MediaWiki REL1_34
Language.php
Go to the documentation of this file.
1<?php
29use CLDRPluralRuleParser\Evaluator;
32
37class Language {
43 const AS_AUTONYMS = LanguageNameUtils::AUTONYMS;
44
50 const ALL = LanguageNameUtils::ALL;
51
58 const SUPPORTED = LanguageNameUtils::SUPPORTED;
59
64
65 public $mVariants, $mCode, $mLoaded = false;
66 public $mMagicExtensions = [];
67 private $mHtmlCode = null;
69 private $mParentLanguage = false;
70
71 public $dateFormatStrings = [];
73
75 protected $namespaceNames;
77
81 public $transformData = [];
82
85
88
89 public static $mLangObjCache = [];
90
96
102
103 // TODO Make these const once we drop HHVM support (T192166)
104 public static $mWeekdayMsgs = [
105 'sunday', 'monday', 'tuesday', 'wednesday', 'thursday',
106 'friday', 'saturday'
107 ];
108
109 public static $mWeekdayAbbrevMsgs = [
110 'sun', 'mon', 'tue', 'wed', 'thu', 'fri', 'sat'
111 ];
112
113 public static $mMonthMsgs = [
114 'january', 'february', 'march', 'april', 'may_long', 'june',
115 'july', 'august', 'september', 'october', 'november',
116 'december'
117 ];
118 public static $mMonthGenMsgs = [
119 'january-gen', 'february-gen', 'march-gen', 'april-gen', 'may-gen', 'june-gen',
120 'july-gen', 'august-gen', 'september-gen', 'october-gen', 'november-gen',
121 'december-gen'
122 ];
123 public static $mMonthAbbrevMsgs = [
124 'jan', 'feb', 'mar', 'apr', 'may', 'jun', 'jul', 'aug',
125 'sep', 'oct', 'nov', 'dec'
126 ];
127
129 'iranian-calendar-m1', 'iranian-calendar-m2', 'iranian-calendar-m3',
130 'iranian-calendar-m4', 'iranian-calendar-m5', 'iranian-calendar-m6',
131 'iranian-calendar-m7', 'iranian-calendar-m8', 'iranian-calendar-m9',
132 'iranian-calendar-m10', 'iranian-calendar-m11', 'iranian-calendar-m12'
133 ];
134
136 'hebrew-calendar-m1', 'hebrew-calendar-m2', 'hebrew-calendar-m3',
137 'hebrew-calendar-m4', 'hebrew-calendar-m5', 'hebrew-calendar-m6',
138 'hebrew-calendar-m7', 'hebrew-calendar-m8', 'hebrew-calendar-m9',
139 'hebrew-calendar-m10', 'hebrew-calendar-m11', 'hebrew-calendar-m12',
140 'hebrew-calendar-m6a', 'hebrew-calendar-m6b'
141 ];
142
144 'hebrew-calendar-m1-gen', 'hebrew-calendar-m2-gen', 'hebrew-calendar-m3-gen',
145 'hebrew-calendar-m4-gen', 'hebrew-calendar-m5-gen', 'hebrew-calendar-m6-gen',
146 'hebrew-calendar-m7-gen', 'hebrew-calendar-m8-gen', 'hebrew-calendar-m9-gen',
147 'hebrew-calendar-m10-gen', 'hebrew-calendar-m11-gen', 'hebrew-calendar-m12-gen',
148 'hebrew-calendar-m6a-gen', 'hebrew-calendar-m6b-gen'
149 ];
150
151 public static $mHijriCalendarMonthMsgs = [
152 'hijri-calendar-m1', 'hijri-calendar-m2', 'hijri-calendar-m3',
153 'hijri-calendar-m4', 'hijri-calendar-m5', 'hijri-calendar-m6',
154 'hijri-calendar-m7', 'hijri-calendar-m8', 'hijri-calendar-m9',
155 'hijri-calendar-m10', 'hijri-calendar-m11', 'hijri-calendar-m12'
156 ];
157
162 public static $durationIntervals = [
163 'millennia' => 31556952000,
164 'centuries' => 3155695200,
165 'decades' => 315569520,
166 'years' => 31556952, // 86400 * ( 365 + ( 24 * 3 + 25 ) / 400 )
167 'weeks' => 604800,
168 'days' => 86400,
169 'hours' => 3600,
170 'minutes' => 60,
171 'seconds' => 1,
172 ];
173
180 private static $fallbackLanguageCache = [];
181
187
191 private static $lre = "\u{202A}"; // U+202A LEFT-TO-RIGHT EMBEDDING
192 private static $rle = "\u{202B}"; // U+202B RIGHT-TO-LEFT EMBEDDING
193 private static $pdf = "\u{202C}"; // U+202C POP DIRECTIONAL FORMATTING
194
206 // @codeCoverageIgnoreStart
207 // phpcs:ignore Generic.Files.LineLength
208 private static $strongDirRegex = '/(?:([\x{41}-\x{5a}\x{61}-\x{7a}\x{aa}\x{b5}\x{ba}\x{c0}-\x{d6}\x{d8}-\x{f6}\x{f8}-\x{2b8}\x{2bb}-\x{2c1}\x{2d0}\x{2d1}\x{2e0}-\x{2e4}\x{2ee}\x{370}-\x{373}\x{376}\x{377}\x{37a}-\x{37d}\x{37f}\x{386}\x{388}-\x{38a}\x{38c}\x{38e}-\x{3a1}\x{3a3}-\x{3f5}\x{3f7}-\x{482}\x{48a}-\x{52f}\x{531}-\x{556}\x{559}-\x{55f}\x{561}-\x{587}\x{589}\x{903}-\x{939}\x{93b}\x{93d}-\x{940}\x{949}-\x{94c}\x{94e}-\x{950}\x{958}-\x{961}\x{964}-\x{980}\x{982}\x{983}\x{985}-\x{98c}\x{98f}\x{990}\x{993}-\x{9a8}\x{9aa}-\x{9b0}\x{9b2}\x{9b6}-\x{9b9}\x{9bd}-\x{9c0}\x{9c7}\x{9c8}\x{9cb}\x{9cc}\x{9ce}\x{9d7}\x{9dc}\x{9dd}\x{9df}-\x{9e1}\x{9e6}-\x{9f1}\x{9f4}-\x{9fa}\x{a03}\x{a05}-\x{a0a}\x{a0f}\x{a10}\x{a13}-\x{a28}\x{a2a}-\x{a30}\x{a32}\x{a33}\x{a35}\x{a36}\x{a38}\x{a39}\x{a3e}-\x{a40}\x{a59}-\x{a5c}\x{a5e}\x{a66}-\x{a6f}\x{a72}-\x{a74}\x{a83}\x{a85}-\x{a8d}\x{a8f}-\x{a91}\x{a93}-\x{aa8}\x{aaa}-\x{ab0}\x{ab2}\x{ab3}\x{ab5}-\x{ab9}\x{abd}-\x{ac0}\x{ac9}\x{acb}\x{acc}\x{ad0}\x{ae0}\x{ae1}\x{ae6}-\x{af0}\x{af9}\x{b02}\x{b03}\x{b05}-\x{b0c}\x{b0f}\x{b10}\x{b13}-\x{b28}\x{b2a}-\x{b30}\x{b32}\x{b33}\x{b35}-\x{b39}\x{b3d}\x{b3e}\x{b40}\x{b47}\x{b48}\x{b4b}\x{b4c}\x{b57}\x{b5c}\x{b5d}\x{b5f}-\x{b61}\x{b66}-\x{b77}\x{b83}\x{b85}-\x{b8a}\x{b8e}-\x{b90}\x{b92}-\x{b95}\x{b99}\x{b9a}\x{b9c}\x{b9e}\x{b9f}\x{ba3}\x{ba4}\x{ba8}-\x{baa}\x{bae}-\x{bb9}\x{bbe}\x{bbf}\x{bc1}\x{bc2}\x{bc6}-\x{bc8}\x{bca}-\x{bcc}\x{bd0}\x{bd7}\x{be6}-\x{bf2}\x{c01}-\x{c03}\x{c05}-\x{c0c}\x{c0e}-\x{c10}\x{c12}-\x{c28}\x{c2a}-\x{c39}\x{c3d}\x{c41}-\x{c44}\x{c58}-\x{c5a}\x{c60}\x{c61}\x{c66}-\x{c6f}\x{c7f}\x{c82}\x{c83}\x{c85}-\x{c8c}\x{c8e}-\x{c90}\x{c92}-\x{ca8}\x{caa}-\x{cb3}\x{cb5}-\x{cb9}\x{cbd}-\x{cc4}\x{cc6}-\x{cc8}\x{cca}\x{ccb}\x{cd5}\x{cd6}\x{cde}\x{ce0}\x{ce1}\x{ce6}-\x{cef}\x{cf1}\x{cf2}\x{d02}\x{d03}\x{d05}-\x{d0c}\x{d0e}-\x{d10}\x{d12}-\x{d3a}\x{d3d}-\x{d40}\x{d46}-\x{d48}\x{d4a}-\x{d4c}\x{d4e}\x{d57}\x{d5f}-\x{d61}\x{d66}-\x{d75}\x{d79}-\x{d7f}\x{d82}\x{d83}\x{d85}-\x{d96}\x{d9a}-\x{db1}\x{db3}-\x{dbb}\x{dbd}\x{dc0}-\x{dc6}\x{dcf}-\x{dd1}\x{dd8}-\x{ddf}\x{de6}-\x{def}\x{df2}-\x{df4}\x{e01}-\x{e30}\x{e32}\x{e33}\x{e40}-\x{e46}\x{e4f}-\x{e5b}\x{e81}\x{e82}\x{e84}\x{e87}\x{e88}\x{e8a}\x{e8d}\x{e94}-\x{e97}\x{e99}-\x{e9f}\x{ea1}-\x{ea3}\x{ea5}\x{ea7}\x{eaa}\x{eab}\x{ead}-\x{eb0}\x{eb2}\x{eb3}\x{ebd}\x{ec0}-\x{ec4}\x{ec6}\x{ed0}-\x{ed9}\x{edc}-\x{edf}\x{f00}-\x{f17}\x{f1a}-\x{f34}\x{f36}\x{f38}\x{f3e}-\x{f47}\x{f49}-\x{f6c}\x{f7f}\x{f85}\x{f88}-\x{f8c}\x{fbe}-\x{fc5}\x{fc7}-\x{fcc}\x{fce}-\x{fda}\x{1000}-\x{102c}\x{1031}\x{1038}\x{103b}\x{103c}\x{103f}-\x{1057}\x{105a}-\x{105d}\x{1061}-\x{1070}\x{1075}-\x{1081}\x{1083}\x{1084}\x{1087}-\x{108c}\x{108e}-\x{109c}\x{109e}-\x{10c5}\x{10c7}\x{10cd}\x{10d0}-\x{1248}\x{124a}-\x{124d}\x{1250}-\x{1256}\x{1258}\x{125a}-\x{125d}\x{1260}-\x{1288}\x{128a}-\x{128d}\x{1290}-\x{12b0}\x{12b2}-\x{12b5}\x{12b8}-\x{12be}\x{12c0}\x{12c2}-\x{12c5}\x{12c8}-\x{12d6}\x{12d8}-\x{1310}\x{1312}-\x{1315}\x{1318}-\x{135a}\x{1360}-\x{137c}\x{1380}-\x{138f}\x{13a0}-\x{13f5}\x{13f8}-\x{13fd}\x{1401}-\x{167f}\x{1681}-\x{169a}\x{16a0}-\x{16f8}\x{1700}-\x{170c}\x{170e}-\x{1711}\x{1720}-\x{1731}\x{1735}\x{1736}\x{1740}-\x{1751}\x{1760}-\x{176c}\x{176e}-\x{1770}\x{1780}-\x{17b3}\x{17b6}\x{17be}-\x{17c5}\x{17c7}\x{17c8}\x{17d4}-\x{17da}\x{17dc}\x{17e0}-\x{17e9}\x{1810}-\x{1819}\x{1820}-\x{1877}\x{1880}-\x{18a8}\x{18aa}\x{18b0}-\x{18f5}\x{1900}-\x{191e}\x{1923}-\x{1926}\x{1929}-\x{192b}\x{1930}\x{1931}\x{1933}-\x{1938}\x{1946}-\x{196d}\x{1970}-\x{1974}\x{1980}-\x{19ab}\x{19b0}-\x{19c9}\x{19d0}-\x{19da}\x{1a00}-\x{1a16}\x{1a19}\x{1a1a}\x{1a1e}-\x{1a55}\x{1a57}\x{1a61}\x{1a63}\x{1a64}\x{1a6d}-\x{1a72}\x{1a80}-\x{1a89}\x{1a90}-\x{1a99}\x{1aa0}-\x{1aad}\x{1b04}-\x{1b33}\x{1b35}\x{1b3b}\x{1b3d}-\x{1b41}\x{1b43}-\x{1b4b}\x{1b50}-\x{1b6a}\x{1b74}-\x{1b7c}\x{1b82}-\x{1ba1}\x{1ba6}\x{1ba7}\x{1baa}\x{1bae}-\x{1be5}\x{1be7}\x{1bea}-\x{1bec}\x{1bee}\x{1bf2}\x{1bf3}\x{1bfc}-\x{1c2b}\x{1c34}\x{1c35}\x{1c3b}-\x{1c49}\x{1c4d}-\x{1c7f}\x{1cc0}-\x{1cc7}\x{1cd3}\x{1ce1}\x{1ce9}-\x{1cec}\x{1cee}-\x{1cf3}\x{1cf5}\x{1cf6}\x{1d00}-\x{1dbf}\x{1e00}-\x{1f15}\x{1f18}-\x{1f1d}\x{1f20}-\x{1f45}\x{1f48}-\x{1f4d}\x{1f50}-\x{1f57}\x{1f59}\x{1f5b}\x{1f5d}\x{1f5f}-\x{1f7d}\x{1f80}-\x{1fb4}\x{1fb6}-\x{1fbc}\x{1fbe}\x{1fc2}-\x{1fc4}\x{1fc6}-\x{1fcc}\x{1fd0}-\x{1fd3}\x{1fd6}-\x{1fdb}\x{1fe0}-\x{1fec}\x{1ff2}-\x{1ff4}\x{1ff6}-\x{1ffc}\x{200e}\x{2071}\x{207f}\x{2090}-\x{209c}\x{2102}\x{2107}\x{210a}-\x{2113}\x{2115}\x{2119}-\x{211d}\x{2124}\x{2126}\x{2128}\x{212a}-\x{212d}\x{212f}-\x{2139}\x{213c}-\x{213f}\x{2145}-\x{2149}\x{214e}\x{214f}\x{2160}-\x{2188}\x{2336}-\x{237a}\x{2395}\x{249c}-\x{24e9}\x{26ac}\x{2800}-\x{28ff}\x{2c00}-\x{2c2e}\x{2c30}-\x{2c5e}\x{2c60}-\x{2ce4}\x{2ceb}-\x{2cee}\x{2cf2}\x{2cf3}\x{2d00}-\x{2d25}\x{2d27}\x{2d2d}\x{2d30}-\x{2d67}\x{2d6f}\x{2d70}\x{2d80}-\x{2d96}\x{2da0}-\x{2da6}\x{2da8}-\x{2dae}\x{2db0}-\x{2db6}\x{2db8}-\x{2dbe}\x{2dc0}-\x{2dc6}\x{2dc8}-\x{2dce}\x{2dd0}-\x{2dd6}\x{2dd8}-\x{2dde}\x{3005}-\x{3007}\x{3021}-\x{3029}\x{302e}\x{302f}\x{3031}-\x{3035}\x{3038}-\x{303c}\x{3041}-\x{3096}\x{309d}-\x{309f}\x{30a1}-\x{30fa}\x{30fc}-\x{30ff}\x{3105}-\x{312d}\x{3131}-\x{318e}\x{3190}-\x{31ba}\x{31f0}-\x{321c}\x{3220}-\x{324f}\x{3260}-\x{327b}\x{327f}-\x{32b0}\x{32c0}-\x{32cb}\x{32d0}-\x{32fe}\x{3300}-\x{3376}\x{337b}-\x{33dd}\x{33e0}-\x{33fe}\x{3400}-\x{4db5}\x{4e00}-\x{9fd5}\x{a000}-\x{a48c}\x{a4d0}-\x{a60c}\x{a610}-\x{a62b}\x{a640}-\x{a66e}\x{a680}-\x{a69d}\x{a6a0}-\x{a6ef}\x{a6f2}-\x{a6f7}\x{a722}-\x{a787}\x{a789}-\x{a7ad}\x{a7b0}-\x{a7b7}\x{a7f7}-\x{a801}\x{a803}-\x{a805}\x{a807}-\x{a80a}\x{a80c}-\x{a824}\x{a827}\x{a830}-\x{a837}\x{a840}-\x{a873}\x{a880}-\x{a8c3}\x{a8ce}-\x{a8d9}\x{a8f2}-\x{a8fd}\x{a900}-\x{a925}\x{a92e}-\x{a946}\x{a952}\x{a953}\x{a95f}-\x{a97c}\x{a983}-\x{a9b2}\x{a9b4}\x{a9b5}\x{a9ba}\x{a9bb}\x{a9bd}-\x{a9cd}\x{a9cf}-\x{a9d9}\x{a9de}-\x{a9e4}\x{a9e6}-\x{a9fe}\x{aa00}-\x{aa28}\x{aa2f}\x{aa30}\x{aa33}\x{aa34}\x{aa40}-\x{aa42}\x{aa44}-\x{aa4b}\x{aa4d}\x{aa50}-\x{aa59}\x{aa5c}-\x{aa7b}\x{aa7d}-\x{aaaf}\x{aab1}\x{aab5}\x{aab6}\x{aab9}-\x{aabd}\x{aac0}\x{aac2}\x{aadb}-\x{aaeb}\x{aaee}-\x{aaf5}\x{ab01}-\x{ab06}\x{ab09}-\x{ab0e}\x{ab11}-\x{ab16}\x{ab20}-\x{ab26}\x{ab28}-\x{ab2e}\x{ab30}-\x{ab65}\x{ab70}-\x{abe4}\x{abe6}\x{abe7}\x{abe9}-\x{abec}\x{abf0}-\x{abf9}\x{ac00}-\x{d7a3}\x{d7b0}-\x{d7c6}\x{d7cb}-\x{d7fb}\x{e000}-\x{fa6d}\x{fa70}-\x{fad9}\x{fb00}-\x{fb06}\x{fb13}-\x{fb17}\x{ff21}-\x{ff3a}\x{ff41}-\x{ff5a}\x{ff66}-\x{ffbe}\x{ffc2}-\x{ffc7}\x{ffca}-\x{ffcf}\x{ffd2}-\x{ffd7}\x{ffda}-\x{ffdc}\x{10000}-\x{1000b}\x{1000d}-\x{10026}\x{10028}-\x{1003a}\x{1003c}\x{1003d}\x{1003f}-\x{1004d}\x{10050}-\x{1005d}\x{10080}-\x{100fa}\x{10100}\x{10102}\x{10107}-\x{10133}\x{10137}-\x{1013f}\x{101d0}-\x{101fc}\x{10280}-\x{1029c}\x{102a0}-\x{102d0}\x{10300}-\x{10323}\x{10330}-\x{1034a}\x{10350}-\x{10375}\x{10380}-\x{1039d}\x{1039f}-\x{103c3}\x{103c8}-\x{103d5}\x{10400}-\x{1049d}\x{104a0}-\x{104a9}\x{10500}-\x{10527}\x{10530}-\x{10563}\x{1056f}\x{10600}-\x{10736}\x{10740}-\x{10755}\x{10760}-\x{10767}\x{11000}\x{11002}-\x{11037}\x{11047}-\x{1104d}\x{11066}-\x{1106f}\x{11082}-\x{110b2}\x{110b7}\x{110b8}\x{110bb}-\x{110c1}\x{110d0}-\x{110e8}\x{110f0}-\x{110f9}\x{11103}-\x{11126}\x{1112c}\x{11136}-\x{11143}\x{11150}-\x{11172}\x{11174}-\x{11176}\x{11182}-\x{111b5}\x{111bf}-\x{111c9}\x{111cd}\x{111d0}-\x{111df}\x{111e1}-\x{111f4}\x{11200}-\x{11211}\x{11213}-\x{1122e}\x{11232}\x{11233}\x{11235}\x{11238}-\x{1123d}\x{11280}-\x{11286}\x{11288}\x{1128a}-\x{1128d}\x{1128f}-\x{1129d}\x{1129f}-\x{112a9}\x{112b0}-\x{112de}\x{112e0}-\x{112e2}\x{112f0}-\x{112f9}\x{11302}\x{11303}\x{11305}-\x{1130c}\x{1130f}\x{11310}\x{11313}-\x{11328}\x{1132a}-\x{11330}\x{11332}\x{11333}\x{11335}-\x{11339}\x{1133d}-\x{1133f}\x{11341}-\x{11344}\x{11347}\x{11348}\x{1134b}-\x{1134d}\x{11350}\x{11357}\x{1135d}-\x{11363}\x{11480}-\x{114b2}\x{114b9}\x{114bb}-\x{114be}\x{114c1}\x{114c4}-\x{114c7}\x{114d0}-\x{114d9}\x{11580}-\x{115b1}\x{115b8}-\x{115bb}\x{115be}\x{115c1}-\x{115db}\x{11600}-\x{11632}\x{1163b}\x{1163c}\x{1163e}\x{11641}-\x{11644}\x{11650}-\x{11659}\x{11680}-\x{116aa}\x{116ac}\x{116ae}\x{116af}\x{116b6}\x{116c0}-\x{116c9}\x{11700}-\x{11719}\x{11720}\x{11721}\x{11726}\x{11730}-\x{1173f}\x{118a0}-\x{118f2}\x{118ff}\x{11ac0}-\x{11af8}\x{12000}-\x{12399}\x{12400}-\x{1246e}\x{12470}-\x{12474}\x{12480}-\x{12543}\x{13000}-\x{1342e}\x{14400}-\x{14646}\x{16800}-\x{16a38}\x{16a40}-\x{16a5e}\x{16a60}-\x{16a69}\x{16a6e}\x{16a6f}\x{16ad0}-\x{16aed}\x{16af5}\x{16b00}-\x{16b2f}\x{16b37}-\x{16b45}\x{16b50}-\x{16b59}\x{16b5b}-\x{16b61}\x{16b63}-\x{16b77}\x{16b7d}-\x{16b8f}\x{16f00}-\x{16f44}\x{16f50}-\x{16f7e}\x{16f93}-\x{16f9f}\x{1b000}\x{1b001}\x{1bc00}-\x{1bc6a}\x{1bc70}-\x{1bc7c}\x{1bc80}-\x{1bc88}\x{1bc90}-\x{1bc99}\x{1bc9c}\x{1bc9f}\x{1d000}-\x{1d0f5}\x{1d100}-\x{1d126}\x{1d129}-\x{1d166}\x{1d16a}-\x{1d172}\x{1d183}\x{1d184}\x{1d18c}-\x{1d1a9}\x{1d1ae}-\x{1d1e8}\x{1d360}-\x{1d371}\x{1d400}-\x{1d454}\x{1d456}-\x{1d49c}\x{1d49e}\x{1d49f}\x{1d4a2}\x{1d4a5}\x{1d4a6}\x{1d4a9}-\x{1d4ac}\x{1d4ae}-\x{1d4b9}\x{1d4bb}\x{1d4bd}-\x{1d4c3}\x{1d4c5}-\x{1d505}\x{1d507}-\x{1d50a}\x{1d50d}-\x{1d514}\x{1d516}-\x{1d51c}\x{1d51e}-\x{1d539}\x{1d53b}-\x{1d53e}\x{1d540}-\x{1d544}\x{1d546}\x{1d54a}-\x{1d550}\x{1d552}-\x{1d6a5}\x{1d6a8}-\x{1d6da}\x{1d6dc}-\x{1d714}\x{1d716}-\x{1d74e}\x{1d750}-\x{1d788}\x{1d78a}-\x{1d7c2}\x{1d7c4}-\x{1d7cb}\x{1d800}-\x{1d9ff}\x{1da37}-\x{1da3a}\x{1da6d}-\x{1da74}\x{1da76}-\x{1da83}\x{1da85}-\x{1da8b}\x{1f110}-\x{1f12e}\x{1f130}-\x{1f169}\x{1f170}-\x{1f19a}\x{1f1e6}-\x{1f202}\x{1f210}-\x{1f23a}\x{1f240}-\x{1f248}\x{1f250}\x{1f251}\x{20000}-\x{2a6d6}\x{2a700}-\x{2b734}\x{2b740}-\x{2b81d}\x{2b820}-\x{2cea1}\x{2f800}-\x{2fa1d}\x{f0000}-\x{ffffd}\x{100000}-\x{10fffd}])|([\x{590}\x{5be}\x{5c0}\x{5c3}\x{5c6}\x{5c8}-\x{5ff}\x{7c0}-\x{7ea}\x{7f4}\x{7f5}\x{7fa}-\x{815}\x{81a}\x{824}\x{828}\x{82e}-\x{858}\x{85c}-\x{89f}\x{200f}\x{fb1d}\x{fb1f}-\x{fb28}\x{fb2a}-\x{fb4f}\x{10800}-\x{1091e}\x{10920}-\x{10a00}\x{10a04}\x{10a07}-\x{10a0b}\x{10a10}-\x{10a37}\x{10a3b}-\x{10a3e}\x{10a40}-\x{10ae4}\x{10ae7}-\x{10b38}\x{10b40}-\x{10e5f}\x{10e7f}-\x{10fff}\x{1e800}-\x{1e8cf}\x{1e8d7}-\x{1edff}\x{1ef00}-\x{1efff}\x{608}\x{60b}\x{60d}\x{61b}-\x{64a}\x{66d}-\x{66f}\x{671}-\x{6d5}\x{6e5}\x{6e6}\x{6ee}\x{6ef}\x{6fa}-\x{710}\x{712}-\x{72f}\x{74b}-\x{7a5}\x{7b1}-\x{7bf}\x{8a0}-\x{8e2}\x{fb50}-\x{fd3d}\x{fd40}-\x{fdcf}\x{fdf0}-\x{fdfc}\x{fdfe}\x{fdff}\x{fe70}-\x{fefe}\x{1ee00}-\x{1eeef}\x{1eef2}-\x{1eeff}]))/u';
209 // @codeCoverageIgnoreEnd
210
217 static function factory( $code ) {
219
220 if ( isset( $wgDummyLanguageCodes[$code] ) ) {
221 $code = $wgDummyLanguageCodes[$code];
222 }
223
224 // get the language object to process
225 $langObj = self::$mLangObjCache[$code] ?? self::newFromCode( $code );
226
227 // merge the language object in to get it up front in the cache
228 self::$mLangObjCache = array_merge( [ $code => $langObj ], self::$mLangObjCache );
229 // get rid of the oldest ones in case we have an overflow
230 self::$mLangObjCache = array_slice( self::$mLangObjCache, 0, $wgLangObjCacheSize, true );
231
232 return $langObj;
233 }
234
242 protected static function newFromCode( $code, $fallback = false ) {
243 $langNameUtils = MediaWikiServices::getInstance()->getLanguageNameUtils();
244 if ( !$langNameUtils->isValidCode( $code ) ) {
245 throw new MWException( "Invalid language code \"$code\"" );
246 }
247
248 if ( !$langNameUtils->isValidBuiltInCode( $code ) ) {
249 // It's not possible to customise this code with class files, so
250 // just return a Language object. This is to support uselang= hacks.
