3 * See docs/magicword.txt.
5 * This program is free software; you can redistribute it and/or modify
6 * it under the terms of the GNU General Public License as published by
7 * the Free Software Foundation; either version 2 of the License, or
8 * (at your option) any later version.
10 * This program is distributed in the hope that it will be useful,
11 * but WITHOUT ANY WARRANTY; without even the implied warranty of
12 * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
13 * GNU General Public License for more details.
15 * You should have received a copy of the GNU General Public License along
16 * with this program; if not, write to the Free Software Foundation, Inc.,
17 * 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA.
18 * http://www.gnu.org/copyleft/gpl.html
25 * This class encapsulates "magic words" such as "#redirect", __NOTOC__, etc.
29 * if (MagicWord::get( 'redirect' )->match( $text ) ) {
34 * Possible future improvements:
35 * * Simultaneous searching for a number of magic words
36 * * MagicWord::$mObjects in shared memory
38 * Please avoid reading the data out of one of these objects and then writing
39 * special case code. If possible, add another match()-like function here.
41 * To add magic words in an extension, use $magicWords in a file listed in
42 * $wgExtensionMessagesFiles[].
48 * $magicWords['en'] = [
49 * 'magicwordkey' => [ 0, 'case_insensitive_magic_word' ],
50 * 'magicwordkey2' => [ 1, 'CASE_sensitive_magic_word2' ],
54 * For magic words which are also Parser variables, add a MagicWordwgVariableIDs
55 * hook. Use string keys.
69 public $mCaseSensitive;
75 private $mRegexStart = '';
78 private $mRegexStartToEnd = '';
81 private $mBaseRegex = '';
84 private $mVariableRegex = '';
87 private $mVariableStartToEndRegex = '';
90 private $mModified = false;
93 private $mFound = false;
95 public static $mVariableIDsInitialised = false;
96 public static $mVariableIDs = [
101 'currentmonthnamegen',
102 'currentmonthabbrev',
161 'numberofactiveusers',
177 /* Array of caching hints for ParserCache */
178 public static $mCacheTTLs = [
179 'currentmonth' => 86400,
180 'currentmonth1' => 86400,
181 'currentmonthname' => 86400,
182 'currentmonthnamegen' => 86400,
183 'currentmonthabbrev' => 86400,
184 'currentday' => 3600,
185 'currentday2' => 3600,
186 'currentdayname' => 3600,
187 'currentyear' => 86400,
188 'currenttime' => 3600,
189 'currenthour' => 3600,
190 'localmonth' => 86400,
191 'localmonth1' => 86400,
192 'localmonthname' => 86400,
193 'localmonthnamegen' => 86400,
194 'localmonthabbrev' => 86400,
197 'localdayname' => 3600,
198 'localyear' => 86400,
201 'numberofarticles' => 3600,
202 'numberoffiles' => 3600,
203 'numberofedits' => 3600,
204 'currentweek' => 3600,
205 'currentdow' => 3600,
208 'numberofusers' => 3600,
209 'numberofactiveusers' => 3600,
210 'numberofpages' => 3600,
211 'currentversion' => 86400,
212 'currenttimestamp' => 3600,
213 'localtimestamp' => 3600,
214 'pagesinnamespace' => 3600,
215 'numberofadmins' => 3600,
216 'numberingroup' => 3600,
219 public static $mDoubleUnderscoreIDs = [
235 public static $mSubstIDs = [
240 public static $mObjects = [];
241 public static $mDoubleUnderscoreArray = null;
245 public function __construct( $id = 0, $syn = [], $cs = false ) {
247 $this->mSynonyms
= (array)$syn;
248 $this->mCaseSensitive
= $cs;
252 * Factory: creates an object representing an ID
258 public static function &get( $id ) {
259 if ( !isset( self
::$mObjects[$id] ) ) {
260 $mw = new MagicWord();
262 self
::$mObjects[$id] = $mw;
264 return self
::$mObjects[$id];
268 * Get an array of parser variable IDs
272 public static function getVariableIDs() {
273 if ( !self
::$mVariableIDsInitialised ) {
275 Hooks
::run( 'MagicWordwgVariableIDs', [ &self
::$mVariableIDs ] );
276 self
::$mVariableIDsInitialised = true;
278 return self
::$mVariableIDs;
282 * Get an array of parser substitution modifier IDs
285 public static function getSubstIDs() {
286 return self
::$mSubstIDs;
290 * Allow external reads of TTL array
295 public static function getCacheTTL( $id ) {
296 if ( array_key_exists( $id, self
