4676 *
4677 * @param string $page Page link
4678 * @param string $details HTML safe text between brackets
4679 * @param bool $oppositedm Add the direction mark opposite to your
4680 * language, to display text properly
4681 * @return HTML escaped string
4682 */
4686 }
4689 return
4694 }
4696 /**
4697 * Generate (prev x| next x) (20|50|100...) type links for paging
4698 *
4699 * @param Title $title Title object to link
4700 * @param int $offset
4701 * @param int $limit
4702 * @param array $query Optional URL query parameter string
4703 * @param bool $atend Optional param for specified if this is the last page
4704 * @return string
4705 */
4708 ) {
4709 // @todo FIXME: Why on earth this needs one message for the text and another one for tooltip?
4711 # Make 'previous' link
4712 $prev = wfMessage( 'prevn' )->inLanguage( $this )->title( $title )->numParams( $limit )->text();
4718 }
4720 # Make 'next' link
4721 $next = wfMessage( 'nextn' )->inLanguage( $this )->title( $title )->numParams( $limit )->text();
4727 }
4729 # Make links to set number of items per page
4734 }
4738 }
4740 /**
4741 * Helper function for viewPrevNext() that generates links
4742 *
4743 * @param Title $title Title object to link
4744 * @param int $offset
4745 * @param int $limit
4746 * @param array $query Extra query parameters
4747 * @param string $link Text to use for the link; will be escaped
4748 * @param string $tooltipMsg Name of the message to use as tooltip
4749 * @param string $class Value of the "class" attribute of the link
4750 * @return string HTML fragment
4751 */
4754 ) {
4761 }
4763 /**
4764 * Get the conversion rule title, if any.
4765 *
4766 * @return string
4767 */
4770 }
4772 /**
4773 * Get the compiled plural rules for the language
4774 * @since 1.20
4775 * @return array Associative array with plural form, and plural rule as key-value pairs
4776 */
4782 $pluralRules = self::$dataCache->getItem( strtolower( $fallbackCode ), 'compiledPluralRules' );
4785 }
4786 }
4787 }
4789 }
4791 /**
4792 * Get the plural rules for the language
4793 * @since 1.20
4794 * @return array Associative array with plural form number and plural rule as key-value pairs
4795 */
4804 }
4805 }
4806 }
4808 }
4810 /**
4811 * Get the plural rule types for the language
4812 * @since 1.22
4813 * @return array Associative array with plural form number and plural rule type as key-value pairs
4814 */
4820 $pluralRuleTypes = self::$dataCache->getItem( strtolower( $fallbackCode ), 'pluralRuleTypes' );
4823 }
4824 }
4825 }
4827 }
4829 /**
4830 * Find the index number of the plural rule appropriate for the given number
4831 * @param int $number
4832 * @return int The index number of the plural rule
4833 */
4838 }
4840 /**
4841 * Find the plural rule type appropriate for the given number
4842 * For example, if the language is set to Arabic, getPluralType(5) should
4843 * return 'few'.
4844 * @since 1.22
4845 * @param int $number
4846 * @return string The name of the plural rule type, e.g. one, two, few, many
4847 */
4855 }
4856 }
4857 }