251 $lang = new Language;
252 $lang->mCode = $code;
253 return $lang;
254 }
255
256 // Check if there is a language class for the code
257 $class = self::classFromCode( $code, $fallback );
258 // LanguageCode does not inherit Language
259 if ( class_exists( $class ) && is_a( $class, 'Language', true ) ) {
260 $lang = new $class;
261 return $lang;
262 }
263
264 // Keep trying the fallback list until we find an existing class
265 $fallbacks = self::getFallbacksFor( $code );
266 foreach ( $fallbacks as $fallbackCode ) {
267 if ( !$langNameUtils->isValidBuiltInCode( $fallbackCode ) ) {
268 throw new MWException( "Invalid fallback '$fallbackCode' in fallback sequence for '$code'" );
269 }
270
271 $class = self::classFromCode( $fallbackCode );
272 if ( class_exists( $class ) ) {
273 $lang = new $class;
274 $lang->mCode = $code;
275 return $lang;
276 }
277 }
278
279 throw new MWException( "Invalid fallback sequence for language '$code'" );
280 }
281
287 public static function clearCaches() {
288 if ( !defined( 'MW_PHPUNIT_TEST' ) && !defined( 'MEDIAWIKI_INSTALL' ) ) {
289 throw new MWException( __METHOD__ . ' must not be used outside tests/installer' );
290 }
291 if ( defined( 'MW_PHPUNIT_TEST' ) ) {
292 MediaWikiServices::getInstance()->resetServiceForTesting( 'LocalisationCache' );
293 MediaWikiServices::getInstance()->resetServiceForTesting( 'LanguageNameUtils' );
294 }
295 self::$mLangObjCache = [];
296 self::$fallbackLanguageCache = [];
297 self::$grammarTransformations = null;
298 }
299
309 public static function isSupportedLanguage( $code ) {
310 return MediaWikiServices::getInstance()->getLanguageNameUtils()
311 ->isSupportedLanguage( $code );
312 }
313
329 public static function isWellFormedLanguageTag( $code, $lenient = false ) {
330 $alpha = '[a-z]';
331 $digit = '[0-9]';
332 $alphanum = '[a-z0-9]';
333 $x = 'x'; # private use singleton
334 $singleton = '[a-wy-z]'; # other singleton
335 $s = $lenient ? '[-_]' : '-';
336
337 $language = "$alpha{2,8}|$alpha{2,3}$s$alpha{3}";
338 $script = "$alpha{4}"; # ISO 15924
339 $region = "(?:$alpha{2}|$digit{3})"; # ISO 3166-1 alpha-2 or UN M.49
340 $variant = "(?:$alphanum{5,8}|$digit$alphanum{3})";
341 $extension = "$singleton(?:$s$alphanum{2,8})+";
342 $privateUse = "$x(?:$s$alphanum{1,8})+";
343
344 # Define certain grandfathered codes, since otherwise the regex is pretty useless.
345 # Since these are limited, this is safe even later changes to the registry --
346 # the only oddity is that it might change the type of the tag, and thus
347 # the results from the capturing groups.
348 # https://www.iana.org/assignments/language-subtag-registry
349
350 $grandfathered = "en{$s}GB{$s}oed"
351 . "|i{$s}(?:ami|bnn|default|enochian|hak|klingon|lux|mingo|navajo|pwn|tao|tay|tsu)"
352 . "|no{$s}(?:bok|nyn)"
353 . "|sgn{$s}(?:BE{$s}(?:fr|nl)|CH{$s}de)"
354 . "|zh{$s}min{$s}nan";
355
356 $variantList = "$variant(?:$s$variant)*";
357 $extensionList = "$extension(?:$s$extension)*";
358
359 $langtag = "(?:($language)"
360 . "(?:$s$script)?"
361 . "(?:$s$region)?"
362 . "(?:$s$variantList)?"
363 . "(?:$s$extensionList)?"
364 . "(?:$s$privateUse)?)";
365
366 # The final breakdown, with capturing groups for each of these components
367 # The variants, extensions, grandfathered, and private-use may have interior '-'
368
369 $root = "^(?:$langtag|$privateUse|$grandfathered)$";
370
371 return (bool)preg_match( "/$root/", strtolower( $code ) );
372 }
373
385 public static function isValidCode( $code ) {
386 return MediaWikiServices::getInstance()->getLanguageNameUtils()->isValidCode( $code );
387 }
388
400 public static function isValidBuiltInCode( $code ) {
401 return MediaWikiServices::getInstance()->getLanguageNameUtils()
402 ->isValidBuiltInCode( $code );
403 }
404
415 public static function isKnownLanguageTag( $tag ) {
416 return MediaWikiServices::getInstance()->getLanguageNameUtils()
417 ->isKnownLanguageTag( $tag );
418 }
419
426 public static function getLocalisationCache() {
427 return MediaWikiServices::getInstance()->getLocalisationCache();
428 }
429
430 function __construct() {
431 $this->mConverter = new FakeConverter( $this );
432 // Set the code to the name of the descendant
433 if ( static::class === 'Language' ) {
434 $this->mCode = 'en';
435 } else {
436 $this->mCode = str_replace( '_', '-', strtolower( substr( static::class, 8 ) ) );
437 }
438 $services = MediaWikiServices::getInstance();
439 $this->localisationCache = $services->getLocalisationCache();
440 $this->langNameUtils = $services->getLanguageNameUtils();
441 }
442
447 function __destruct() {
448 foreach ( $this as $name => $value ) {
449 unset( $this->$name );
450 }
451 }
452
457 function initContLang() {
458 }
459
464 public function getFallbackLanguages() {
465 return self::getFallbacksFor( $this->mCode );
466 }
467
472 public function getBookstoreList() {
473 return $this->localisationCache->getItem( $this->mCode, 'bookstoreList' );
474 }
475
482 public function getNamespaces() {
483 if ( is_null( $this->namespaceNames ) ) {
485
486 $validNamespaces = MediaWikiServices::getInstance()->getNamespaceInfo()->
487 getCanonicalNamespaces();
488
489 $this->namespaceNames = $wgExtraNamespaces +
490 $this->localisationCache->getItem( $this->mCode, 'namespaceNames' );
491 $this->namespaceNames += $validNamespaces;
492
493 $this->namespaceNames[NS_PROJECT] = $wgMetaNamespace;
494 if ( $wgMetaNamespaceTalk ) {
495 $this->namespaceNames[NS_PROJECT_TALK] = $wgMetaNamespaceTalk;
496 } else {
497 $talk = $this->namespaceNames[NS_PROJECT_TALK];
498 $this->namespaceNames[NS_PROJECT_TALK] =
499 $this->fixVariableInNamespace( $talk );
500 }
501
502 # Sometimes a language will be localised but not actually exist on this wiki.
503 foreach ( $this->namespaceNames as $key => $text ) {
504 if ( !isset( $validNamespaces[$key] ) ) {
505 unset( $this->namespaceNames[$key] );
506 }
507 }
508
509 # The above mixing may leave namespaces out of canonical order.
510 # Re-order by namespace ID number...
511 ksort( $this->namespaceNames );
512
513 Hooks::run( 'LanguageGetNamespaces', [ &$this->namespaceNames ] );
514 }
515
517 }
518
523 public function setNamespaces( array $namespaces ) {
524 $this->namespaceNames = $namespaces;
525 $this->mNamespaceIds = null;
526 }
527
531 public function resetNamespaces() {
532 $this->namespaceNames = null;
533 $this->mNamespaceIds = null;
534 $this->namespaceAliases = null;
535 }
536
543 public function getFormattedNamespaces() {
544 $ns = $this->getNamespaces();
545 foreach ( $ns as $k => $v ) {
546 $ns[$k] = strtr( $v, '_', ' ' );
547 }
548 return $ns;
549 }
550
562 public function getNsText( $index ) {
563 $ns = $this->getNamespaces();
564 return $ns[$index] ?? false;
565 }
566
580 public function getFormattedNsText( $index ) {
581 $ns = $this->getNsText( $index );
582 return strtr( $ns, '_', ' ' );
583 }
584
593 public function getGenderNsText( $index, $gender ) {
595
597 (array)$this->localisationCache->getItem( $this->mCode, 'namespaceGenderAliases' );
598
599 return $ns[$index][$gender] ?? $this->getNsText( $index );
600 }
601
608 public function needsGenderDistinction() {
610 if ( count( $wgExtraGenderNamespaces ) > 0 ) {
611 // $wgExtraGenderNamespaces overrides everything
612 return true;
613 } elseif ( isset( $wgExtraNamespaces[NS_USER] ) && isset( $wgExtraNamespaces[NS_USER_TALK] ) ) {
615 // $wgExtraNamespaces overrides any gender aliases specified in i18n files
616 return false;
617 } else {
618 // Check what is in i18n files
619 $aliases = $this->localisationCache->getItem( $this->mCode, 'namespaceGenderAliases' );
620 return count( $aliases ) > 0;
621 }
622 }
623
632 function getLocalNsIndex( $text ) {
633 $lctext = $this->lc( $text );
634 $ids = $this->getNamespaceIds();
635 return $ids[$lctext] ?? false;
636 }
637
641 public function getNamespaceAliases() {
642 if ( is_null( $this->namespaceAliases ) ) {
643 $aliases = $this->localisationCache->getItem( $this->mCode, 'namespaceAliases' );
644 if ( !$aliases ) {
645 $aliases = [];
646 } else {
647 foreach ( $aliases as $name => $index ) {
648 if ( $index === NS_PROJECT_TALK ) {
649 unset( $aliases[$name] );
650 $name = $this->fixVariableInNamespace( $name );
651 $aliases[$name] = $index;
652 }
653 }
654 }
655
657 $genders = $wgExtraGenderNamespaces + (array)$this->localisationCache
658 ->getItem( $this->mCode, 'namespaceGenderAliases' );
659 foreach ( $genders as $index => $forms ) {
660 foreach ( $forms as $alias ) {
661 $aliases[$alias] = $index;
662 }
663 }
664
665 # Also add converted namespace names as aliases, to avoid confusion.
666 $convertedNames = [];
667 foreach ( $this->getVariants() as $variant ) {
668 if ( $variant === $this->mCode ) {
669 continue;
670 }
671 foreach ( $this->getNamespaces() as $ns => $_ ) {
672 $convertedNames[$this->getConverter()->convertNamespace( $ns, $variant )] = $ns;
673 }
674 }
675
676 $this->namespaceAliases = $aliases + $convertedNames;
677
678 # Filter out aliases to namespaces that don't exist, e.g. from extensions
679 # that aren't loaded here but are included in the l10n cache.
680 # (array_intersect preserves keys from its first argument)
681 $this->namespaceAliases = array_intersect(
682 $this->namespaceAliases,
683 array_keys( $this->getNamespaces() )
684 );
685 }
686
688 }
689
693 public function getNamespaceIds() {
694 if ( is_null( $this->mNamespaceIds ) ) {
695 global $wgNamespaceAliases;
696 # Put namespace names and aliases into a hashtable.
697 # If this is too slow, then we should arrange it so that it is done
698 # before caching. The catch is that at pre-cache time, the above
699 # class-specific fixup hasn't been done.