::$mCacheTTLs ) ) {
297 return self
::$mCacheTTLs[$id];
304 * Get a MagicWordArray of double-underscore entities
306 * @return MagicWordArray
308 public static function getDoubleUnderscoreArray() {
309 if ( is_null( self
::$mDoubleUnderscoreArray ) ) {
310 Hooks
::run( 'GetDoubleUnderscoreIDs', [ &self
::$mDoubleUnderscoreIDs ] );
311 self
::$mDoubleUnderscoreArray = new MagicWordArray( self
::$mDoubleUnderscoreIDs );
313 return self
::$mDoubleUnderscoreArray;
317 * Clear the self::$mObjects variable
318 * For use in parser tests
320 public static function clearCache() {
321 self
::$mObjects = [];
325 * Initialises this object with an ID
328 * @throws MWException
330 public function load( $id ) {
333 $wgContLang->getMagic( $this );
334 if ( !$this->mSynonyms
) {
335 $this->mSynonyms
= [ 'brionmademeputthishere' ];
336 throw new MWException( "Error: invalid magic word '$id'" );
341 * Preliminary initialisation
344 public function initRegex() {
345 // Sort the synonyms by length, descending, so that the longest synonym
346 // matches in precedence to the shortest
347 $synonyms = $this->mSynonyms
;
348 usort( $synonyms, [ $this, 'compareStringLength' ] );
351 foreach ( $synonyms as $synonym ) {
352 // In case a magic word contains /, like that's going to happen;)
353 $escSyn[] = preg_quote( $synonym, '/' );
355 $this->mBaseRegex
= implode( '|', $escSyn );
357 $case = $this->mCaseSensitive ?
'' : 'iu';
358 $this->mRegex
= "/{$this->mBaseRegex}/{$case}";
359 $this->mRegexStart
= "/^(?:{$this->mBaseRegex})/{$case}";
360 $this->mRegexStartToEnd
= "/^(?:{$this->mBaseRegex})$/{$case}";
361 $this->mVariableRegex
= str_replace( "\\$1", "(.*?)", $this->mRegex
);
362 $this->mVariableStartToEndRegex
= str_replace( "\\$1", "(.*?)",
363 "/^(?:{$this->mBaseRegex})$/{$case}" );
367 * A comparison function that returns -1, 0 or 1 depending on whether the
368 * first string is longer, the same length or shorter than the second
376 public function compareStringLength( $s1, $s2 ) {
381 } elseif ( $l1 > $l2 ) {
389 * Gets a regex representing matching the word
393 public function getRegex() {
394 if ( $this->mRegex
== '' ) {
397 return $this->mRegex
;
401 * Gets the regexp case modifier to use, i.e. i or nothing, to be used if
402 * one is using MagicWord::getBaseRegex(), otherwise it'll be included in
403 * the complete expression
407 public function getRegexCase() {
408 if ( $this->mRegex
=== '' ) {
412 return $this->mCaseSensitive ?
'' : 'iu';
416 * Gets a regex matching the word, if it is at the string start
420 public function getRegexStart() {
421 if ( $this->mRegex
== '' ) {
424 return $this->mRegexStart
;
428 * Gets a regex matching the word from start to end of a string
433 public function getRegexStartToEnd() {
434 if ( $this->mRegexStartToEnd
== '' ) {
437 return $this->mRegexStartToEnd
;
441 * regex without the slashes and what not
445 public function getBaseRegex() {
446 if ( $this->mRegex
== '' ) {
449 return $this->mBaseRegex
;
453 * Returns true if the text contains the word
455 * @param string $text
459 public function match( $text ) {
460 return (bool)preg_match( $this->getRegex(), $text );
464 * Returns true if the text starts with the word
466 * @param string $text
470 public function matchStart( $text ) {
471 return (bool)preg_match( $this->getRegexStart(), $text );
475 * Returns true if the text matched the word
477 * @param string $text
482 public function matchStartToEnd( $text ) {
483 return (bool)preg_match( $this->getRegexStartToEnd(), $text );
487 * Returns NULL if there's no match, the value of $1 otherwise
488 * The return code is the matched string, if there's no variable
489 * part in the regex and the matched variable part ($1) if there
492 * @param string $text
496 public function matchVariableStartToEnd( $text ) {
498 $matchcount = preg_match( $this->getVariableStartToEndRegex(), $text, $matches );
499 if ( $matchcount == 0 ) {
502 # multiple matched parts (variable match); some will be empty because of
503 # synonyms. The variable will be the second non-empty one so remove any
504 # blank elements and re-sort the indices.