700 $this->mNamespaceIds = [];
701 foreach ( $this->getNamespaces() as $index => $name ) {
702 $this->mNamespaceIds[$this->lc( $name )] = $index;
703 }
704 foreach ( $this->getNamespaceAliases() as $name => $index ) {
705 $this->mNamespaceIds[$this->lc( $name )] = $index;
706 }
707 if ( $wgNamespaceAliases ) {
708 foreach ( $wgNamespaceAliases as $name => $index ) {
709 $this->mNamespaceIds[$this->lc( $name )] = $index;
710 }
711 }
712 }
713 return $this->mNamespaceIds;
714 }
715
723 public function getNsIndex( $text ) {
724 $lctext = $this->lc( $text );
725 $ns = MediaWikiServices::getInstance()->getNamespaceInfo()->
726 getCanonicalIndex( $lctext );
727 if ( $ns !== null ) {
728 return $ns;
729 }
730 $ids = $this->getNamespaceIds();
731 return $ids[$lctext] ?? false;
732 }
733
741 public function getVariantname( $code, $usemsg = true ) {
742 $msg = "variantname-$code";
743 if ( $usemsg && wfMessage( $msg )->exists() ) {
744 return $this->getMessageFromDB( $msg );
745 }
746 $name = $this->langNameUtils->getLanguageName( $code );
747 if ( $name ) {
748 return $name; # if it's defined as a language name, show that
749 } else {
750 # otherwise, output the language code
751 return $code;
752 }
753 }
754
758 public function getDatePreferences() {
759 return $this->localisationCache->getItem( $this->mCode, 'datePreferences' );
760 }
761
765 function getDateFormats() {
766 return $this->localisationCache->getItem( $this->mCode, 'dateFormats' );
767 }
768
772 public function getDefaultDateFormat() {
773 $df = $this->localisationCache->getItem( $this->mCode, 'defaultDateFormat' );
774 if ( $df === 'dmy or mdy' ) {
775 global $wgAmericanDates;
776 return $wgAmericanDates ? 'mdy' : 'dmy';
777 } else {
778 return $df;
779 }
780 }
781
785 public function getDatePreferenceMigrationMap() {
786 return $this->localisationCache->getItem( $this->mCode, 'datePreferenceMigrationMap' );
787 }
788
792 public function getExtraUserToggles() {
793 wfDeprecated( __METHOD__, '1.34' );
794 return [];
795 }
796
801 function getUserToggle( $tog ) {
802 return $this->getMessageFromDB( "tog-$tog" );
803 }
804
818 public static function fetchLanguageNames( $inLanguage = self::AS_AUTONYMS, $include = 'mw' ) {
819 return MediaWikiServices::getInstance()->getLanguageNameUtils()
820 ->getLanguageNames( $inLanguage, $include );
821 }
822
832 public static function fetchLanguageName(
833 $code,
834 $inLanguage = self::AS_AUTONYMS,
835 $include = self::ALL
836 ) {
837 return MediaWikiServices::getInstance()->getLanguageNameUtils()
838 ->getLanguageName( $code, $inLanguage, $include );
839 }
840
847 public function getMessageFromDB( $msg ) {
848 return $this->msg( $msg )->text();
849 }
850
857 protected function msg( $msg ) {
858 return wfMessage( $msg )->inLanguage( $this );
859 }
860
865 public function getMonthName( $key ) {
866 return $this->getMessageFromDB( self::$mMonthMsgs[$key - 1] );
867 }
868
872 public function getMonthNamesArray() {
873 $monthNames = [ '' ];
874 for ( $i = 1; $i < 13; $i++ ) {
875 $monthNames[] = $this->getMonthName( $i );
876 }
877 return $monthNames;
878 }
879
884 public function getMonthNameGen( $key ) {
885 return $this->getMessageFromDB( self::$mMonthGenMsgs[$key - 1] );
886 }
887
892 public function getMonthAbbreviation( $key ) {
893 return $this->getMessageFromDB( self::$mMonthAbbrevMsgs[$key - 1] );
894 }
895
899 public function getMonthAbbreviationsArray() {
900 $monthNames = [ '' ];
901 for ( $i = 1; $i < 13; $i++ ) {
902 $monthNames[] = $this->getMonthAbbreviation( $i );
903 }
904 return $monthNames;
905 }
906
911 public function getWeekdayName( $key ) {
912 return $this->getMessageFromDB( self::$mWeekdayMsgs[$key - 1] );
913 }
914
919 function getWeekdayAbbreviation( $key ) {
920 return $this->getMessageFromDB( self::$mWeekdayAbbrevMsgs[$key - 1] );
921 }
922
927 function getIranianCalendarMonthName( $key ) {
928 return $this->getMessageFromDB( self::$mIranianCalendarMonthMsgs[$key - 1] );
929 }
930
935 function getHebrewCalendarMonthName( $key ) {
936 return $this->getMessageFromDB( self::$mHebrewCalendarMonthMsgs[$key - 1] );
937 }
938
943 function getHebrewCalendarMonthNameGen( $key ) {
944 return $this->getMessageFromDB( self::$mHebrewCalendarMonthGenMsgs[$key - 1] );
945 }
946
951 function getHijriCalendarMonthName( $key ) {
952 return $this->getMessageFromDB( self::$mHijriCalendarMonthMsgs[$key - 1] );
953 }
954
963 private static function dateTimeObjFormat( &$dateTimeObj, $ts, $zone, $code ) {
964 if ( !$dateTimeObj ) {
965 $dateTimeObj = DateTime::createFromFormat(
966 'YmdHis', $ts, $zone ?: new DateTimeZone( 'UTC' )
967 );
968 }
969 return $dateTimeObj->format( $code );
970 }
971
1041 public function sprintfDate( $format, $ts, DateTimeZone $zone = null, &$ttl = 'unused' ) {
1042 $s = '';
1043 $raw = false;
1044 $roman = false;
1045 $hebrewNum = false;
1046 $dateTimeObj = false;
1047 $rawToggle = false;
1048 $iranian = false;
1049 $hebrew = false;
1050 $hijri = false;
1051 $thai = false;
1052 $minguo = false;
1053 $tenno = false;
1054
1055 $usedSecond = false;
1056 $usedMinute = false;
1057 $usedHour = false;
1058 $usedAMPM = false;
1059 $usedDay = false;
1060 $usedWeek = false;
1061 $usedMonth = false;
1062 $usedYear = false;
1063 $usedISOYear = false;
1064 $usedIsLeapYear = false;
1065
1066 $usedHebrewMonth = false;
1067 $usedIranianMonth = false;
1068 $usedHijriMonth = false;
1069 $usedHebrewYear = false;
1070 $usedIranianYear = false;
1071 $usedHijriYear = false;
1072 $usedTennoYear = false;
1073
1074 if ( strlen( $ts ) !== 14 ) {
1075 throw new MWException( __METHOD__ . ": The timestamp $ts should have 14 characters" );
1076 }
1077
1078 if ( !ctype_digit( $ts ) ) {
1079 throw new MWException( __METHOD__ . ": The timestamp $ts should be a number" );
1080 }
1081
1082 $formatLength = strlen( $format );
1083 for ( $p = 0; $p < $formatLength; $p++ ) {
1084 $num = false;
1085 $code = $format[$p];
1086 if ( $code == 'x' && $p < $formatLength - 1 ) {
1087 $code .= $format[++$p];
1088 }
1089
1090 if ( ( $code === 'xi'
1091 || $code === 'xj'
1092 || $code === 'xk'
1093 || $code === 'xm'
1094 || $code === 'xo'
1095 || $code === 'xt' )
1096 && $p < $formatLength - 1 ) {
1097 $code .= $format[++$p];
1098 }
1099
1100 switch ( $code ) {
1101 case 'xx':
1102 $s .= 'x';
1103 break;
1104 case 'xn':
1105 $raw = true;
1106 break;
1107 case 'xN':
1108 $rawToggle = !$rawToggle;
1109 break;
1110 case 'xr':
1111 $roman = true;
1112 break;
1113 case 'xh':
1114 $hebrewNum = true;
1115 break;
1116 case 'xg':
1117 $usedMonth = true;
1118 $s .= $this->getMonthNameGen( substr( $ts, 4, 2 ) );
1119 break;
1120 case 'xjx':
1121 $usedHebrewMonth = true;
1122 if ( !$hebrew ) {
1123 $hebrew = self::tsToHebrew( $ts );
1124 }
1125 $s .= $this->getHebrewCalendarMonthNameGen( $hebrew[1] );
1126 break;
1127 case 'd':
1128 $usedDay = true;
1129 $num = substr( $ts, 6, 2 );
1130 break;
1131 case 'D':
1132 $usedDay = true;
1133 $s .= $this->getWeekdayAbbreviation(
1134 self::dateTimeObjFormat( $dateTimeObj, $ts, $zone, 'w' ) + 1
1135 );
1136 break;
1137 case 'j':
1138 $usedDay = true;
1139 $num = intval( substr( $ts, 6, 2 ) );
1140 break;
1141 case 'xij':
1142 $usedDay = true;
1143 if ( !$iranian ) {
1144 $iranian = self::tsToIranian( $ts );
1145 }
1146 $num = $iranian[2];
1147 break;
1148 case 'xmj':
1149 $usedDay = true;
1150 if ( !$hijri ) {
1151 $hijri = self::tsToHijri( $ts );
1152 }
1153 $num = $hijri[2];
1154 break;
1155 case 'xjj':
1156 $usedDay = true;
1157 if ( !$hebrew ) {
1158 $hebrew = self::tsToHebrew( $ts );
1159 }
1160 $num = $hebrew[2];
1161 break;
1162 case 'l':
1163 $usedDay = true;
1164 $s .= $this->getWeekdayName(
1165 self::dateTimeObjFormat( $dateTimeObj, $ts, $zone, 'w' ) + 1
1166 );
1167 break;
1168 case 'F':
1169 $usedMonth = true;
1170 $s .= $this->getMonthName( substr( $ts, 4, 2 ) );
1171 break;
1172 case 'xiF':
1173 $usedIranianMonth = true;
1174 if ( !$iranian ) {
1175 $iranian = self::tsToIranian( $ts );
1176 }
1177 $s .= $this->getIranianCalendarMonthName( $iranian[1] );
1178 break;
1179 case 'xmF':
1180 $usedHijriMonth = true;
1181 if ( !$hijri ) {
1182 $hijri = self::tsToHijri( $ts );
1183 }
1184 $s .= $this->getHijriCalendarMonthName( $hijri[1] );
1185 break;
1186 case 'xjF':
1187 $usedHebrewMonth = true;
1188 if ( !$hebrew ) {
1189 $hebrew = self::tsToHebrew( $ts );
1190 }
1191 $s .= $this->getHebrewCalendarMonthName( $hebrew[1] );
1192 break;
1193 case 'm':
1194 $usedMonth = true;
1195 $num = substr( $ts, 4, 2 );
1196 break;
1197 case 'M':
1198 $usedMonth = true;
1199 $s .= $this->getMonthAbbreviation( substr( $ts, 4, 2 ) );
1200 break;
1201 case 'n':
1202 $usedMonth = true;
1203 $num = intval( substr( $ts, 4, 2 ) );
1204 break;
1205 case 'xin':
1206 $usedIranianMonth = true;
1207 if ( !$iranian ) {
1208 $iranian = self::tsToIranian( $ts );
1209 }
1210 $num = $iranian[1];
1211 break;
1212 case 'xmn':
1213 $usedHijriMonth = true;
1214 if ( !$hijri ) {
1215 $hijri = self::tsToHijri( $ts );
1216 }
1217 $num = $hijri[1];
1218 break;
1219 case 'xjn':
1220 $usedHebrewMonth = true;
1221 if ( !$hebrew ) {
1222 $hebrew = self::tsToHebrew( $ts );
1223 }
1224 $num = $hebrew[1];
1225 break;
1226 case 'xjt':
1227 $usedHebrewMonth = true;
1228 if ( !$hebrew ) {
1229 $hebrew = self::tsToHebrew( $ts );
1230 }
1231 $num = $hebrew[3];
1232 break;
1233 case 'Y':
1234 $usedYear = true;
1235 $num = substr( $ts, 0, 4 );
1236 break;
1237 case 'xiY':
1238 $usedIranianYear = true;
1239 if ( !$iranian ) {
1240 $iranian = self::tsToIranian( $ts );
1241 }
1242 $num = $iranian[0];
1243 break;
1244 case 'xmY':
1245 $usedHijriYear = true;
1246 if ( !$hijri ) {
1247 $hijri = self::tsToHijri( $ts );
1248 }
1249 $num = $hijri[0];
1250 break;
1251 case 'xjY':
1252 $usedHebrewYear = true;
1253 if ( !$hebrew ) {
1254 $hebrew = self::tsToHebrew( $ts );
1255 }
1256 $num = $hebrew[0];
1257 break;
1258 case 'xkY':
1259 $usedYear = true;
1260 if ( !$thai ) {
1261 $thai = self::tsToYear( $ts, 'thai' );
1262 }
1263 $num = $thai[0];
1264 break;
1265 case 'xoY':
1266 $usedYear = true;
1267 if ( !$minguo ) {
1268 $minguo = self::tsToYear( $ts, 'minguo' );
1269 }
1270 $num = $minguo[0];
1271 break;
1272 case 'xtY':
1273 $usedTennoYear = true;
1274 if ( !$tenno ) {
1275 $tenno = self::tsToYear( $ts, 'tenno' );
1276 }
1277 $num = $tenno[0];
1278 break;
1279 case 'y':
1280 $usedYear = true;
1281 $num = substr( $ts, 2, 2 );
1282 break;
1283 case 'xiy':
1284 $usedIranianYear = true;
1285 if ( !$iranian ) {
1286 $iranian = self::tsToIranian( $ts );
1287 }
1288 $num = substr( $iranian[0], -2 );
1289 break;
1290 case 'xit':
1291 $usedIranianYear = true;
1292 if ( !$iranian ) {
1293 $iranian = self::tsToIranian( $ts );
1294 }
1295 $num = self::$IRANIAN_DAYS[$iranian[1] - 1];
1296 break;
1297 case 'xiz':
1298 $usedIranianYear = true;
1299 if ( !$iranian ) {
1300 $iranian = self::tsToIranian( $ts );
1301 }
1302 $num = $iranian[3];
1303 break;
1304 case 'a':
1305 $usedAMPM = true;
1306 $s .= intval( substr( $ts, 8, 2 ) ) < 12 ? 'am' : 'pm';
1307 break;
1308 case 'A':
1309 $usedAMPM = true;
1310 $s .= intval( substr( $ts, 8, 2 ) ) < 12 ? 'AM' : 'PM';
1311 break;
1312 case 'g':
1313 $usedHour = true;
1314 $h = substr( $ts, 8, 2 );
1315 $num = $h % 12 ?: 12;
1316 break;
1317 case 'G':
1318 $usedHour = true;
1319 $num = intval( substr( $ts, 8, 2 ) );
1320 break;
1321 case 'h':
1322 $usedHour = true;
1323 $h = substr( $ts, 8, 2 );
1324 $num = sprintf( '%02d', $h % 12 ?: 12 );
1325 break;
1326 case 'H':
1327 $usedHour = true;
1328 $num = substr( $ts, 8, 2 );
1329 break;
1330 case 'i':
1331 $usedMinute = true;
1332 $num = substr( $ts, 10, 2 );
1333 break;
1334 case 's':
1335 $usedSecond = true;
1336 $num = substr( $ts, 12, 2 );
1337 break;
1338 case 'c':
1339 case 'r':
1340 $usedSecond = true;
1341 // fall through
1342 case 'e':
1343 case 'O':
1344 case 'P':
1345 case 'T':
1346 $s .= self::dateTimeObjFormat( $dateTimeObj, $ts, $zone, $code );
1347 break;
1348 case 'w':
1349 case 'N':
1350 case 'z':
1351 $usedDay = true;
1352 $num = self::dateTimeObjFormat( $dateTimeObj, $ts, $zone, $code );
1353 break;
1354 case 'W':
1355 $usedWeek = true;
1356 $num = self::dateTimeObjFormat( $dateTimeObj, $ts, $zone, $code );
1357 break;
1358 case 't':
1359 $usedMonth = true;
1360 $num = self::dateTimeObjFormat( $dateTimeObj, $ts, $zone, $code );
1361 break;
1362 case 'L':
1363 $usedIsLeapYear = true;
1364 $num = self::dateTimeObjFormat( $dateTimeObj, $ts, $zone, $code );
1365 break;
1366 case 'o':
1367 $usedISOYear = true;
1368 $num = self::dateTimeObjFormat( $dateTimeObj, $ts, $zone, $code );
1369 break;
1370 case 'U':
1371 $usedSecond = true;
1372 // fall through
1373 case 'I':
1374 case 'Z':
1375 $num = self::dateTimeObjFormat( $dateTimeObj, $ts, $zone, $code );
1376 break;
1377 case '\\':
1378 # Backslash escaping
1379 if ( $p < $formatLength - 1 ) {
1380 $s .= $format[++$p];
1381 } else {
1382 $s .= '\\';
1383 }
1384 break;
1385 case '"':
1386 # Quoted literal
1387 if ( $p < $formatLength - 1 ) {
1388 $endQuote = strpos( $format, '"', $p + 1 );
1389 if ( $endQuote === false ) {
1390 # No terminating quote, assume literal "
1391 $s .= '"';
1392 } else {
1393 $s .= substr( $format, $p + 1, $endQuote - $p - 1 );
1394 $p = $endQuote;
1395 }
1396 } else {
1397 # Quote at end of string, assume literal "
1398 $s .= '"';
1399 }
1400 break;
1401 default:
1402 $s .= $format[$p];
1403 }
1404 if ( $num !== false ) {
1405 if ( $rawToggle || $raw ) {
1406 $s .= $num;
1407 $raw = false;
1408 } elseif ( $roman ) {
1409 $s .= self::romanNumeral( $num );
1410 $roman = false;
1411 } elseif ( $hebrewNum ) {
1412 $s .= self::hebrewNumeral( $num );
1413 $hebrewNum = false;
1414 } else {
1415 $s .= $this->formatNum( $num, true );
1416 }
1417 }
1418 }
1419
1420 if ( $ttl === 'unused' ) {
1421 // No need to calculate the TTL, the caller wont use it anyway.
1422 } elseif ( $usedSecond ) {
1423 $ttl = 1;
1424 } elseif ( $usedMinute ) {
1425 $ttl = 60 - substr( $ts, 12, 2 );
1426 } elseif ( $usedHour ) {
1427 $ttl = 3600 - substr( $ts, 10, 2 ) * 60 - substr( $ts, 12, 2 );
1428 } elseif ( $usedAMPM ) {
1429 $ttl = 43200 - ( substr( $ts, 8, 2 ) % 12 ) * 3600 -
1430 substr( $ts, 10, 2 ) * 60 - substr( $ts, 12, 2 );
1431 } elseif (
1432 $usedDay ||
1433 $usedHebrewMonth ||
1434 $usedIranianMonth ||
1435 $usedHijriMonth ||
1436 $usedHebrewYear ||
1437 $usedIranianYear ||
1438 $usedHijriYear ||
1439 $usedTennoYear
1440 ) {
1441 // @todo Someone who understands the non-Gregorian calendars
1442 // should write proper logic for them so that they don't need purged every day.
1443 $ttl = 86400 - substr( $ts, 8, 2 ) * 3600 -
1444 substr( $ts, 10, 2 ) * 60 - substr( $ts, 12, 2 );
1445 } else {
1446 $possibleTtls = [];
1447 $timeRemainingInDay = 86400 - substr( $ts, 8, 2 ) * 3600 -
1448 substr( $ts, 10, 2 ) * 60 - substr( $ts, 12, 2 );
1449 if ( $usedWeek ) {
1450 $possibleTtls[] =
1451 ( 7 - self::dateTimeObjFormat( $dateTimeObj, $ts, $zone, 'N' ) ) * 86400 +
1452 $timeRemainingInDay;
1453 } elseif ( $usedISOYear ) {
1454 // December 28th falls on the last ISO week of the year, every year.
1455 // The last ISO week of a year can be 52 or 53.
1456 $lastWeekOfISOYear = DateTime::createFromFormat(
1457 'Ymd',
1458 substr( $ts, 0, 4 ) . '1228',
1459 $zone ?: new DateTimeZone( 'UTC' )
1460 )->format( 'W' );
1461 $currentISOWeek = self::dateTimeObjFormat( $dateTimeObj, $ts, $zone, 'W' );
1462 $weeksRemaining = $lastWeekOfISOYear - $currentISOWeek;
1463 $timeRemainingInWeek =
1464 ( 7 - self::dateTimeObjFormat( $dateTimeObj, $ts, $zone, 'N' ) ) * 86400
1465 + $timeRemainingInDay;
1466 $possibleTtls[] = $weeksRemaining * 604800 + $timeRemainingInWeek;
1467 }
1468
1469 if ( $usedMonth ) {
1470 $possibleTtls[] =
1471 ( self::dateTimeObjFormat( $dateTimeObj, $ts, $zone, 't' ) -
1472 substr( $ts, 6, 2 ) ) * 86400
1473 + $timeRemainingInDay;
1474 } elseif ( $usedYear ) {
1475 $possibleTtls[] =
1476 ( self::dateTimeObjFormat( $dateTimeObj, $ts, $zone, 'L' ) + 364 -
1477 self::dateTimeObjFormat( $dateTimeObj, $ts, $zone, 'z' ) ) * 86400
1478 + $timeRemainingInDay;
1479 } elseif ( $usedIsLeapYear ) {
1480 $year = substr( $ts, 0, 4 );
1481 $timeRemainingInYear =
1482 ( self::dateTimeObjFormat( $dateTimeObj, $ts, $zone, 'L' ) + 364 -
1483 self::dateTimeObjFormat( $dateTimeObj, $ts, $zone, 'z' ) ) * 86400
1484 + $timeRemainingInDay;
1485 $mod = $year % 4;
1486 if ( $mod || ( !( $year % 100 ) && $year % 400 ) ) {
1487 // this isn't a leap year. see when the next one starts
1488 $nextCandidate = $year - $mod + 4;
1489 if ( $nextCandidate % 100 || !( $nextCandidate % 400 ) ) {
1490 $possibleTtls[] = ( $nextCandidate - $year - 1 ) * 365 * 86400 +
1491 $timeRemainingInYear;
1492 } else {
1493 $possibleTtls[] = ( $nextCandidate - $year + 3 ) * 365 * 86400 +
1494 $timeRemainingInYear;
1495 }
1496 } else {
1497 // this is a leap year, so the next year isn't
1498 $possibleTtls[] = $timeRemainingInYear;
1499 }
1500 }
1501
1502 if ( $possibleTtls ) {
1503 $ttl = min( $possibleTtls );
1504 }
1505 }
1506
1507 return $s;
1508 }
1509
1510 private static $GREG_DAYS = [ 31, 28, 31, 30, 31, 30, 31, 31, 30, 31, 30, 31 ];
1511 private static $IRANIAN_DAYS = [ 31, 31, 31, 31, 31, 31, 30, 30, 30, 30, 30, 29 ];
1512
1525 private static function tsToIranian( $ts ) {
1526 $gy = substr( $ts, 0, 4 ) - 1600;
1527 $gm = substr( $ts, 4, 2 ) - 1;
1528 $gd = substr( $ts, 6, 2 ) - 1;
1529
1530 # Days passed from the beginning (including leap years)
1531 $gDayNo = 365 * $gy
1532 + floor( ( $gy + 3 ) / 4 )
1533 - floor( ( $gy + 99 ) / 100 )
1534 + floor( ( $gy + 399 ) / 400 );
1535
1536 // Add days of the past months of this year
1537 for ( $i = 0; $i < $gm; $i++ ) {
1538 $gDayNo += self::$GREG_DAYS[$i];
1539 }
1540
1541 // Leap years
1542 if ( $gm > 1 && ( ( $gy % 4 === 0 && $gy % 100 !== 0 || ( $gy % 400 == 0 ) ) ) ) {
1543 $gDayNo++;
1544 }
1545
1546 // Days passed in current month
1547 $gDayNo += (int)$gd;
1548
1549 $jDayNo = $gDayNo - 79;
1550
1551 $jNp = floor( $jDayNo / 12053 );
1552 $jDayNo %= 12053;
1553
1554 $jy = 979 + 33 * $jNp + 4 * floor( $jDayNo / 1461 );
1555 $jDayNo %= 1461;
1556
1557 if ( $jDayNo >= 366 ) {
1558 $jy += floor( ( $jDayNo - 1 ) / 365 );
1559 $jDayNo = floor( ( $jDayNo - 1 ) % 365 );
1560 }
1561
1562 $jz = $jDayNo;
1563
1564 for ( $i = 0; $i < 11 && $jDayNo >= self::$IRANIAN_DAYS[$i]; $i++ ) {
1565 $jDayNo -= self::$IRANIAN_DAYS[$i];
1566 }
1567
1568 $jm = $i + 1;
1569 $jd = $jDayNo + 1;
1570
1571 return [ $jy, $jm, $jd, $jz ];
1572 }
1573
1585 private static function tsToHijri( $ts ) {
1586 $year = substr( $ts, 0, 4 );
1587 $month = substr( $ts, 4, 2 );
1588 $day = substr( $ts, 6, 2 );
1589
1590 $zyr = $year;
1591 $zd = $day;
1592 $zm = $month;
1593 $zy = $zyr;
1594
1595 if (
1596 ( $zy > 1582 ) || ( ( $zy == 1582 ) && ( $zm > 10 ) ) ||
1597 ( ( $zy == 1582 ) && ( $zm == 10 ) && ( $zd > 14 ) )
1598 ) {
1599 $zjd = (int)( ( 1461 * ( $zy + 4800 + (int)( ( $zm - 14 ) / 12 ) ) ) / 4 ) +
1600 (int)( ( 367 * ( $zm - 2 - 12 * ( (int)( ( $zm - 14 ) / 12 ) ) ) ) / 12 ) -
1601 (int)( ( 3 * (int)( ( ( $zy + 4900 + (int)( ( $zm - 14 ) / 12 ) ) / 100 ) ) ) / 4 ) +
1602 $zd - 32075;
1603 } else {
1604 $zjd = 367 * $zy - (int)( ( 7 * ( $zy + 5001 + (int)( ( $zm - 9 ) / 7 ) ) ) / 4 ) +
1605 (int)( ( 275 * $zm ) / 9 ) + $zd + 1729777;
1606 }
1607
1608 $zl = $zjd - 1948440 + 10632;
1609 $zn = (int)( ( $zl - 1 ) / 10631 );
1610 $zl = $zl - 10631 * $zn + 354;
1611 $zj = ( (int)( ( 10985 - $zl ) / 5316 ) ) * ( (int)( ( 50 * $zl ) / 17719 ) ) +
1612 ( (int)( $zl / 5670 ) ) * ( (int)( ( 43 * $zl ) / 15238 ) );
1613 $zl = $zl - ( (int)( ( 30 - $zj ) / 15 ) ) * ( (int)( ( 17719 * $zj ) / 50 ) ) -
1614 ( (int)( $zj / 16 ) ) * ( (int)( ( 15238 * $zj ) / 43 ) ) + 29;
1615 $zm = (int)( ( 24 * $zl ) / 709 );
1616 $zd = $zl - (int)( ( 709 * $zm ) / 24 );
1617 $zy = 30 * $zn + $zj - 30;
1618
1619 return [ $zy, $zm, $zd ];
1620 }
1621
1637 private static function tsToHebrew( $ts ) {
1638 # Parse date
1639 $year = substr( $ts, 0, 4 );
1640 $month = substr( $ts, 4, 2 );
1641 $day = substr( $ts, 6, 2 );
1642
1643 # Calculate Hebrew year
1644 $hebrewYear = $year + 3760;
1645
1646 # Month number when September = 1, August = 12
1647 $month += 4;
1648 if ( $month > 12 ) {
1649 # Next year
1650 $month -= 12;
1651 $year++;
1652 $hebrewYear++;
1653 }
1654
1655 # Calculate day of year from 1 September
1656 $dayOfYear = $day;
1657 for ( $i = 1; $i < $month; $i++ ) {
1658 if ( $i == 6 ) {
1659 # February
1660 $dayOfYear += 28;
1661 # Check if the year is leap
1662 if ( $year % 400 == 0 || ( $year % 4 == 0 && $year % 100 > 0 ) ) {
1663 $dayOfYear++;
1664 }
1665 } elseif ( $i == 8 || $i == 10 || $i == 1 || $i == 3 ) {
1666 $dayOfYear += 30;
1667 } else {
1668 $dayOfYear += 31;
1669 }
1670 }
1671
1672 # Calculate the start of the Hebrew year
1673 $start = self::hebrewYearStart( $hebrewYear );
1674
1675 # Calculate next year's start
1676 if ( $dayOfYear <= $start ) {
1677 # Day is before the start of the year - it is the previous year
1678 # Next year's start
1679 $nextStart = $start;
1680 # Previous year
1681 $year--;
1682 $hebrewYear--;
1683 # Add days since previous year's 1 September
1684 $dayOfYear += 365;
1685 if ( ( $year % 400 == 0 ) || ( $year % 100 != 0 && $year % 4 == 0 ) ) {
1686 # Leap year
1687 $dayOfYear++;
1688 }
1689 # Start of the new (previous) year
1690 $start = self::hebrewYearStart( $hebrewYear );
1691 } else {
1692 # Next year's start
1693 $nextStart = self::hebrewYearStart( $hebrewYear + 1 );
1694 }
1695
1696 # Calculate Hebrew day of year
1697 $hebrewDayOfYear = $dayOfYear - $start;
1698
1699 # Difference between year's days
1700 $diff = $nextStart - $start;
1701 # Add 12 (or 13 for leap years) days to ignore the difference between
1702 # Hebrew and Gregorian year (353 at least vs. 365/6) - now the
1703 # difference is only about the year type
1704 if ( ( $year % 400 == 0 ) || ( $year % 100 != 0 && $year % 4 == 0 ) ) {
1705 $diff += 13;
1706 } else {
1707 $diff += 12;
1708 }
1709
1710 # Check the year pattern, and is leap year
1711 # 0 means an incomplete year, 1 means a regular year, 2 means a complete year
1712 # This is mod 30, to work on both leap years (which add 30 days of Adar I)
1713 # and non-leap years
1714 $yearPattern = $diff % 30;
1715 # Check if leap year
1716 $isLeap = $diff >= 30;
1717
1718 # Calculate day in the month from number of day in the Hebrew year
1719 # Don't check Adar - if the day is not in Adar, we will stop before;
1720 # if it is in Adar, we will use it to check if it is Adar I or Adar II
1721 $hebrewDay = $hebrewDayOfYear;
1722 $hebrewMonth = 1;
1723 $days = 0;
1724 while ( $hebrewMonth <= 12 ) {
1725 # Calculate days in this month
1726 if ( $isLeap && $hebrewMonth == 6 ) {
1727 # Leap year - has Adar I, with 30 days, and Adar II, with 29 days
1728 $days = 30;
1729 if ( $hebrewDay <= $days ) {
1730 # Day in Adar I
1731 $hebrewMonth = 13;
1732 } else {
1733 # Subtract the days of Adar I
1734 $hebrewDay -= $days;
1735 # Try Adar II
1736 $days = 29;
1737 if ( $hebrewDay <= $days ) {
1738 # Day in Adar II
1739 $hebrewMonth = 14;
1740 }
1741 }
1742 } elseif ( $hebrewMonth == 2 && $yearPattern == 2 ) {
1743 # Cheshvan in a complete year (otherwise as the rule below)
1744 $days = 30;
1745 } elseif ( $hebrewMonth == 3 && $yearPattern == 0 ) {
1746 # Kislev in an incomplete year (otherwise as the rule below)
1747 $days = 29;
1748 } else {
1749 # Odd months have 30 days, even have 29
1750 $days = 30 - ( $hebrewMonth - 1 ) % 2;
1751 }
1752 if ( $hebrewDay <= $days ) {
1753 # In the current month
1754 break;
1755 } else {
1756 # Subtract the days of the current month
1757 $hebrewDay -= $days;
1758 # Try in the next month
1759 $hebrewMonth++;
1760 }
1761 }
1762
1763 return [ $hebrewYear, $hebrewMonth, $hebrewDay, $days ];
1764 }
1765
1775 private static function hebrewYearStart( $year ) {
1776 $a = intval( ( 12 * ( $year - 1 ) + 17 ) % 19 );
1777 $b = intval( ( $year - 1 ) % 4 );
1778 $m = 32.044093161144 + 1.5542417966212 * $a + $b / 4.0 - 0.0031777940220923 * ( $year - 1 );
1779 if ( $m < 0 ) {
1780 $m--;
1781 }
1782 $Mar = intval( $m );
1783 if ( $m < 0 ) {
1784 $m++;
1785 }
1786 $m -= $Mar;
1787
1788 $c = intval( ( $Mar + 3 * ( $year - 1 ) + 5 * $b + 5 ) % 7 );
1789 if ( $c == 0 && $a > 11 && $m >= 0.89772376543210 ) {
1790 $Mar++;
1791 } elseif ( $c == 1 && $a > 6 && $m >= 0.63287037037037 ) {
1792 $Mar += 2;
1793 } elseif ( $c == 2 || $c == 4 || $c == 6 ) {
1794 $Mar++;
1795 }
1796
1797 $Mar += intval( ( $year - 3761 ) / 100 ) - intval( ( $year - 3761 ) / 400 ) - 24;
1798 return $Mar;
1799 }
1800
1813 private static function tsToYear( $ts, $cName ) {
1814 $gy = substr( $ts, 0, 4 );
1815 $gm = substr( $ts, 4, 2 );
1816 $gd = substr( $ts, 6, 2 );
1817
1818 if ( !strcmp( $cName, 'thai' ) ) {
1819 # Thai solar dates
1820 # Add 543 years to the Gregorian calendar
1821 # Months and days are identical
1822 $gy_offset = $gy + 543;
1823 # fix for dates between 1912 and 1941
1824 # https://en.wikipedia.org/?oldid=836596673#New_year
1825 if ( $gy >= 1912 && $gy <= 1940 ) {
1826 if ( $gm <= 3 ) {
1827 $gy_offset--;
1828 }
1829 $gm = ( $gm - 3 ) % 12;
1830 }
1831 } elseif ( ( !strcmp( $cName, 'minguo' ) ) || !strcmp( $cName, 'juche' ) ) {
1832 # Minguo dates
1833 # Deduct 1911 years from the Gregorian calendar
1834 # Months and days are identical
1835 $gy_offset = $gy - 1911;
1836 } elseif ( !strcmp( $cName, 'tenno' ) ) {
1837 # Nengō dates up to Meiji period
1838 # Deduct years from the Gregorian calendar
1839 # depending on the nengo periods
1840 # Months and days are identical
1841 if ( ( $gy < 1912 )
1842 || ( ( $gy == 1912 ) && ( $gm < 7 ) )
1843 || ( ( $gy == 1912 ) && ( $gm == 7 ) && ( $gd < 31 ) )
1844 ) {
1845 # Meiji period
1846 $gy_gannen = $gy - 1868 + 1;
1847 $gy_offset = $gy_gannen;
1848 if ( $gy_gannen == 1 ) {
1849 $gy_offset = '元';
1850 }
1851 $gy_offset = '明治' . $gy_offset;
1852 } elseif (
1853 ( ( $gy == 1912 ) && ( $gm == 7 ) && ( $gd == 31 ) ) ||
1854 ( ( $gy == 1912 ) && ( $gm >= 8 ) ) ||
1855 ( ( $gy > 1912 ) && ( $gy < 1926 ) ) ||
1856 ( ( $gy == 1926 ) && ( $gm < 12 ) ) ||
1857 ( ( $gy == 1926 ) && ( $gm == 12 ) && ( $gd < 26 ) )
1858 ) {
1859 # Taishō period
1860 $gy_gannen = $gy - 1912 + 1;
1861 $gy_offset = $gy_gannen;
1862 if ( $gy_gannen == 1 ) {
1863 $gy_offset = '元';
1864 }
1865 $gy_offset = '大正' . $gy_offset;
1866 } elseif (
1867 ( ( $gy == 1926 ) && ( $gm == 12 ) && ( $gd >= 26 ) ) ||
1868 ( ( $gy > 1926 ) && ( $gy < 1989 ) ) ||
1869 ( ( $gy == 1989 ) && ( $gm == 1 ) && ( $gd < 8 ) )
1870 ) {
1871 # Shōwa period
1872 $gy_gannen = $gy - 1926 + 1;
1873 $gy_offset = $gy_gannen;
1874 if ( $gy_gannen == 1 ) {
1875 $gy_offset = '元';
1876 }
1877 $gy_offset = '昭和' . $gy_offset;
1878 } elseif (
1879 ( ( $gy == 1989 ) && ( $gm == 1 ) && ( $gd >= 8 ) ) ||
1880 ( ( $gy > 1989 ) && ( $gy < 2019 ) ) ||
1881 ( ( $gy == 2019 ) && ( $gm < 5 ) )
1882 ) {
1883 # Heisei period
1884 $gy_gannen = $gy - 1989 + 1;
1885 $gy_offset = $gy_gannen;
1886 if ( $gy_gannen == 1 ) {
1887 $gy_offset = '元';
1888 }
1889 $gy_offset = '平成' . $gy_offset;
1890 } else {
1891 # Reiwa period
1892 $gy_gannen = $gy - 2019 + 1;
1893 $gy_offset = $gy_gannen;
1894 if ( $gy_gannen == 1 ) {
1895 $gy_offset = '元';
1896 }
1897 $gy_offset = '令和' . $gy_offset;
1898 }
1899 } else {
1900 $gy_offset = $gy;
1901 }
1902
1903 return [ $gy_offset, $gm, $gd ];
1904 }
1905
1919 private static function strongDirFromContent( $text = '' ) {
1920 if ( !preg_match( self::$strongDirRegex, $text, $matches ) ) {
1921 return null;
1922 }
1923 if ( $matches[1] === '' ) {
1924 return 'rtl';
1925 }
1926 return 'ltr';
1927 }
1928
1936 static function romanNumeral( $num ) {
1937 static $table = [
1938 [ '', 'I', 'II', 'III', 'IV', 'V', 'VI', 'VII', 'VIII', 'IX', 'X' ],
1939 [ '', 'X', 'XX', 'XXX', 'XL', 'L', 'LX', 'LXX', 'LXXX', 'XC', 'C' ],
1940 [ '', 'C', 'CC', 'CCC', 'CD', 'D', 'DC', 'DCC', 'DCCC', 'CM', 'M' ],
1941 [ '', 'M', 'MM', 'MMM', 'MMMM', 'MMMMM', 'MMMMMM', 'MMMMMMM',
1942 'MMMMMMMM', 'MMMMMMMMM', 'MMMMMMMMMM' ]
1943 ];
1944
1945 $num = intval( $num );
1946 if ( $num > 10000 || $num <= 0 ) {
1947 return $num;
1948 }
1949
1950 $s = '';
1951 for ( $pow10 = 1000, $i = 3; $i >= 0; $pow10 /= 10, $i-- ) {
1952 if ( $num >= $pow10 ) {
1953 $s .= $table[$i][(int)floor( $num / $pow10 )];
1954 }
1955 $num = $num % $pow10;
1956 }
1957 return $s;
1958 }
1959
1967 static function hebrewNumeral( $num ) {
1968 static $table = [
1969 [ '', 'א', 'ב', 'ג', 'ד', 'ה', 'ו', 'ז', 'ח', 'ט', 'י' ],
1970 [ '', 'י', 'כ', 'ל', 'מ', 'נ', 'ס', 'ע', 'פ', 'צ', 'ק' ],
1971 [ '',
1972 [ 'ק' ],
1973 [ 'ר' ],
1974 [ 'ש' ],
1975 [ 'ת' ],
1976 [ 'ת', 'ק' ],
1977 [ 'ת', 'ר' ],
1978 [ 'ת', 'ש' ],
1979 [ 'ת', 'ת' ],
1980 [ 'ת', 'ת', 'ק' ],
1981 [ 'ת', 'ת', 'ר' ],
1982 ],
1983 [ '', 'א', 'ב', 'ג', 'ד', 'ה', 'ו', 'ז', 'ח', 'ט', 'י' ]
1984 ];
1985
1986 $num = intval( $num );
1987 if ( $num > 9999 || $num <= 0 ) {
1988 return $num;
1989 }
1990
1991 // Round thousands have special notations
1992 if ( $num === 1000 ) {
1993 return "א' אלף";
1994 } elseif ( $num % 1000 === 0 ) {
1995 return $table[0][$num / 1000] . "' אלפים";
1996 }
1997
1998 $letters = [];
1999
2000 for ( $pow10 = 1000, $i = 3; $i >= 0; $pow10 /= 10, $i-- ) {
2001 if ( $num >= $pow10 ) {
2002 if ( $num === 15 || $num === 16 ) {
2003 $letters[] = $table[0][9];
2004 $letters[] = $table[0][$num - 9];
2005 $num = 0;
2006 } else {
2007 $letters = array_merge(
2008 $letters,
2009 (array)$table[$i][intval( $num / $pow10 )]
2010 );
2011
2012 if ( $pow10 === 1000 ) {
2013 $letters[] = "'";
2014 }
2015 }
2016 }
2017
2018 $num = $num % $pow10;
2019 }
2020
2021 $preTransformLength = count( $letters );
2022 if ( $preTransformLength === 1 ) {
2023 // Add geresh (single quote) to one-letter numbers
2024 $letters[] = "'";
2025 } else {
2026 $lastIndex = $preTransformLength - 1;
2027 $letters[$lastIndex] = str_replace(
2028 [ 'כ', 'מ', 'נ', 'פ', 'צ' ],
2029 [ 'ך', 'ם', 'ן', 'ף', 'ץ' ],
2030 $letters[$lastIndex]
2031 );
2032
2033 // Add gershayim (double quote) to multiple-letter numbers,
2034 // but exclude numbers with only one letter after the thousands
2035 // (1001-1009, 1020, 1030, 2001-2009, etc.)
2036 if ( $letters[1] === "'" && $preTransformLength === 3 ) {
2037 $letters[] = "'";
2038 } else {
2039 array_splice( $letters, -1, 0, '"' );
2040 }
2041 }
2042
2043 return implode( $letters );
2044 }
2045
2054 public function userAdjust( $ts, $tz = false ) {
2055 global $wgUser, $wgLocalTZoffset;
2056
2057 if ( $tz === false ) {
2058 $tz = $wgUser->getOption( 'timecorrection' );
2059 }
2060
2061 $data = explode( '|', $tz, 3 );
2062
2063 if ( $data[0] == 'ZoneInfo' ) {
2064 try {
2065 $userTZ = new DateTimeZone( $data[2] );
2066 $date = new DateTime( $ts, new DateTimeZone( 'UTC' ) );
2067 $date->setTimezone( $userTZ );
2068 return $date->format( 'YmdHis' );
2069 } catch ( Exception $e ) {
2070 // Unrecognized timezone, default to 'Offset' with the stored offset.
2071 $data[0] = 'Offset';
2072 }
2073 }
2074
2075 if ( $data[0] == 'System' || $tz == '' ) {
2076 # Global offset in minutes.
2077 $minDiff = $wgLocalTZoffset;
2078 } elseif ( $data[0] == 'Offset' ) {
2079 $minDiff = intval( $data[1] );
2080 } else {
2081 $data = explode( ':', $tz );
2082 if ( count( $data ) == 2 ) {
2083 $data[0] = intval( $data[0] );
2084 $data[1] = intval( $data[1] );
2085 $minDiff = abs( $data[0] ) * 60 + $data[1];
2086 if ( $data[0] < 0 ) {
2087 $minDiff = -$minDiff;
2088 }
2089 } else {
2090 $minDiff = intval( $data[0] ) * 60;
2091 }
2092 }
2093
2094 # No difference ? Return time unchanged
2095 if ( $minDiff == 0 ) {
2096 return $ts;
2097 }
2098
2099 Wikimedia\suppressWarnings(); // E_STRICT system time bitching
2100 # Generate an adjusted date; take advantage of the fact that mktime
2101 # will normalize out-of-range values so we don't have to split $minDiff
2102 # into hours and minutes.