507 $matches = array_values( array_filter( $matches ) );
509 if ( count( $matches ) == 1 ) {
518 * Returns true if the text matches the word, and alters the
519 * input string, removing all instances of the word
521 * @param string $text
525 public function matchAndRemove( &$text ) {
526 $this->mFound
= false;
527 $text = preg_replace_callback(
529 [ $this, 'pregRemoveAndRecord' ],
533 return $this->mFound
;
537 * @param string $text
540 public function matchStartAndRemove( &$text ) {
541 $this->mFound
= false;
542 $text = preg_replace_callback(
543 $this->getRegexStart(),
544 [ $this, 'pregRemoveAndRecord' ],
548 return $this->mFound
;
552 * Used in matchAndRemove()
556 public function pregRemoveAndRecord() {
557 $this->mFound
= true;
562 * Replaces the word with something else
564 * @param string $replacement
565 * @param string $subject
570 public function replace( $replacement, $subject, $limit = -1 ) {
573 StringUtils
::escapeRegexReplacement( $replacement ),
577 $this->mModified
= $res !== $subject;
582 * Variable handling: {{SUBST:xxx}} style words
583 * Calls back a function to determine what to replace xxx with
584 * Input word must contain $1
586 * @param string $text
587 * @param callable $callback
591 public function substituteCallback( $text, $callback ) {
592 $res = preg_replace_callback( $this->getVariableRegex(), $callback, $text );
593 $this->mModified
= $res !== $text;
598 * Matches the word, where $1 is a wildcard
602 public function getVariableRegex() {
603 if ( $this->mVariableRegex
== '' ) {
606 return $this->mVariableRegex
;
610 * Matches the entire string, where $1 is a wildcard
614 public function getVariableStartToEndRegex() {
615 if ( $this->mVariableStartToEndRegex
== '' ) {
618 return $this->mVariableStartToEndRegex
;
622 * Accesses the synonym list directly
628 public function getSynonym( $i ) {
629 return $this->mSynonyms
[$i];
635 public function getSynonyms() {
636 return $this->mSynonyms
;
640 * Returns true if the last call to replace() or substituteCallback()
641 * returned a modified text, otherwise false.
645 public function getWasModified() {
646 return $this->mModified
;
650 * $magicarr is an associative array of (magic word ID => replacement)
651 * This method uses the php feature to do several replacements at the same time,
652 * thereby gaining some efficiency. The result is placed in the out variable
653 * $result. The return value is true if something was replaced.
654 * @deprecated since 1.25, unused
656 * @param array $magicarr
657 * @param string $subject
658 * @param string $result
662 public function replaceMultiple( $magicarr, $subject, &$result ) {
663 wfDeprecated( __METHOD__
, '1.25' );
666 foreach ( $magicarr as $id => $replacement ) {
667 $mw = MagicWord
::get( $id );
668 $search[] = $mw->getRegex();
669 $replace[] = $replacement;
672 $result = preg_replace( $search, $replace, $subject );
673 return $result !== $subject;
677 * Adds all the synonyms of this MagicWord to an array, to allow quick
678 * lookup in a list of magic words
680 * @param array $array
681 * @param string $value
683 public function addToArray( &$array, $value ) {
685 foreach ( $this->mSynonyms
as $syn ) {
686 $array[$wgContLang->lc( $syn )] = $value;
693 public function isCaseSensitive() {
694 return $this->mCaseSensitive
;
700 public function getId() {