2103 $t = mktime( (
2104 (int)substr( $ts, 8, 2 ) ), # Hours
2105 (int)substr( $ts, 10, 2 ) + $minDiff, # Minutes
2106 (int)substr( $ts, 12, 2 ), # Seconds
2107 (int)substr( $ts, 4, 2 ), # Month
2108 (int)substr( $ts, 6, 2 ), # Day
2109 (int)substr( $ts, 0, 4 ) ); # Year
2110
2111 $date = date( 'YmdHis', $t );
2112 Wikimedia\restoreWarnings();
2113
2114 return $date;
2115 }
2116
2132 function dateFormat( $usePrefs = true ) {
2133 global $wgUser;
2134
2135 if ( is_bool( $usePrefs ) ) {
2136 if ( $usePrefs ) {
2137 $datePreference = $wgUser->getDatePreference();
2138 } else {
2139 $datePreference = (string)User::getDefaultOption( 'date' );
2140 }
2141 } else {
2142 $datePreference = (string)$usePrefs;
2143 }
2144
2145 // return int
2146 if ( $datePreference == '' ) {
2147 return 'default';
2148 }
2149
2150 return $datePreference;
2151 }
2152
2163 function getDateFormatString( $type, $pref ) {
2164 $wasDefault = false;
2165 if ( $pref == 'default' ) {
2166 $wasDefault = true;
2167 $pref = $this->getDefaultDateFormat();
2168 }
2169
2170 if ( !isset( $this->dateFormatStrings[$type][$pref] ) ) {
2171 $df =
2172 $this->localisationCache->getSubitem( $this->mCode, 'dateFormats', "$pref $type" );
2173
2174 if ( $type === 'pretty' && $df === null ) {
2175 $df = $this->getDateFormatString( 'date', $pref );
2176 }
2177
2178 if ( !$wasDefault && $df === null ) {
2179 $pref = $this->getDefaultDateFormat();
2180 $df = $this->getLocalisationCache()
2181 ->getSubitem( $this->mCode, 'dateFormats', "$pref $type" );
2182 }
2183
2184 $this->dateFormatStrings[$type][$pref] = $df;
2185 }
2186 return $this->dateFormatStrings[$type][$pref];
2187 }
2188
2199 public function date( $ts, $adj = false, $format = true, $timecorrection = false ) {
2200 $ts = wfTimestamp( TS_MW, $ts );
2201 if ( $adj ) {
2202 $ts = $this->userAdjust( $ts, $timecorrection );
2203 }
2204 $df = $this->getDateFormatString( 'date', $this->dateFormat( $format ) );
2205 return $this->sprintfDate( $df, $ts );
2206 }
2207
2218 public function time( $ts, $adj = false, $format = true, $timecorrection = false ) {
2219 $ts = wfTimestamp( TS_MW, $ts );
2220 if ( $adj ) {
2221 $ts = $this->userAdjust( $ts, $timecorrection );
2222 }
2223 $df = $this->getDateFormatString( 'time', $this->dateFormat( $format ) );
2224 return $this->sprintfDate( $df, $ts );
2225 }
2226
2238 public function timeanddate( $ts, $adj = false, $format = true, $timecorrection = false ) {
2239 $ts = wfTimestamp( TS_MW, $ts );
2240 if ( $adj ) {
2241 $ts = $this->userAdjust( $ts, $timecorrection );
2242 }
2243 $df = $this->getDateFormatString( 'both', $this->dateFormat( $format ) );
2244 return $this->sprintfDate( $df, $ts );
2245 }
2246
2257 public function formatDuration( $seconds, array $chosenIntervals = [] ) {
2258 $intervals = $this->getDurationIntervals( $seconds, $chosenIntervals );
2259
2260 $segments = [];
2261
2262 foreach ( $intervals as $intervalName => $intervalValue ) {
2263 // Messages: duration-seconds, duration-minutes, duration-hours, duration-days, duration-weeks,
2264 // duration-years, duration-decades, duration-centuries, duration-millennia
2265 $message = wfMessage( 'duration-' . $intervalName )->numParams( $intervalValue );
2266 $segments[] = $message->inLanguage( $this )->escaped();
2267 }
2268
2269 return $this->listToText( $segments );
2270 }
2271
2283 public function getDurationIntervals( $seconds, array $chosenIntervals = [] ) {
2284 if ( empty( $chosenIntervals ) ) {
2285 $chosenIntervals = [
2286 'millennia',
2287 'centuries',
2288 'decades',
2289 'years',
2290 'days',
2291 'hours',
2292 'minutes',
2293 'seconds'
2294 ];
2295 }
2296
2297 $intervals = array_intersect_key( self::$durationIntervals, array_flip( $chosenIntervals ) );
2298 $sortedNames = array_keys( $intervals );
2299 $smallestInterval = array_pop( $sortedNames );
2300
2301 $segments = [];
2302
2303 foreach ( $intervals as $name => $length ) {
2304 $value = floor( $seconds / $length );
2305
2306 if ( $value > 0 || ( $name == $smallestInterval && empty( $segments ) ) ) {
2307 $seconds -= $value * $length;
2308 $segments[$name] = $value;
2309 }
2310 }
2311
2312 return $segments;
2313 }
2314
2334 private function internalUserTimeAndDate( $type, $ts, User $user, array $options ) {
2335 $ts = wfTimestamp( TS_MW, $ts );
2336 $options += [ 'timecorrection' => true, 'format' => true ];
2337 if ( $options['timecorrection'] !== false ) {
2338 if ( $options['timecorrection'] === true ) {
2339 $offset = $user->getOption( 'timecorrection' );
2340 } else {
2341 $offset = $options['timecorrection'];
2342 }
2343 $ts = $this->userAdjust( $ts, $offset );
2344 }
2345 if ( $options['format'] === true ) {
2346 $format = $user->getDatePreference();
2347 } else {
2348 $format = $options['format'];
2349 }
2350 $df = $this->getDateFormatString( $type, $this->dateFormat( $format ) );
2351 return $this->sprintfDate( $df, $ts );
2352 }
2353
2373 public function userDate( $ts, User $user, array $options = [] ) {
2374 return $this->internalUserTimeAndDate( 'date', $ts, $user, $options );
2375 }
2376
2396 public function userTime( $ts, User $user, array $options = [] ) {
2397 return $this->internalUserTimeAndDate( 'time', $ts, $user, $options );
2398 }
2399
2419 public function userTimeAndDate( $ts, User $user, array $options = [] ) {
2420 return $this->internalUserTimeAndDate( 'both', $ts, $user, $options );
2421 }
2422
2438 public function getHumanTimestamp(
2439 MWTimestamp $time, MWTimestamp $relativeTo = null, User $user = null
2440 ) {
2441 if ( $relativeTo === null ) {
2442 $relativeTo = new MWTimestamp();
2443 }
2444 if ( $user === null ) {
2445 $user = RequestContext::getMain()->getUser();
2446 }
2447
2448 // Adjust for the user's timezone.
2449 $offsetThis = $time->offsetForUser( $user );
2450 $offsetRel = $relativeTo->offsetForUser( $user );
2451
2452 $ts = '';
2453 if ( Hooks::run( 'GetHumanTimestamp', [ &$ts, $time, $relativeTo, $user, $this ] ) ) {
2454 $ts = $this->getHumanTimestampInternal( $time, $relativeTo, $user );
2455 }
2456
2457 // Reset the timezone on the objects.
2458 $time->timestamp->sub( $offsetThis );
2459 $relativeTo->timestamp->sub( $offsetRel );
2460
2461 return $ts;
2462 }
2463
2476 MWTimestamp $ts, MWTimestamp $relativeTo, User $user
2477 ) {
2478 $diff = $ts->diff( $relativeTo );
2479 $diffDay = (bool)( (int)$ts->timestamp->format( 'w' ) -
2480 (int)$relativeTo->timestamp->format( 'w' ) );
2481 $days = $diff->days ?: (int)$diffDay;
2482 if ( $diff->invert || $days > 5
2483 && $ts->timestamp->format( 'Y' ) !== $relativeTo->timestamp->format( 'Y' )
2484 ) {
2485 // Timestamps are in different years: use full timestamp
2486 // Also do full timestamp for future dates
2490 $format = $this->getDateFormatString( 'both', $user->getDatePreference() ?: 'default' );
2491 $ts = $this->sprintfDate( $format, $ts->getTimestamp( TS_MW ) );
2492 } elseif ( $days > 5 ) {
2493 // Timestamps are in same year, but more than 5 days ago: show day and month only.
2494 $format = $this->getDateFormatString( 'pretty', $user->getDatePreference() ?: 'default' );
2495 $ts = $this->sprintfDate( $format, $ts->getTimestamp( TS_MW ) );
2496 } elseif ( $days > 1 ) {
2497 // Timestamp within the past week: show the day of the week and time
2498 $format = $this->getDateFormatString( 'time', $user->getDatePreference() ?: 'default' );
2499 $weekday = self::$mWeekdayMsgs[$ts->timestamp->format( 'w' )];
2500 // Messages:
2501 // sunday-at, monday-at, tuesday-at, wednesday-at, thursday-at, friday-at, saturday-at
2502 $ts = wfMessage( "$weekday-at" )
2503 ->inLanguage( $this )
2504 ->params( $this->sprintfDate( $format, $ts->getTimestamp( TS_MW ) ) )
2505 ->text();
2506 } elseif ( $days == 1 ) {
2507 // Timestamp was yesterday: say 'yesterday' and the time.
2508 $format = $this->getDateFormatString( 'time', $user->getDatePreference() ?: 'default' );
2509 $ts = wfMessage( 'yesterday-at' )
2510 ->inLanguage( $this )
2511 ->params( $this->sprintfDate( $format, $ts->getTimestamp( TS_MW ) ) )
2512 ->text();
2513 } elseif ( $diff->h > 1 || $diff->h == 1 && $diff->i > 30 ) {
2514 // Timestamp was today, but more than 90 minutes ago: say 'today' and the time.
2515 $format = $this->getDateFormatString( 'time', $user->getDatePreference() ?: 'default' );
2516 $ts = wfMessage( 'today-at' )
2517 ->inLanguage( $this )
2518 ->params( $this->sprintfDate( $format, $ts->getTimestamp( TS_MW ) ) )
2519 ->text();
2520
2521 // From here on in, the timestamp was soon enough ago so that we can simply say
2522 // XX units ago, e.g., "2 hours ago" or "5 minutes ago"
2523 } elseif ( $diff->h == 1 ) {
2524 // Less than 90 minutes, but more than an hour ago.
2525 $ts = wfMessage( 'hours-ago' )->inLanguage( $this )->numParams( 1 )->text();
2526 } elseif ( $diff->i >= 1 ) {
2527 // A few minutes ago.
2528 $ts = wfMessage( 'minutes-ago' )->inLanguage( $this )->numParams( $diff->i )->text();
2529 } elseif ( $diff->s >= 30 ) {
2530 // Less than a minute, but more than 30 sec ago.
2531 $ts = wfMessage( 'seconds-ago' )->inLanguage( $this )->numParams( $diff->s )->text();
2532 } else {
2533 // Less than 30 seconds ago.
2534 $ts = wfMessage( 'just-now' )->text();
2535 }
2536
2537 return $ts;
2538 }
2539
2544 public function getMessage( $key ) {
2545 return $this->localisationCache->getSubitem( $this->mCode, 'messages', $key );
2546 }
2547
2551 function getAllMessages() {
2552 return $this->localisationCache->getItem( $this->mCode, 'messages' );
2553 }
2554
2561 public function iconv( $in, $out, $string ) {
2562 # Even with //IGNORE iconv can whine about illegal characters in
2563 # *input* string. We just ignore those too.
2564 # REF: https://bugs.php.net/bug.php?id=37166
2565 # REF: https://phabricator.wikimedia.org/T18885
2566 Wikimedia\suppressWarnings();
2567 $text = iconv( $in, $out . '//IGNORE', $string );
2568 Wikimedia\restoreWarnings();
2569 return $text;
2570 }
2571
2572 // callback functions for ucwords(), ucwordbreaks()
2573
2579 return $this->ucfirst( $matches[1] );
2580 }
2581
2587 return mb_strtoupper( $matches[0] );
2588 }
2589
2595 return mb_strtoupper( $matches[0] );
2596 }
2597
2605 public function ucfirst( $str ) {
2606 $o = ord( $str );
2607 if ( $o < 96 ) { // if already uppercase...
2608 return $str;
2609 } elseif ( $o < 128 ) {
2610 return ucfirst( $str ); // use PHP's ucfirst()
2611 } else {
2612 // fall back to more complex logic in case of multibyte strings
2613 return $this->uc( $str, true );
2614 }
2615 }
2616
2625 public function uc( $str, $first = false ) {
2626 if ( $first ) {
2627 if ( $this->isMultibyte( $str ) ) {
2628 return $this->mbUpperChar( mb_substr( $str, 0, 1 ) ) . mb_substr( $str, 1 );
2629 } else {
2630 return ucfirst( $str );
2631 }
2632 } else {
2633 return $this->isMultibyte( $str ) ? mb_strtoupper( $str ) : strtoupper( $str );
2634 }
2635 }
2636
2650 protected function mbUpperChar( $char ) {
2652 if ( array_key_exists( $char, $wgOverrideUcfirstCharacters ) ) {
2653 return $wgOverrideUcfirstCharacters[$char];
2654 } else {
2655 return mb_strtoupper( $char );
2656 }
2657 }
2658
2663 function lcfirst( $str ) {
2664 $o = ord( $str );
2665 if ( !$o ) {
2666 return strval( $str );
2667 } elseif ( $o >= 128 ) {
2668 return $this->lc( $str, true );
2669 } elseif ( $o > 96 ) {
2670 return $str;
2671 } else {
2672 $str[0] = strtolower( $str[0] );
2673 return $str;
2674 }
2675 }
2676
2682 function lc( $str, $first = false ) {
2683 if ( $first ) {
2684 if ( $this->isMultibyte( $str ) ) {
2685 return mb_strtolower( mb_substr( $str, 0, 1 ) ) . mb_substr( $str, 1 );
2686 } else {
2687 return strtolower( substr( $str, 0, 1 ) ) . substr( $str, 1 );
2688 }
2689 } else {
2690 return $this->isMultibyte( $str ) ? mb_strtolower( $str ) : strtolower( $str );
2691 }
2692 }
2693
2698 function isMultibyte( $str ) {
2699 return strlen( $str ) !== mb_strlen( $str );
2700 }
2701
2706 function ucwords( $str ) {
2707 if ( $this->isMultibyte( $str ) ) {
2708 $str = $this->lc( $str );
2709
2710 // regexp to find first letter in each word (i.e. after each space)
2711 $replaceRegexp = "/^([a-z]|[\\xc0-\\xff][\\x80-\\xbf]*)| ([a-z]|[\\xc0-\\xff][\\x80-\\xbf]*)/";
2712
2713 // function to use to capitalize a single char
2714 return preg_replace_callback(
2715 $replaceRegexp,
2716 [ $this, 'ucwordsCallbackMB' ],
2717 $str
2718 );
2719 } else {
2720 return ucwords( strtolower( $str ) );
2721 }
2722 }
2723
2730 function ucwordbreaks( $str ) {
2731 if ( $this->isMultibyte( $str ) ) {
2732 $str = $this->lc( $str );
2733
2734 // since \b doesn't work for UTF-8, we explicitely define word break chars
2735 $breaks = "[ \-\‍(\‍)\}\{\.,\?!]";
2736
2737 // find first letter after word break
2738 $replaceRegexp = "/^([a-z]|[\\xc0-\\xff][\\x80-\\xbf]*)|" .
2739 "$breaks([a-z]|[\\xc0-\\xff][\\x80-\\xbf]*)/";
2740
2741 return preg_replace_callback(
2742 $replaceRegexp,
2743 [ $this, 'ucwordbreaksCallbackMB' ],
2744 $str
2745 );
2746 } else {
2747 return preg_replace_callback(
2748 '/\b([\w\x80-\xff]+)\b/',
2749 [ $this, 'ucwordbreaksCallbackAscii' ],
2750 $str
2751 );
2752 }
2753 }
2754
2770 function caseFold( $s ) {
2771 return $this->uc( $s );
2772 }
2773
2780 if ( is_array( $s ) ) {
2781 throw new MWException( 'Given array to checkTitleEncoding.' );
2782 }
2783 if ( StringUtils::isUtf8( $s ) ) {
2784 return $s;
2785 }
2786
2787 return $this->iconv( $this->fallback8bitEncoding(), 'utf-8', $s );
2788 }
2789
2794 return $this->localisationCache->getItem( $this->mCode, 'fallback8bitEncoding' );
2795 }
2796
2805 function hasWordBreaks() {
2806 return true;
2807 }
2808
2816 function segmentByWord( $string ) {
2817 return $string;
2818 }
2819
2827 function normalizeForSearch( $string ) {
2828 return self::convertDoubleWidth( $string );
2829 }
2830
2839 protected static function convertDoubleWidth( $string ) {
2840 static $full = null;
2841 static $half = null;
2842
2843 if ( $full === null ) {
2844 $fullWidth = "0123456789ABCDEFGHIJKLMNOPQRSTUVWXYZabcdefghijklmnopqrstuvwxyz";
2845 $halfWidth = "0123456789ABCDEFGHIJKLMNOPQRSTUVWXYZabcdefghijklmnopqrstuvwxyz";
2846 $full = str_split( $fullWidth, 3 );
2847 $half = str_split( $halfWidth );
2848 }
2849
2850 $string = str_replace( $full, $half, $string );
2851 return $string;
2852 }
2853
2859 protected static function insertSpace( $string, $pattern ) {
2860 $string = preg_replace( $pattern, " $1 ", $string );
2861 $string = preg_replace( '/ +/', ' ', $string );
2862 return $string;
2863 }
2864
2869 function convertForSearchResult( $termsArray ) {
2870 # some languages, e.g. Chinese, need to do a conversion
2871 # in order for search results to be displayed correctly
2872 return $termsArray;
2873 }
2874
2881 function firstChar( $s ) {
2882 $matches = [];
2883 preg_match(
2884 '/^([\x00-\x7f]|[\xc0-\xdf][\x80-\xbf]|' .
2885 '[\xe0-\xef][\x80-\xbf]{2}|[\xf0-\xf7][\x80-\xbf]{3})/',
2886 $s,
2887 $matches
2888 );
2889
2890 if ( isset( $matches[1] ) ) {
2891 if ( strlen( $matches[1] ) != 3 ) {
2892 return $matches[1];
2893 }
2894
2895 // Break down Hangul syllables to grab the first jamo
2896 $code = UtfNormal\Utils::utf8ToCodepoint( $matches[1] );
2897 if ( $code < 0xac00 || 0xd7a4 <= $code ) {
2898 return $matches[1];
2899 } elseif ( $code < 0xb098 ) {
2900 return "\u{3131}";
2901 } elseif ( $code < 0xb2e4 ) {
2902 return "\u{3134}";
2903 } elseif ( $code < 0xb77c ) {
2904 return "\u{3137}";
2905 } elseif ( $code < 0xb9c8 ) {
2906 return "\u{3139}";
2907 } elseif ( $code < 0xbc14 ) {
2908 return "\u{3141}";
2909 } elseif ( $code < 0xc0ac ) {
2910 return "\u{3142}";
2911 } elseif ( $code < 0xc544 ) {
2912 return "\u{3145}";
2913 } elseif ( $code < 0xc790 ) {
2914 return "\u{3147}";
2915 } elseif ( $code < 0xcc28 ) {
2916 return "\u{3148}";
2917 } elseif ( $code < 0xce74 ) {
2918 return "\u{314A}";
2919 } elseif ( $code < 0xd0c0 ) {
2920 return "\u{314B}";
2921 } elseif ( $code < 0xd30c ) {
2922 return "\u{314C}";
2923 } elseif ( $code < 0xd558 ) {
2924 return "\u{314D}";
2925 } else {
2926 return "\u{314E}";
2927 }
2928 } else {
2929 return '';
2930 }
2931 }
2932
2944 public function normalize( $s ) {
2945 global $wgAllUnicodeFixes;
2946 $s = UtfNormal\Validator::cleanUp( $s );
2947 if ( $wgAllUnicodeFixes ) {
2948 $s = $this->transformUsingPairFile( 'normalize-ar.php', $s );
2949 $s = $this->transformUsingPairFile( 'normalize-ml.php', $s );
2950 }
2951
2952 return $s;
2953 }
2954
2969 protected function transformUsingPairFile( $file, $string ) {
2970 if ( !isset( $this->transformData[$file] ) ) {
2971 global $IP;
2972 $data = require "$IP/languages/data/{$file}";
2973 $this->transformData[$file] = new ReplacementArray( $data );
2974 }
2975 return $this->transformData[$file]->replace( $string );
2976 }
2977
2983 function isRTL() {
2984 return $this->localisationCache->getItem( $this->mCode, 'rtl' );
2985 }
2986
2991 function getDir() {
2992 return $this->isRTL() ? 'rtl' : 'ltr';
2993 }
2994
3003 function alignStart() {
3004 return $this->isRTL() ? 'right' : 'left';
3005 }
3006
3015 function alignEnd() {
3016 return $this->isRTL() ? 'left' : 'right';
3017 }
3018
3030 function getDirMarkEntity( $opposite = false ) {
3031 if ( $opposite ) {
3032 return $this->isRTL() ? '&lrm;' : '&rlm;';
3033 }
3034 return $this->isRTL() ? '&rlm;' : '&lrm;';
3035 }
3036
3047 function getDirMark( $opposite = false ) {
3048 $lrm = "\u{200E}"; # LEFT-TO-RIGHT MARK, commonly abbreviated LRM
3049 $rlm = "\u{200F}"; # RIGHT-TO-LEFT MARK, commonly abbreviated RLM
3050 if ( $opposite ) {
3051 return $this->isRTL() ? $lrm : $rlm;
3052 }
3053 return $this->isRTL() ? $rlm : $lrm;
3054 }
3055
3060 return $this->localisationCache->getItem( $this->mCode, 'capitalizeAllNouns' );
3061 }
3062
3070 function getArrow( $direction = 'forwards' ) {
3071 switch ( $direction ) {
3072 case 'forwards':
3073 return $this->isRTL() ? '←' : '→';
3074 case 'backwards':
3075 return $this->isRTL() ? '→' : '←';
3076 case 'left':
3077 return '←';
3078 case 'right':
3079 return '→';
3080 case 'up':
3081 return '↑';
3082 case 'down':
3083 return '↓';
3084 }
3085 }
3086
3093 return $this->localisationCache->getItem( $this->mCode, 'linkPrefixExtension' );
3094 }
3095
3100 function getMagicWords() {
3101 return $this->localisationCache->getItem( $this->mCode, 'magicWords' );
3102 }
3103
3109 function getMagic( $mw ) {
3110 $rawEntry = $this->mMagicExtensions[$mw->mId] ??
3111 $this->localisationCache->getSubitem( $this->mCode, 'magicWords', $mw->mId );
3112
3113 if ( !is_array( $rawEntry ) ) {
3114 wfWarn( "\"$rawEntry\" is not a valid magic word for \"$mw->mId\"" );
3115 } else {
3116 $mw->mCaseSensitive = $rawEntry[0];
3117 $mw->mSynonyms = array_slice( $rawEntry, 1 );
3118 }
3119 }
3120
3126 function addMagicWordsByLang( $newWords ) {
3127 $fallbackChain = $this->getFallbackLanguages();
3128 $fallbackChain = array_reverse( $fallbackChain );
3129 foreach ( $fallbackChain as $code ) {
3130 if ( isset( $newWords[$code] ) ) {
3131 $this->mMagicExtensions = $newWords[$code] + $this->mMagicExtensions;
3132 }
3133 }
3134 }
3135
3142 // Cache aliases because it may be slow to load them
3143 if ( is_null( $this->mExtendedSpecialPageAliases ) ) {
3144 // Initialise array
3145 $this->mExtendedSpecialPageAliases =
3146 $this->localisationCache->getItem( $this->mCode, 'specialPageAliases' );
3147 }
3148
3149 return $this->mExtendedSpecialPageAliases;
3150 }
3151
3158 function emphasize( $text ) {
3159 return "<em>$text</em>";
3160 }
3161
3184 public function formatNum( $number, $nocommafy = false ) {
3185 global $wgTranslateNumerals;
3186 if ( !$nocommafy ) {
3187 $number = $this->commafy( $number );
3188 $s = $this->separatorTransformTable();
3189 if ( $s ) {
3190 $number = strtr( $number, $s );
3191 }
3192 }
3193
3194 if ( $wgTranslateNumerals ) {
3195 $s = $this->digitTransformTable();
3196 if ( $s ) {
3197 $number = strtr( $number, $s );
3198 }
3199 }
3200
3201 return (string)$number;
3202 }
3203
3212 public function formatNumNoSeparators( $number ) {
3213 return $this->formatNum( $number, true );
3214 }
3215
3220 public function parseFormattedNumber( $number ) {
3221 $s = $this->digitTransformTable();
3222 if ( $s ) {
3223 // eliminate empty array values such as ''. (T66347)
3224 $s = array_filter( $s );
3225 $number = strtr( $number, array_flip( $s ) );
3226 }
3227
3228 $s = $this->separatorTransformTable();
3229 if ( $s ) {
3230 // eliminate empty array values such as ''. (T66347)
3231 $s = array_filter( $s );
3232 $number = strtr( $number, array_flip( $s ) );
3233 }
3234
3235 $number = strtr( $number, [ ',' => '' ] );
3236 return $number;
3237 }
3238
3245 function commafy( $number ) {
3246 $digitGroupingPattern = $this->digitGroupingPattern();
3247 $minimumGroupingDigits = $this->minimumGroupingDigits();
3248 if ( $number === null ) {
3249 return '';
3250 }
3251
3252 if ( !$digitGroupingPattern || $digitGroupingPattern === "###,###,###" ) {
3253 // Default grouping is at thousands, use the same for ###,###,### pattern too.
3254 // In some languages it's conventional not to insert a thousands separator
3255 // in numbers that are four digits long (1000-9999).
3256 if ( $minimumGroupingDigits ) {
3257 // Number of '#' characters after last comma in the grouping pattern.
3258 // The pattern is hardcoded here, but this would vary for different patterns.
3259 $primaryGroupingSize = 3;
3260 // Maximum length of a number to suppress digit grouping for.
3261 $maximumLength = $minimumGroupingDigits + $primaryGroupingSize - 1;
3262 if ( preg_match( '/^\-?\d{1,' . $maximumLength . '}(\.\d+)?$/', $number ) ) {
3263 return $number;
3264 }
3265 }
3266 return strrev( (string)preg_replace( '/(\d{3})(?=\d)(?!\d*\.)/', '$1,', strrev( $number ) ) );
3267 } else {
3268 // Ref: http://cldr.unicode.org/translation/number-patterns
3269 $sign = "";
3270 if ( intval( $number ) < 0 ) {
3271 // For negative numbers apply the algorithm like positive number and add sign.
3272 $sign = "-";
3273 $number = substr( $number, 1 );
3274 }
3275 $integerPart = [];
3276 $decimalPart = [];
3277 $numMatches = preg_match_all( "/(#+)/", $digitGroupingPattern, $matches );
3278 preg_match( "/\d+/", $number, $integerPart );
3279 preg_match( "/\.\d*/", $number, $decimalPart );
3280 $groupedNumber = ( count( $decimalPart ) > 0 ) ? $decimalPart[0] : "";
3281 if ( $groupedNumber === $number ) {
3282 // the string does not have any number part. Eg: .12345
3283 return $sign . $groupedNumber;
3284 }
3285 $start = $end = ( $integerPart ) ? strlen( $integerPart[0] ) : 0;
3286 while ( $start > 0 ) {
3287 $match = $matches[0][$numMatches - 1];
3288 $matchLen = strlen( $match );
3289 $start = $end - $matchLen;
3290 if ( $start < 0 ) {
3291 $start = 0;
3292 }
3293 $groupedNumber = substr( $number, $start, $end - $start ) . $groupedNumber;
3294 $end = $start;
3295 if ( $numMatches > 1 ) {
3296 // use the last pattern for the rest of the number
3297 $numMatches--;
3298 }
3299 if ( $start > 0 ) {
3300 $groupedNumber = "," . $groupedNumber;
3301 }
3302 }
3303 return $sign . $groupedNumber;
3304 }
3305 }
3306
3311 return $this->localisationCache->getItem( $this->mCode, 'digitGroupingPattern' );
3312 }
3313
3318 return $this->localisationCache->getItem( $this->mCode, 'digitTransformTable' );
3319 }
3320
3325 return $this->localisationCache->getItem( $this->mCode, 'separatorTransformTable' );
3326 }
3327
3332 return $this->localisationCache->getItem( $this->mCode, 'minimumGroupingDigits' );
3333 }
3334
3343 public function listToText( array $list ) {
3344 $itemCount = count( $list );
3345 if ( $itemCount < 1 ) {
3346 return '';
3347 }
3348 $text = array_pop( $list );
3349 if ( $itemCount > 1 ) {
3350 $and = $this->msg( 'and' )->escaped();
3351 $space = $this->msg( 'word-separator' )->escaped();
3352 $comma = '';
3353 if ( $itemCount > 2 ) {
3354 $comma = $this->msg( 'comma-separator' )->escaped();
3355 }
3356 $text = implode( $comma, $list ) . $and . $space . $text;
3357 }
3358 return $text;
3359 }
3360
3367 function commaList( array $list ) {
3368 return implode(
3369 wfMessage( 'comma-separator' )->inLanguage( $this )->escaped(),
3370 $list
3371 );
3372 }
3373
3380 function semicolonList( array $list ) {
3381 return implode(
3382 wfMessage( 'semicolon-separator' )->inLanguage( $this )->escaped(),
3383 $list
3384 );
3385 }
3386
3392 function pipeList( array $list ) {
3393 return implode(
3394 wfMessage( 'pipe-separator' )->inLanguage( $this )->escaped(),
3395 $list
3396 );
3397 }
3398
3414 function truncateForDatabase( $string, $length, $ellipsis = '...', $adjustLength = true ) {
3415 return $this->truncateInternal(
3416 $string, $length, $ellipsis, $adjustLength, 'strlen', 'substr'
3417 );
3418 }
3419
3438 function truncateForVisual( $string, $length, $ellipsis = '...', $adjustLength = true ) {
3439 // Passing encoding to mb_strlen and mb_substr is optional.
3440 // Encoding defaults to mb_internal_encoding(), which is set to UTF-8 in Setup.php, so
3441 // explicit specification of encoding is skipped.
3442 // Note: Both multibyte methods are callables invoked in truncateInternal.
3443 return $this->truncateInternal(
3444 $string, $length, $ellipsis, $adjustLength, 'mb_strlen', 'mb_substr'
3445 );
3446 }
3447
3464 private function truncateInternal(
3465 $string, $length, $ellipsis, $adjustLength, callable $measureLength, callable $getSubstring
3466 ) {
3467 # Check if there is no need to truncate
3468 if ( $measureLength( $string ) <= abs( $length ) ) {
3469 return $string; // no need to truncate
3470 }
3471
3472 # Use the localized ellipsis character
3473 if ( $ellipsis == '...' ) {
3474 $ellipsis = wfMessage( 'ellipsis' )->inLanguage( $this )->escaped();
3475 }
3476 if ( $length == 0 ) {
3477 return $ellipsis; // convention
3478 }
3479
3480 $stringOriginal = $string;
3481 # If ellipsis length is >= $length then we can't apply $adjustLength
3482 if ( $adjustLength && $measureLength( $ellipsis ) >= abs( $length ) ) {
3483 $string = $ellipsis; // this can be slightly unexpected
3484 # Otherwise, truncate and add ellipsis...
3485 } else {
3486 $ellipsisLength = $adjustLength ? $measureLength( $ellipsis ) : 0;
3487 if ( $length > 0 ) {
3488 $length -= $ellipsisLength;
3489 $string = $getSubstring( $string, 0, $length ); // xyz...
3490 $string = $this->removeBadCharLast( $string );
3491 $string = rtrim( $string ) . $ellipsis;
3492 } else {
3493 $length += $ellipsisLength;
3494 $string = $getSubstring( $string, $length ); // ...xyz
3495 $string = $this->removeBadCharFirst( $string );
3496 $string = $ellipsis . ltrim( $string );
3497 }
3498 }
3499
3500 # Do not truncate if the ellipsis makes the string longer/equal (T24181).
3501 # This check is *not* redundant if $adjustLength, due to the single case where
3502 # LEN($ellipsis) > ABS($limit arg); $stringOriginal could be shorter than $string.
3503 if ( $measureLength( $string ) < $measureLength( $stringOriginal ) ) {
3504 return $string;
3505 } else {
3506 return $stringOriginal;
3507 }
3508 }
3509
3517 protected function removeBadCharLast( $string ) {
3518 if ( $string != '' ) {
3519 $char = ord( $string[strlen( $string ) - 1] );
3520 $m = [];
3521 if ( $char >= 0xc0 ) {
3522 # We got the first byte only of a multibyte char; remove it.
3523 $string = substr( $string, 0, -1 );
3524 } elseif ( $char >= 0x80 &&
3525 // Use the /s modifier (PCRE_DOTALL) so (.*) also matches newlines
3526 preg_match( '/^(.*)(?:[\xe0-\xef][\x80-\xbf]|' .
3527 '[\xf0-\xf7][\x80-\xbf]{1,2})$/s', $string, $m )
3528 ) {
3529 # We chopped in the middle of a character; remove it
3530 $string = $m[1];
3531 }
3532 }
3533 return $string;
3534 }
3535
3543 protected function removeBadCharFirst( $string ) {
3544 if ( $string != '' ) {
3545 $char = ord( $string[0] );
3546 if ( $char >= 0x80 && $char < 0xc0 ) {
3547 # We chopped in the middle of a character; remove the whole thing
3548 $string = preg_replace( '/^[\x80-\xbf]+/', '', $string );
3549 }
3550 }
3551 return $string;
3552 }
3553
3569 function truncateHtml( $text, $length, $ellipsis = '...' ) {
3570 # Use the localized ellipsis character
3571 if ( $ellipsis == '...' ) {
3572 $ellipsis = wfMessage( 'ellipsis' )->inLanguage( $this )->escaped();
3573 }
3574 # Check if there is clearly no need to truncate
3575 if ( $length <= 0 ) {
3576 return $ellipsis; // no text shown, nothing to format (convention)
3577 } elseif ( strlen( $text ) <= $length ) {
3578 return $text; // string short enough even *with* HTML (short-circuit)
3579 }
3580
3581 $dispLen = 0; // innerHTML legth so far
3582 $testingEllipsis = false; // checking if ellipses will make string longer/equal?
3583 $tagType = 0; // 0-open, 1-close
3584 $bracketState = 0; // 1-tag start, 2-tag name, 0-neither
3585 $entityState = 0; // 0-not entity, 1-entity
3586 $tag = $ret = ''; // accumulated tag name, accumulated result string
3587 $openTags = []; // open tag stack
3588 $maybeState = null; // possible truncation state
3589
3590 $textLen = strlen( $text );
3591 $neLength = max( 0, $length - strlen( $ellipsis ) ); // non-ellipsis len if truncated
3592 for ( $pos = 0; true; ++$pos ) {
3593 # Consider truncation once the display length has reached the maximim.
3594 # We check if $dispLen > 0 to grab tags for the $neLength = 0 case.
3595 # Check that we're not in the middle of a bracket/entity...
3596 if ( $dispLen && $dispLen >= $neLength && $bracketState == 0 && !$entityState ) {
3597 if ( !$testingEllipsis ) {
3598 $testingEllipsis = true;
3599 # Save where we are; we will truncate here unless there turn out to
3600 # be so few remaining characters that truncation is not necessary.
3601 if ( !$maybeState ) { // already saved? ($neLength = 0 case)
3602 $maybeState = [ $ret, $openTags ]; // save state
3603 }
3604 } elseif ( $dispLen > $length && $dispLen > strlen( $ellipsis ) ) {
3605 # String in fact does need truncation, the truncation point was OK.
3606 // @phan-suppress-next-line PhanTypeInvalidExpressionArrayDestructuring
3607 list( $ret, $openTags ) = $maybeState; // reload state
3608 $ret = $this->removeBadCharLast( $ret ); // multi-byte char fix
3609 $ret .= $ellipsis; // add ellipsis
3610 break;
3611 }
3612 }
3613 if ( $pos >= $textLen ) {
3614 break; // extra iteration just for above checks
3615 }
3616
3617 # Read the next char...
3618 $ch = $text[$pos];
3619 $lastCh = $pos ? $text[$pos - 1] : '';
3620 $ret .= $ch; // add to result string
3621 if ( $ch == '<' ) {
3622 $this->truncate_endBracket( $tag, $tagType, $lastCh, $openTags ); // for bad HTML
3623 $entityState = 0; // for bad HTML
3624 $bracketState = 1; // tag started (checking for backslash)
3625 } elseif ( $ch == '>' ) {
3626 $this->truncate_endBracket( $tag, $tagType, $lastCh, $openTags );
3627 $entityState = 0; // for bad HTML
3628 $bracketState = 0; // out of brackets
3629 } elseif ( $bracketState == 1 ) {
3630 if ( $ch == '/' ) {
3631 $tagType = 1; // close tag (e.g. "</span>")
3632 } else {
3633 $tagType = 0; // open tag (e.g. "<span>")
3634 $tag .= $ch;
3635 }
3636 $bracketState = 2; // building tag name
3637 } elseif ( $bracketState == 2 ) {
3638 if ( $ch != ' ' ) {
3639 $tag .= $ch;
3640 } else {
3641 // Name found (e.g. "<a href=..."), add on tag attributes...
3642 $pos += $this->truncate_skip( $ret, $text, "<>", $pos + 1 );
3643 }
3644 } elseif ( $bracketState == 0 ) {
3645 if ( $entityState ) {
3646 if ( $ch == ';' ) {
3647 $entityState = 0;
3648 $dispLen++; // entity is one displayed char
3649 }
3650 } else {
3651 if ( $neLength == 0 && !$maybeState ) {
3652 // Save state without $ch. We want to *hit* the first
3653 // display char (to get tags) but not *use* it if truncating.
3654 $maybeState = [ substr( $ret, 0, -1 ), $openTags ];
3655 }
3656 if ( $ch == '&' ) {
3657 $entityState = 1; // entity found, (e.g. "&#160;")
3658 } else {
3659 $dispLen++; // this char is displayed
3660 // Add the next $max display text chars after this in one swoop...
3661 $max = ( $testingEllipsis ? $length : $neLength ) - $dispLen;
3662 $skipped = $this->truncate_skip( $ret, $text, "<>&", $pos + 1, $max );
3663 $dispLen += $skipped;
3664 $pos += $skipped;
3665 }
3666 }
3667 }
3668 }
3669 // Close the last tag if left unclosed by bad HTML
3670 $this->truncate_endBracket( $tag, $text[$textLen - 1], $tagType, $openTags );
3671 while ( count( $openTags ) > 0 ) {
3672 $ret .= '</' . array_pop( $openTags ) . '>'; // close open tags
3673 }
3674 return $ret;
3675 }
3676
3688 private function truncate_skip( &$ret, $text, $search, $start, $len = null ) {
3689 if ( $len === null ) {
3690 $len = -1; // -1 means "no limit" for strcspn
3691 } elseif ( $len < 0 ) {
3692 $len = 0; // sanity
3693 }
3694 $skipCount = 0;
3695 if ( $start < strlen( $text ) ) {
3696 $skipCount = strcspn( $text, $search, $start, $len );
3697 $ret .= substr( $text, $start, $skipCount );
3698 }
3699 return $skipCount;
3700 }
3701
3711 private function truncate_endBracket( &$tag, $tagType, $lastCh, &$openTags ) {
3712 $tag = ltrim( $tag );
3713 if ( $tag != '' ) {
3714 if ( $tagType == 0 && $lastCh != '/' ) {
3715 $openTags[] = $tag; // tag opened (didn't close itself)
3716 } elseif ( $tagType == 1 ) {
3717 if ( $openTags && $tag == $openTags[count( $openTags ) - 1] ) {
3718 array_pop( $openTags ); // tag closed
3719 }
3720 }
3721 $tag = '';
3722 }
3723 }
3724
3733 function convertGrammar( $word, $case ) {
3734 global $wgGrammarForms;
3735 if ( isset( $wgGrammarForms[$this->getCode()][$case][$word] ) ) {
3736 return $wgGrammarForms[$this->getCode()][$case][$word];
3737 }
3738
3739 $grammarTransformations = $this->getGrammarTransformations();
3740
3741 if ( isset( $grammarTransformations[$case] ) ) {
3742 $forms = $grammarTransformations[$case];
3743
3744 // Some names of grammar rules are aliases for other rules.
3745 // In such cases the value is a string rather than object,
3746 // so load the actual rules.
3747 if ( is_string( $forms ) ) {
3748 $forms = $grammarTransformations[$forms];
3749 }
3750
3751 foreach ( array_values( $forms ) as $rule ) {
3752 $form = $rule[0];
3753
3754 if ( $form === '@metadata' ) {
3755 continue;
3756 }
3757
3758 $replacement = $rule[1];
3759
3760 $regex = '/' . addcslashes( $form, '/' ) . '/u';
3761 $patternMatches = preg_match( $regex, $word );
3762
3763 if ( $patternMatches === false ) {
3765 'An error occurred while processing grammar. ' .
3766 "Word: '$word'. Regex: /$form/."
3767 );
3768 } elseif ( $patternMatches === 1 ) {
3769 $word = preg_replace( $regex, $replacement, $word );
3770
3771 break;
3772 }
3773 }
3774 }
3775
3776 return $word;
3777 }
3778
3784 function getGrammarForms() {
3785 global $wgGrammarForms;
3786 if ( isset( $wgGrammarForms[$this->getCode()] )
3787 && is_array( $wgGrammarForms[$this->getCode()] )
3788 ) {
3789 return $wgGrammarForms[$this->getCode()];
3790 }
3791
3792 return [];
3793 }
3794
3804 public function getGrammarTransformations() {
3805 $languageCode = $this->getCode();
3806
3807 if ( self::$grammarTransformations === null ) {
3808 self::$grammarTransformations = new MapCacheLRU( 10 );
3809 }
3810
3811 if ( self::$grammarTransformations->has( $languageCode ) ) {
3812 return self::$grammarTransformations->get( $languageCode );
3813 }
3814
3815 $data = [];
3816
3817 $grammarDataFile = __DIR__ . "/data/grammarTransformations/$languageCode.json";
3818 if ( is_readable( $grammarDataFile ) ) {
3819 $data = FormatJson::decode(
3820 file_get_contents( $grammarDataFile ),
3821 true
3822 );
3823
3824 if ( $data === null ) {
3825 throw new MWException( "Invalid grammar data for \"$languageCode\"." );
3826 }
3827
3828 self::$grammarTransformations->set( $languageCode, $data );
3829 }
3830
3831 return $data;
3832 }
3833
3853 function gender( $gender, $forms ) {
3854 if ( !count( $forms ) ) {
3855 return '';
3856 }
3857 $forms = $this->preConvertPlural( $forms, 2 );
3858 if ( $gender === 'male' ) {
3859 return $forms[0];
3860 }
3861 if ( $gender === 'female' ) {
3862 return $forms[1];
3863 }
3864 return $forms[2] ?? $forms[0];
3865 }
3866
3882 function convertPlural( $count, $forms ) {
3883 // Handle explicit n=pluralform cases
3884 $forms = $this->handleExplicitPluralForms( $count, $forms );
3885 if ( is_string( $forms ) ) {
3886 return $forms;
3887 }
3888 if ( !count( $forms ) ) {
3889 return '';
3890 }
3891
3892 $pluralForm = $this->getPluralRuleIndexNumber( $count );
3893 $pluralForm = min( $pluralForm, count( $forms ) - 1 );
3894 return $forms[$pluralForm];
3895 }
3896
3912 protected function handleExplicitPluralForms( $count, array $forms ) {
3913 foreach ( $forms as $index => $form ) {
3914 if ( preg_match( '/\d+=/i', $form ) ) {
3915 $pos = strpos( $form, '=' );
3916 if ( substr( $form, 0, $pos ) === (string)$count ) {
3917 return substr( $form, $pos + 1 );
3918 }
3919 unset( $forms[$index] );
3920 }
3921 }
3922 return array_values( $forms );
3923 }
3924
3933 protected function preConvertPlural( /* Array */ $forms, $count ) {
3934 return array_pad( $forms, $count, end( $forms ) );
3935 }
3936
3953 public function embedBidi( $text = '' ) {
3954 $dir = self::strongDirFromContent( $text );
3955 if ( $dir === 'ltr' ) {
3956 // Wrap in LEFT-TO-RIGHT EMBEDDING ... POP DIRECTIONAL FORMATTING
3957 return self::$lre . $text . self::$pdf;
3958 }
3959 if ( $dir === 'rtl' ) {
3960 // Wrap in RIGHT-TO-LEFT EMBEDDING ... POP DIRECTIONAL FORMATTING
3961 return self::$rle . $text . self::$pdf;
3962 }
3963 // No strong directionality: do not wrap
3964 return $text;
3965 }
3966
3980 function translateBlockExpiry( $str, User $user = null, $now = 0 ) {
3981 $duration = SpecialBlock::getSuggestedDurations( $this );
3982 foreach ( $duration as $show => $value ) {
3983 if ( strcmp( $str, $value ) == 0 ) {
3984 return htmlspecialchars( trim( $show ) );
3985 }
3986 }
3987
3988 if ( wfIsInfinity( $str ) ) {
3989 foreach ( $duration as $show => $value ) {
3990 if ( wfIsInfinity( $value ) ) {
3991 return htmlspecialchars( trim( $show ) );
3992 }
3993 }
3994 }
3995
3996 // If all else fails, return a standard duration or timestamp description.
3997 $time = strtotime( $str, $now );
3998 if ( $time === false ) { // Unknown format. Return it as-is in case.
3999 return $str;
4000 } elseif ( $time !== strtotime( $str, $now + 1 ) ) { // It's a relative timestamp.
4001 // The result differs based on current time, so the difference
4002 // is a fixed duration length.
4003 return $this->formatDuration( $time - $now );
4004 } else { // It's an absolute timestamp.
4005 if ( $time === 0 ) {
4006 // wfTimestamp() handles 0 as current time instead of epoch.
4007 $time = '19700101000000';
4008 }
4009 if ( $user ) {
4010 return $this->userTimeAndDate( $time, $user );
4011 }
4012 return $this->timeanddate( $time );
4013 }
4014 }
4015
4023 public function segmentForDiff( $text ) {
4024 return $text;
4025 }
4026
4033 public function unsegmentForDiff( $text ) {
4034 return $text;
4035 }
4036
4043 public function getConverter() {
4044 return $this->mConverter;
4045 }
4046
4055 public function autoConvert( $text, $variant = false ) {
4056 return $this->mConverter->autoConvert( $text, $variant );
4057 }
4058
4065 public function autoConvertToAllVariants( $text ) {
4066 return $this->mConverter->autoConvertToAllVariants( $text );
4067 }
4068
4080 public function convert( $text ) {
4081 return $this->mConverter->convert( $text );
4082 }
4083
4090 public function convertTitle( $title ) {
4091 return $this->mConverter->convertTitle( $title );
4092 }
4093
4102 public function convertNamespace( $ns, $variant = null ) {
4103 return $this->mConverter->convertNamespace( $ns, $variant );
4104 }
4105
4111 public function hasVariants() {
4112 return count( $this->getVariants() ) > 1;
4113 }
4114
4125 public function hasVariant( $variant ) {
4126 return $variant && ( $variant === $this->mConverter->validateVariant( $variant ) );
4127 }
4128
4135 public function convertHtml( $text ) {
4136 return htmlspecialchars( $this->convert( $text ) );
4137 }
4138
4143 public function convertCategoryKey( $key ) {
4144 return $this->mConverter->convertCategoryKey( $key );
4145 }
4146
4153 public function getVariants() {
4154 return $this->mConverter->getVariants();
4155 }
4156
4160 public function getPreferredVariant() {
4161 return $this->mConverter->getPreferredVariant();
4162 }
4163
4167 public function getDefaultVariant() {
4168 return $this->mConverter->getDefaultVariant();
4169 }
4170
4174 public function getURLVariant() {
4175 return $this->mConverter->getURLVariant();
4176 }
4177
4190 public function findVariantLink( &$link, &$nt, $ignoreOtherCond = false ) {
4191 $this->mConverter->findVariantLink( $link, $nt, $ignoreOtherCond );
4192 }
4193
4201 return $this->mConverter->getExtraHashOptions();
4202 }
4203
4211 public function getParsedTitle() {
4212 return $this->mConverter->getParsedTitle();
4213 }
4214
4222 $this->mConverter->updateConversionTable( $title );
4223 }
4224
4231 public function linkTrail() {
4232 return $this->localisationCache->getItem( $this->mCode, 'linkTrail' );
4233 }
4234
4241 public function linkPrefixCharset() {
4242 return $this->localisationCache->getItem( $this->mCode, 'linkPrefixCharset' );
4243 }
4244
4252 public function getParentLanguage() {
4253 if ( $this->mParentLanguage !== false ) {
4254 return $this->mParentLanguage;
4255 }
4256
4257 $code = explode( '-', $this->getCode() )[0];
4258 if ( !in_array( $code, LanguageConverter::$languagesWithVariants ) ) {
4259 $this->mParentLanguage = null;
4260 return null;
4261 }
4262 $lang = self::factory( $code );
4263 if ( !$lang->hasVariant( $this->getCode() ) ) {
4264 $this->mParentLanguage = null;
4265 return null;
4266 }
4267
4268 $this->mParentLanguage = $lang;
4269 return $lang;
4270 }
4271
4279 public function equals( Language $lang ) {
4280 return $lang === $this || $lang->getCode() === $this->mCode;
4281 }
4282
4291 public function getCode() {
4292 return $this->mCode;
4293 }
4294
4305 public function getHtmlCode() {
4306 if ( is_null( $this->mHtmlCode ) ) {
4307 $this->mHtmlCode = LanguageCode::bcp47( $this->getCode() );
4308 }
4309 return $this->mHtmlCode;
4310 }
4311
4319 public static function getCodeFromFileName( $filename, $prefix = 'Language', $suffix = '.php' ) {
4320 $m = null;
4321 preg_match( '/' . preg_quote( $prefix, '/' ) . '([A-Z][a-z_]+)' .
4322 preg_quote( $suffix, '/' ) . '/', $filename, $m );
4323 if ( !count( $m ) ) {
4324 return false;
4325 }
4326 return str_replace( '_', '-', strtolower( $m[1] ) );
4327 }
4328
4334 public static function classFromCode( $code, $fallback = true ) {
4335 if ( $fallback && $code == 'en' ) {
4336 return 'Language';
4337 } else {
4338 return 'Language' . str_replace( '-', '_', ucfirst( $code ) );
4339 }
4340 }
4341
4352 public static function getFileName( $prefix, $code, $suffix = '.php' ) {
4353 return MediaWikiServices::getInstance()->getLanguageNameUtils()
4354 ->getFileName( $prefix, $code, $suffix );
4355 }
4356
4362 public static function getMessagesFileName( $code ) {
4363 return MediaWikiServices::getInstance()->getLanguageNameUtils()
4364 ->getMessagesFileName( $code );
4365 }
4366
4374 public static function getJsonMessagesFileName( $code ) {
4375 return MediaWikiServices::getInstance()->getLanguageNameUtils()
4376 ->getJsonMessagesFileName( $code );
4377 }
4378
4386 public static function getFallbackFor( $code ) {
4387 $fallbacks = self::getFallbacksFor( $code );
4388 if ( $fallbacks ) {
4389 return $fallbacks[0];
4390 }
4391 return false;
4392 }
4393
4404 public static function getFallbacksFor( $code, $mode = self::MESSAGES_FALLBACKS ) {
4405 if ( $code === 'en' || !self::isValidBuiltInCode( $code ) ) {
4406 return [];
4407 }
4408 switch ( $mode ) {
4409 case self::MESSAGES_FALLBACKS:
4410 // For unknown languages, fallbackSequence returns an empty array,
4411 // hardcode fallback to 'en' in that case as English messages are
4412 // always defined.
4413 return self::getLocalisationCache()->getItem( $code, 'fallbackSequence' ) ?: [ 'en' ];
4414 case self::STRICT_FALLBACKS:
4415 // Use this mode when you don't want to fallback to English unless
4416 // explicitly defined, for example when you have language-variant icons
4417 // and an international language-independent fallback.
4418 return self::getLocalisationCache()->getItem( $code, 'originalFallbackSequence' );
4419 default:
4420 throw new MWException( "Invalid fallback mode \"$mode\"" );
4421 }
4422 }
4423
4432 public static function getFallbacksIncludingSiteLanguage( $code ) {
4433 global $wgLanguageCode;
4434
4435 // Usually, we will only store a tiny number of fallback chains, so we
4436 // keep them in static memory.
4437 $cacheKey = "{$code}-{$wgLanguageCode}";
4438
4439 if ( !array_key_exists( $cacheKey, self::$fallbackLanguageCache ) ) {
4440 $fallbacks = self::getFallbacksFor( $code );
4441
4442 // Append the site's fallback chain, including the site language itself
4443 $siteFallbacks = self::getFallbacksFor( $wgLanguageCode );
4444 array_unshift( $siteFallbacks, $wgLanguageCode );
4445
4446 // Eliminate any languages already included in the chain
4447 $siteFallbacks = array_diff( $siteFallbacks, $fallbacks );
4448
4449 self::$fallbackLanguageCache[$cacheKey] = [ $fallbacks, $siteFallbacks ];
4450 }
4451 return self::$fallbackLanguageCache[$cacheKey];
4452 }
4453
4463 public static function getMessagesFor( $code ) {
4464 return self::getLocalisationCache()->getItem( $code, 'messages' );
4465 }
4466
4475 public static function getMessageFor( $key, $code ) {
4476 return self::getLocalisationCache()->getSubitem( $code, 'messages', $key );
4477 }
4478
4487 public static function getMessageKeysFor( $code ) {
4488 return self::getLocalisationCache()->getSubitemList( $code, 'messages' );
4489 }
4490
4495 function fixVariableInNamespace( $talk ) {
4496 if ( strpos( $talk, '$1' ) === false ) {
4497 return $talk;
4498 }
4499
4500 global $wgMetaNamespace;
4501 $talk = str_replace( '$1', $wgMetaNamespace, $talk );
4502
4503 # Allow grammar transformations
4504 # Allowing full message-style parsing would make simple requests
4505 # such as action=raw much more expensive than they need to be.
4506 # This will hopefully cover most cases.
4507 $talk = preg_replace_callback( '/{{grammar:(.*?)\|(.*?)}}/i',
4508 [ $this, 'replaceGrammarInNamespace' ], $talk );
4509 return str_replace( ' ', '_', $talk );
4510 }
4511
4517 return $this->convertGrammar( trim( $m[2] ), trim( $m[1] ) );
4518 }
4519
4530 public function formatExpiry( $expiry, $format = true, $infinity = 'infinity' ) {
4531 static $dbInfinity;
4532 if ( $dbInfinity === null ) {
4533 $dbInfinity = wfGetDB( DB_REPLICA )->getInfinity();
4534 }
4535
4536 if ( $expiry == '' || $expiry === 'infinity' || $expiry == $dbInfinity ) {
4537 return $format === true
4538 ? $this->getMessageFromDB( 'infiniteblock' )
4539 : $infinity;
4540 } else {
4541 return $format === true
4542 ? $this->timeanddate( $expiry, /* User preference timezone */ true )
4543 : wfTimestamp( $format, $expiry );
4544 }
4545 }
4546
4561 function formatTimePeriod( $seconds, $format = [] ) {
4562 if ( !is_array( $format ) ) {
4563 $format = [ 'avoid' => $format ]; // For backwards compatibility
4564 }
4565 if ( !isset( $format['avoid'] ) ) {
4566 $format['avoid'] = false;
4567 }
4568 if ( !isset( $format['noabbrevs'] ) ) {
4569 $format['noabbrevs'] = false;
4570 }
4571 $secondsMsg = wfMessage(
4572 $format['noabbrevs'] ? 'seconds' : 'seconds-abbrev' )->inLanguage( $this );
4573 $minutesMsg = wfMessage(
4574 $format['noabbrevs'] ? 'minutes' : 'minutes-abbrev' )->inLanguage( $this );
4575 $hoursMsg = wfMessage(
4576 $format['noabbrevs'] ? 'hours' : 'hours-abbrev' )->inLanguage( $this );
4577 $daysMsg = wfMessage(
4578 $format['noabbrevs'] ? 'days' : 'days-abbrev' )->inLanguage( $this );
4579
4580 if ( round( $seconds * 10 ) < 100 ) {
4581 $s = $this->formatNum( sprintf( "%.1f", round( $seconds * 10 ) / 10 ) );
4582 $s = $secondsMsg->params( $s )->text();
4583 } elseif ( round( $seconds ) < 60 ) {
4584 $s = $this->formatNum( round( $seconds ) );
4585 $s = $secondsMsg->params( $s )->text();
4586 } elseif ( round( $seconds ) < 3600 ) {
4587 $minutes = floor( $seconds / 60 );
4588 $secondsPart = round( fmod( $seconds, 60 ) );
4589 if ( $secondsPart == 60 ) {
4590 $secondsPart = 0;
4591 $minutes++;
4592 }
4593 $s = $minutesMsg->params( $this->formatNum( $minutes ) )->text();
4594 $s .= ' ';
4595 $s .= $secondsMsg->params( $this->formatNum( $secondsPart ) )->text();
4596 } elseif ( round( $seconds ) <= 2 * 86400 ) {
4597 $hours = floor( $seconds / 3600 );
4598 $minutes = floor( ( $seconds - $hours * 3600 ) / 60 );
4599 $secondsPart = round( $seconds - $hours * 3600 - $minutes * 60 );
4600 if ( $secondsPart == 60 ) {
4601 $secondsPart = 0;
4602 $minutes++;
4603 }
4604 if ( $minutes == 60 ) {
4605 $minutes = 0;
4606 $hours++;
4607 }
4608 $s = $hoursMsg->params( $this->formatNum( $hours ) )->text();
4609 $s .= ' ';
4610 $s .= $minutesMsg->params( $this->formatNum( $minutes ) )->text();
4611 if ( !in_array( $format['avoid'], [ 'avoidseconds', 'avoidminutes', 'avoidhours' ] ) ) {
4612 $s .= ' ' . $secondsMsg->params( $this->formatNum( $secondsPart ) )->text();
4613 }
4614 } else {
4615 $days = floor( $seconds / 86400 );
4616 if ( $format['avoid'] === 'avoidhours' ) {
4617 $hours = round( ( $seconds - $days * 86400 ) / 3600 );
4618 if ( $hours == 24 ) {
4619 $hours = 0;
4620 $days++;
4621 }
4622 $s = $daysMsg->params( $this->formatNum( $days ) )->text();
4623 } elseif ( $format['avoid'] === 'avoidminutes' ) {
4624 $hours = round( ( $seconds - $days * 86400 ) / 3600 );
4625 if ( $hours == 24 ) {
4626 $hours = 0;
4627 $days++;
4628 }
4629 $s = $daysMsg->params( $this->formatNum( $days ) )->text();
4630 $s .= ' ';
4631 $s .= $hoursMsg->params( $this->formatNum( $hours ) )->text();
4632 } elseif ( $format['avoid'] === 'avoidseconds' ) {
4633 $hours = floor( ( $seconds - $days * 86400 ) / 3600 );
4634 $minutes = round( ( $seconds - $days * 86400 - $hours * 3600 ) / 60 );
4635 if ( $minutes == 60 ) {
4636 $minutes = 0;
4637 $hours++;
4638 }
4639 if ( $hours == 24 ) {
4640 $hours = 0;
4641 $days++;
4642 }
4643 $s = $daysMsg->params( $this->formatNum( $days ) )->text();
4644 $s .= ' ';
4645 $s .= $hoursMsg->params( $this->formatNum( $hours ) )->text();
4646 $s .= ' ';
4647 $s .= $minutesMsg->params( $this->formatNum( $minutes ) )->text();
4648 } else {
4649 $s = $daysMsg->params( $this->formatNum( $days ) )->text();
4650 $s .= ' ';
4651 $s .= $this->formatTimePeriod( $seconds - $days * 86400, $format );
4652 }
4653 }
4654 return $s;
4655 }
4656
4668 function formatBitrate( $bps ) {
4669 return $this->formatComputingNumbers( $bps, 1000, "bitrate-$1bits" );
4670 }
4671
4678 function formatComputingNumbers( $size, $boundary, $messageKey ) {
4679 if ( $size <= 0 ) {
4680 return str_replace( '$1', $this->formatNum( $size ),
4681 $this->getMessageFromDB( str_replace( '$1', '', $messageKey ) )
4682 );
4683 }
4684 $sizes = [ '', 'kilo', 'mega', 'giga', 'tera', 'peta', 'exa', 'zeta', 'yotta' ];
4685 $index = 0;
4686
4687 $maxIndex = count( $sizes ) - 1;
4688 while ( $size >= $boundary && $index < $maxIndex ) {
4689 $index++;
4690 $size /= $boundary;
4691 }
4692
4693 // For small sizes no decimal places necessary
4694 $round = 0;
4695 if ( $index > 1 ) {
4696 // For MB and bigger two decimal places are smarter
4697 $round = 2;
4698 }
4699 $msg = str_replace( '$1', $sizes[$index], $messageKey );
4700
4701 $size = round( $size, $round );
4702 $text = $this->getMessageFromDB( $msg );
4703 return str_replace( '$1', $this->formatNum( $size ), $text );
4704 }
4705
4716 function formatSize( $size ) {
4717 return $this->formatComputingNumbers( $size, 1024, "size-$1bytes" );
4718 }
4719
4729 function specialList( $page, $details, $oppositedm = true ) {
4730 if ( !$details ) {
4731 return $page;
4732 }
4733
4734 $dirmark = ( $oppositedm ? $this->getDirMark( true ) : '' ) . $this->getDirMark();
4735 return $page .
4736 $dirmark .
4737 $this->msg( 'word-separator' )->escaped() .
4738 $this->msg( 'parentheses' )->rawParams( $details )->escaped();
4739 }
4740
4753 public function viewPrevNext( Title $title, $offset, $limit,
4754 array $query = [], $atend = false
4755 ) {
4756 wfDeprecated( __METHOD__, '1.34' );
4757 // @todo FIXME: Why on earth this needs one message for the text and another one for tooltip?
4758
4759 # Make 'previous' link
4760 $prev = wfMessage( 'prevn' )->inLanguage( $this )->title( $title )->numParams( $limit )->text();
4761 if ( $offset > 0 ) {
4762 $plink = $this->numLink( $title, max( $offset - $limit, 0 ), $limit,
4763 $query, $prev, 'prevn-title', 'mw-prevlink' );
4764 } else {
4765 $plink = htmlspecialchars( $prev );
4766 }
4767
4768 # Make 'next' link
4769 $next = wfMessage( 'nextn' )->inLanguage( $this )->title( $title )->numParams( $limit )->text();
4770 if ( $atend ) {
4771 $nlink = htmlspecialchars( $next );
4772 } else {
4773 $nlink = $this->numLink( $title, $offset + $limit, $limit,
4774 $query, $next, 'nextn-title', 'mw-nextlink' );
4775 }
4776
4777 # Make links to set number of items per page
4778 $numLinks = [];
4779 foreach ( [ 20, 50, 100, 250, 500 ] as $num ) {
4780 $numLinks[] = $this->numLink( $title, $offset, $num,
4781 $query, $this->formatNum( $num ), 'shown-title', 'mw-numlink' );
4782 }
4783
4784 return wfMessage( 'viewprevnext' )->inLanguage( $this )->title( $title
4785 )->rawParams( $plink, $nlink, $this->pipeList( $numLinks ) )->escaped();
4786 }
4787
4800 private function numLink( Title $title, $offset, $limit, array $query, $link,
4801 $tooltipMsg, $class
4802 ) {
4803 $query = [ 'limit' => $limit, 'offset' => $offset ] + $query;
4804 $tooltip = wfMessage( $tooltipMsg )->inLanguage( $this )->title( $title )
4805 ->numParams( $limit )->text();
4806
4807 return Html::element( 'a', [ 'href' => $title->getLocalURL( $query ),
4808 'title' => $tooltip, 'class' => $class ], $link );
4809 }
4810
4816 public function getConvRuleTitle() {
4817 return $this->mConverter->getConvRuleTitle();
4818 }
4819
4825 public function getCompiledPluralRules() {
4826 $pluralRules =
4827 $this->localisationCache->getItem( strtolower( $this->mCode ), 'compiledPluralRules' );
4828 $fallbacks = self::getFallbacksFor( $this->mCode );
4829 if ( !$pluralRules ) {
4830 foreach ( $fallbacks as $fallbackCode ) {
4831 $pluralRules = $this->localisationCache
4832 ->getItem( strtolower( $fallbackCode ), 'compiledPluralRules' );
4833 if ( $pluralRules ) {
4834 break;
4835 }
4836 }
4837 }
4838 return $pluralRules;
4839 }
4840
4846 public function getPluralRules() {
4847 $pluralRules =
4848 $this->localisationCache->getItem( strtolower( $this->mCode ), 'pluralRules' );
4849 $fallbacks = self::getFallbacksFor( $this->mCode );
4850 if ( !$pluralRules ) {
4851 foreach ( $fallbacks as $fallbackCode ) {
4852 $pluralRules = $this->localisationCache
4853 ->getItem( strtolower( $fallbackCode ), 'pluralRules' );
4854 if ( $pluralRules ) {
4855 break;
4856 }
4857 }
4858 }
4859 return $pluralRules;
4860 }
4861
4867 public function getPluralRuleTypes() {
4868 $pluralRuleTypes =
4869 $this->localisationCache->getItem( strtolower( $this->mCode ), 'pluralRuleTypes' );
4870 $fallbacks = self::getFallbacksFor( $this->mCode );
4871 if ( !$pluralRuleTypes ) {
4872 foreach ( $fallbacks as $fallbackCode ) {
4873 $pluralRuleTypes = $this->localisationCache
4874 ->getItem( strtolower( $fallbackCode ), 'pluralRuleTypes' );
4875 if ( $pluralRuleTypes ) {
4876 break;
4877 }
4878 }
4879 }
4880 return $pluralRuleTypes;
4881 }
4882
4888 public function getPluralRuleIndexNumber( $number ) {
4889 $pluralRules = $this->getCompiledPluralRules();
4890 $form = Evaluator::evaluateCompiled( $number, $pluralRules );
4891 return $form;
4892 }
4893
4902 public function getPluralRuleType( $number ) {
4903 $index = $this->getPluralRuleIndexNumber( $number );
4904 $pluralRuleTypes = $this->getPluralRuleTypes();
4905 return $pluralRuleTypes[$index] ?? 'other';
4906 }
4907}
$wgLanguageCode
Site language code.
$wgExtraGenderNamespaces
Same as above, but for namespaces with gender distinction.
$wgTranslateNumerals
For Hindi and Arabic use local numerals instead of Western style (0-9) numerals in interface.
array $wgOverrideUcfirstCharacters
List of Unicode characters for which capitalization is overridden in Language::ucfirst.
$wgGrammarForms
Some languages need different word forms, usually for different cases.
$wgExtraNamespaces
Additional namespaces.
$wgAllUnicodeFixes
Set this to always convert certain Unicode sequences to modern ones regardless of the content languag...
$wgDummyLanguageCodes
Functionally the same as $wgExtraLanguageCodes, but deprecated.
$wgNamespaceAliases
Namespace aliases.
$wgMetaNamespace
Name of the project namespace.
$wgMetaNamespaceTalk
Name of the project talk namespace.
$wgLocalTZoffset
Set an offset from UTC in minutes to use for the default timezone setting for anonymous users and new...
$wgLangObjCacheSize
Language cache size, or really how many languages can we handle simultaneously without degrading to c...
wfWarn( $msg, $callerOffset=1, $level=E_USER_NOTICE)
Send a warning either to the debug log or in a PHP error depending on $wgDevelopmentWarnings.
wfGetDB( $db, $groups=[], $wiki=false)
Get a Database object.
wfLogWarning( $msg, $callerOffset=1, $level=E_USER_WARNING)
Send a warning as a PHP error and the debug log.
wfIsInfinity( $str)
Determine input string is represents as infinity.
wfTimestamp( $outputtype=TS_UNIX, $ts=0)
Get a timestamp string in one of various formats.
wfMessage( $key,... $params)
This is the function for getting translated interface messages.
wfDeprecated( $function, $version=false, $component=false, $callerOffset=2)
Throws a warning that $function is deprecated.
$fallback
$namespaceAliases
$namespaceNames
$digitGroupingPattern
$minimumGroupingDigits
$IP
Definition WebStart.php:41
A fake language variant converter.
Base class for language conversion.
Internationalisation code.
Definition Language.php:37
const STRICT_FALLBACKS
Return a strict fallback chain in getFallbacksFor.
Definition Language.php:101
const MESSAGES_FALLBACKS
Return a fallback chain for messages in getFallbacksFor.
Definition Language.php:95
hasVariants()
Check if this is a language with variants.
initContLang()
Hook which will be called if this is the content language.
Definition Language.php:457
static $mWeekdayMsgs
Definition Language.php:104
getLocalNsIndex( $text)
Get a namespace key by value, case insensitive.
Definition Language.php:632
date( $ts, $adj=false, $format=true, $timecorrection=false)
linkTrail()
A regular expression to match legal word-trailing characters which should be merged onto a link of th...
static $strongDirRegex
Directionality test regex for embedBidi().
Definition Language.php:208
formatExpiry( $expiry, $format=true, $infinity='infinity')
Decode an expiry (block, protection, etc) which has come from the DB.
getPluralRuleTypes()
Get the plural rule types for the language.
static tsToIranian( $ts)
Algorithm by Roozbeh Pournader and Mohammad Toossi to convert Gregorian dates to Iranian dates.
getDefaultVariant()
convertCategoryKey( $key)
array null $namespaceNames
Definition Language.php:75
static getMessageKeysFor( $code)
Get all message keys for a given language.
static isWellFormedLanguageTag( $code, $lenient=false)
Returns true if a language code string is a well-formed language tag according to RFC 5646.
Definition Language.php:329
getVariantname( $code, $usemsg=true)
short names for language variants used for language conversion links.
Definition Language.php:741
static $mHebrewCalendarMonthGenMsgs
Definition Language.php:143
static isValidBuiltInCode( $code)
Returns true if a language code is of a valid form for the purposes of internal customisation of Medi...
Definition Language.php:400
static getMessageFor( $key, $code)
Get a message for a given language.
gender( $gender, $forms)
Provides an alternative text depending on specified gender.
static tsToYear( $ts, $cName)
Algorithm to convert Gregorian dates to Thai solar dates, Minguo dates or Minguo dates.
truncateForVisual( $string, $length, $ellipsis='...', $adjustLength=true)
Truncate a string to a specified number of characters, appending an optional string (e....
ucwords( $str)
formatNum( $number, $nocommafy=false)
Normally we output all numbers in plain en_US style, that is 293,291.235 for twohundredninetythreetho...
static clearCaches()
Intended for tests that may change configuration in a way that invalidates caches.
Definition Language.php:287
static $pdf
Definition Language.php:193
digitGroupingPattern()
static array $fallbackLanguageCache
Cache for language fallbacks.
Definition Language.php:180
static $lre
Unicode directional formatting characters, for embedBidi()
Definition Language.php:191
getSpecialPageAliases()
Get special page names, as an associative array canonical name => array of valid names,...
getParsedTitle()
For languages that support multiple variants, the title of an article may be displayed differently in...
LanguageNameUtils $langNameUtils
Definition Language.php:87
static isSupportedLanguage( $code)
Checks whether any localisation is available for that language tag in MediaWiki (MessagesXx....
Definition Language.php:309
hasVariant( $variant)
Strict check if the language has the specific variant.
static classFromCode( $code, $fallback=true)
transformUsingPairFile( $file, $string)
Transform a string using serialized data stored in the given file (which must be in the serialized su...
formatDuration( $seconds, array $chosenIntervals=[])
Takes a number of seconds and turns it into a text using values such as hours and minutes.
getMessage( $key)
const ALL
Return all known languages in fetchLanguageName(s).
Definition Language.php:50
pipeList(array $list)
Same as commaList, but separate it with the pipe instead.
internalUserTimeAndDate( $type, $ts, User $user, array $options)
Internal helper function for userDate(), userTime() and userTimeAndDate()
ucwordbreaksCallbackAscii( $matches)
lc( $str, $first=false)
Language false $mParentLanguage
Definition Language.php:69
fallback8bitEncoding()
getPluralRuleType( $number)
Find the plural rule type appropriate for the given number For example, if the language is set to Ara...
normalize( $s)
Convert a UTF-8 string to normal form C.
userAdjust( $ts, $tz=false)
Used by date() and time() to adjust the time output.
LocalisationCache $localisationCache
Definition Language.php:84
getArrow( $direction='forwards')
An arrow, depending on the language direction.
mbUpperChar( $char)
Convert character to uppercase, allowing overrides of the default mb_upper behaviour,...
equals(Language $lang)
Compare with an other language object.
ucwordbreaksCallbackMB( $matches)
convert( $text)
convert text to different variants of a language.
getParentLanguage()
Get the "parent" language which has a converter to convert a "compatible" language (in another varian...
getGenderNsText( $index, $gender)
Returns gender-dependent namespace alias if available.
Definition Language.php:593
translateBlockExpiry( $str, User $user=null, $now=0)
truncateHtml( $text, $length, $ellipsis='...')
Truncate a string of valid HTML to a specified length in bytes, appending an optional string (e....
getAllMessages()
getGrammarTransformations()
Get the grammar transformations data for the language.
__destruct()
Reduce memory usage PhanTypeSuspiciousNonTraversableForeach.
Definition Language.php:447
getFormattedNsText( $index)
A convenience function that returns the same thing as getNsText() except with '_' changed to ' ',...
Definition Language.php:580
getURLVariant()
getMessageFromDB( $msg)
Get a message from the MediaWiki namespace.
Definition Language.php:847
capitalizeAllNouns()
getDirMark( $opposite=false)
A hidden direction mark (LRM or RLM), depending on the language direction.
truncate_skip(&$ret, $text, $search, $start, $len=null)
truncateHtml() helper function like strcspn() but adds the skipped chars to $ret
__construct()
Definition Language.php:430
getFormattedNamespaces()
A convenience function that returns getNamespaces() with spaces instead of underscores in values.
Definition Language.php:543
getExtraHashOptions()
returns language specific options used by User::getPageRenderHash() for example, the preferred langua...
$transformData
ReplacementArray object caches.
Definition Language.php:81
userTimeAndDate( $ts, User $user, array $options=[])
Get the formatted date and time for the given timestamp and formatted for the given user.
setNamespaces(array $namespaces)
Arbitrarily set all of the namespace names at once.
Definition Language.php:523
commafy( $number)
Adds commas to a given number.
truncate_endBracket(&$tag, $tagType, $lastCh, &$openTags)
truncateHtml() helper function (a) push or pop $tag from $openTags as needed (b) clear $tag value
firstChar( $s)
Get the first character of a string.
$mExtendedSpecialPageAliases
Definition Language.php:72
getGrammarForms()
Get the grammar forms for the content language.
static getFallbacksFor( $code, $mode=self::MESSAGES_FALLBACKS)
Get the ordered list of fallback languages.
getNamespaces()
Returns an array of localised namespaces indexed by their numbers.
Definition Language.php:482
static convertDoubleWidth( $string)
convert double-width roman characters to single-width.
static tsToHebrew( $ts)
Converting Gregorian dates to Hebrew dates.
specialList( $page, $details, $oppositedm=true)
Make a list item, used by various special pages.
getHtmlCode()
Get the code in BCP 47 format which we can use inside of html lang="" tags.
needsGenderDistinction()
Whether this language uses gender-dependent namespace aliases.
Definition Language.php:608
static array $durationIntervals
Definition Language.php:162
getNsIndex( $text)
Get a namespace key by value, case insensitive.
Definition Language.php:723
linkPrefixCharset()
A regular expression character set to match legal word-prefixing characters which should be merged on...
listToText(array $list)
Take a list of strings and build a locale-friendly comma-separated list, using the local comma-separa...
caseFold( $s)
Return a case-folded representation of $s.
truncateInternal( $string, $length, $ellipsis, $adjustLength, callable $measureLength, callable $getSubstring)
Internal method used for truncation.
static hebrewNumeral( $num)
Hebrew Gematria number formatting up to 9999.
getCode()
Get the internal language code for this language object.
static $mIranianCalendarMonthMsgs
Definition Language.php:128
uc( $str, $first=false)
Convert a string to uppercase.
static $mHebrewCalendarMonthMsgs
Definition Language.php:135
normalizeForSearch( $string)
Some languages have special punctuation need to be normalized.
truncateForDatabase( $string, $length, $ellipsis='...', $adjustLength=true)
Truncate a string to a specified length in bytes, appending an optional string (e....
static newFromCode( $code, $fallback=false)
Create a language object for a given language code.
Definition Language.php:242
embedBidi( $text='')
Wraps argument with unicode control characters for directionality safety.
getMagic( $mw)
Fill a MagicWord object with data from here.
viewPrevNext(Title $title, $offset, $limit, array $query=[], $atend=false)
Generate (prev x| next x) (20|50|100...) type links for paging.
resetNamespaces()
Resets all of the namespace caches.
Definition Language.php:531
getMagicWords()
Get all magic words from cache.
minimumGroupingDigits()
emphasize( $text)
Italic is unsuitable for some languages.
timeanddate( $ts, $adj=false, $format=true, $timecorrection=false)
static getFallbacksIncludingSiteLanguage( $code)
Get the ordered list of fallback languages, ending with the fallback language chain for the site lang...
convertForSearchResult( $termsArray)
static $mMonthGenMsgs
Definition Language.php:118
getPreferredVariant()
getDirMarkEntity( $opposite=false)
A hidden direction mark (LRM or RLM), depending on the language direction.
convertGrammar( $word, $case)
Grammatical transformations, needed for inflected languages Invoked by putting {{grammar:case|word}} ...
autoConvert( $text, $variant=false)
convert text to a variant
getHumanTimestampInternal(MWTimestamp $ts, MWTimestamp $relativeTo, User $user)
Convert an MWTimestamp into a pretty human-readable timestamp using the given user preferences and re...
getHumanTimestamp(MWTimestamp $time, MWTimestamp $relativeTo=null, User $user=null)
Get the timestamp in a human-friendly relative format, e.g., "3 days ago".
convertNamespace( $ns, $variant=null)
Convert a namespace index to a string in the preferred variant.
preConvertPlural($forms, $count)
Checks that convertPlural was given an array and pads it to requested amount of forms by copying the ...
formatTimePeriod( $seconds, $format=[])
Formats a time given in seconds into a string representation of that time.
static getFallbackFor( $code)
Get the first fallback for a given language.
isRTL()
For right-to-left language support.
separatorTransformTable()
segmentForDiff( $text)
languages like Chinese need to be segmented in order for the diff to be of any use
static $mMonthAbbrevMsgs
Definition Language.php:123
replaceGrammarInNamespace( $m)
getBookstoreList()
Exports $wgBookstoreListEn.
Definition Language.php:472
getDir()
Return the correct HTML 'dir' attribute value for this language.
parseFormattedNumber( $number)
static factory( $code)
Get a cached or new language object for a given language code.
Definition Language.php:217
$mMagicExtensions
Definition Language.php:66
addMagicWordsByLang( $newWords)
Add magic words to the extension array.
linkPrefixExtension()
To allow "foo[[bar]]" to extend the link over the whole word "foobar".
getFallbackLanguages()
Definition Language.php:464
userDate( $ts, User $user, array $options=[])
Get the formatted date for the given timestamp and formatted for the given user.
getPluralRules()
Get the plural rules for the language.
autoConvertToAllVariants( $text)
convert text to all supported variants
const AS_AUTONYMS
Return autonyms in fetchLanguageName(s).
Definition Language.php:43
iconv( $in, $out, $string)
formatSize( $size)
Format a size in bytes for output, using an appropriate unit (B, KB, MB, GB, TB, PB,...
semicolonList(array $list)
Take a list of strings and build a locale-friendly semicolon-separated list, using the local semicolo...
fixVariableInNamespace( $talk)
static $rle
Definition Language.php:192
static insertSpace( $string, $pattern)
commaList(array $list)
Take a list of strings and build a locale-friendly comma-separated list, using the local comma-separa...
dateFormat( $usePrefs=true)
This is meant to be used by time(), date(), and timeanddate() to get the date preference they're supp...
static getMessagesFor( $code)
Get all messages for a given language WARNING: this may take a long time.
findVariantLink(&$link, &$nt, $ignoreOtherCond=false)
If a language supports multiple variants, it is possible that non-existing link in one variant actual...
const SUPPORTED
Return in fetchLanguageName(s) only the languages for which we have at least some localisation.
Definition Language.php:58
handleExplicitPluralForms( $count, array $forms)
Handles explicit plural forms for Language::convertPlural()
alignEnd()
Return 'right' or 'left' as appropriate alignment for line-end for this language's text direction.
getDateFormatString( $type, $pref)
Get a format string for a given type and preference.
convertPlural( $count, $forms)
Plural form transformations, needed for some languages.
getDurationIntervals( $seconds, array $chosenIntervals=[])
Takes a number of seconds and returns an array with a set of corresponding intervals.
static getLocalisationCache()
Get the LocalisationCache instance.
Definition Language.php:426
static $mLangObjCache
Definition Language.php:89
static getFileName( $prefix, $code, $suffix='.php')
Get the name of a file for a certain language code.
formatNumNoSeparators( $number)
Front-end for non-commafied formatNum.
static isKnownLanguageTag( $tag)
Returns true if a language code is an IETF tag known to MediaWiki.
Definition Language.php:415
removeBadCharLast( $string)
Remove bytes that represent an incomplete Unicode character at the end of string (e....
static getJsonMessagesFileName( $code)
getNamespaceAliases()
Definition Language.php:641
ucwordbreaks( $str)
capitalize words at word breaks
ucfirst( $str)
Make a string's first character uppercase.
static strongDirFromContent( $text='')
Gets directionality of the first strongly directional codepoint, for embedBidi()
userTime( $ts, User $user, array $options=[])
Get the formatted time for the given timestamp and formatted for the given user.
getNsText( $index)
Get a namespace value by key.
Definition Language.php:562
hasWordBreaks()
Most writing systems use whitespace to break up words.
LanguageConverter FakeConverter $mConverter
Definition Language.php:63
getConvRuleTitle()
Get the conversion rule title, if any.
formatBitrate( $bps)
Format a bitrate for output, using an appropriate unit (bps, kbps, Mbps, Gbps, Tbps,...
convertTitle( $title)
Convert a Title object to a string in the preferred variant.
numLink(Title $title, $offset, $limit, array $query, $link, $tooltipMsg, $class)
Helper function for viewPrevNext() that generates links.
static getMessagesFileName( $code)
getNamespaceIds()
Definition Language.php:693
static MapCacheLRU null $grammarTransformations
Cache for grammar rules data.
Definition Language.php:186
getVariants()
Get the list of variants supported by this language see sample implementation in LanguageZh....
time( $ts, $adj=false, $format=true, $timecorrection=false)
static getCodeFromFileName( $filename, $prefix='Language', $suffix='.php')
Get the language code from a file name.
removeBadCharFirst( $string)
Remove bytes that represent an incomplete Unicode character at the start of string (e....
static $mHijriCalendarMonthMsgs
Definition Language.php:151
static $mWeekdayAbbrevMsgs
Definition Language.php:109
unsegmentForDiff( $text)
and unsegment to show the result
formatComputingNumbers( $size, $boundary, $messageKey)
ucwordsCallbackMB( $matches)
isMultibyte( $str)
lcfirst( $str)
getPluralRuleIndexNumber( $number)
Find the index number of the plural rule appropriate for the given number.
updateConversionTable(Title $title)
Refresh the cache of conversion tables when MediaWiki:Conversiontable* is updated.
segmentByWord( $string)
Some languages such as Chinese require word segmentation, Specify such segmentation when overridden i...
checkTitleEncoding( $s)
TODO: $s is not always a string per T218883.
convertHtml( $text)
Perform output conversion on a string, and encode for safe HTML output.
getConverter()
Return the LanguageConverter used in the Language.
static tsToHijri( $ts)
Converting Gregorian dates to Hijri dates.
digitTransformTable()
static isValidCode( $code)
Returns true if a language code string is of a valid form, whether or not it exists.
Definition Language.php:385
static $mMonthMsgs
Definition Language.php:113
$dateFormatStrings
Definition Language.php:71
static romanNumeral( $num)
Roman number formatting up to 10000.
alignStart()
Return 'left' or 'right' as appropriate alignment for line-start for this language's text direction.
static hebrewYearStart( $year)
This calculates the Hebrew year start, as days since 1 September.
getCompiledPluralRules()
Get the compiled plural rules for the language.
Class for caching the contents of localisation files, Messages*.php and *.i18n.php.
MediaWiki exception.
Library for creating and parsing MW-style timestamps.
offsetForUser(User $user)
Adjust the timestamp depending on the given user's preferences.
Handles a simple LRU key/value map with a maximum number of entries.
A service that provides utilities to do with language names and codes.
isValidCode( $code)
Returns true if a language code string is of a valid form, whether or not it exists.
isValidBuiltInCode( $code)
Returns true if a language code is of a valid form for the purposes of internal customisation of Medi...
MediaWikiServices is the service locator for the application scope of MediaWiki.
Wrapper around strtr() that holds replacements.
static getSuggestedDurations(Language $lang=null, $includeOther=true)
Get an array of suggested block durations from MediaWiki:Ipboptions.
static isUtf8( $value)
Test whether a string is valid UTF-8.
Represents a title within MediaWiki.
Definition Title.php:42
The User object encapsulates all of the user-specific settings (user_id, name, rights,...
Definition User.php:51
getOption( $oname, $defaultOverride=null, $ignoreHidden=false)
Get the user's current setting for a given option.
Definition User.php:3022
static getDefaultOption( $opt)
Get a given default option value.
Definition User.php:1734
getDatePreference()
Get the user's preferred date format.
Definition User.php:3345
const NS_USER
Definition Defines.php:71
const NS_PROJECT_TALK
Definition Defines.php:74
const NS_USER_TALK
Definition Defines.php:72
const NS_PROJECT
Definition Defines.php:73
const DB_REPLICA
Definition defines.php:25
return true
Definition router.php:94
if(PHP_SAPI !='cli-server') if(!isset( $_SERVER['SCRIPT_FILENAME'])) $file
Item class for a filearchive table row.
Definition router.php:42
if(!isset( $args[0])) $lang