5 * See docs/magicword.txt
12 * This class encapsulates "magic words" such as #redirect, __NOTOC__, etc.
14 * if (MagicWord::get( 'redirect' )->match( $text ) )
16 * Possible future improvements:
17 * * Simultaneous searching for a number of magic words
18 * * MagicWord::$mObjects in shared memory
20 * Please avoid reading the data out of one of these objects and then writing
21 * special case code. If possible, add another match()-like function here.
23 * To add magic words in an extension, use the LanguageGetMagic hook. For
24 * magic words which are also Parser variables, add a MagicWordwgVariableIDs
25 * hook. Use string keys.
33 var $mId, $mSynonyms, $mCaseSensitive;
35 var $mRegexStart = '';
37 var $mVariableRegex = '';
38 var $mVariableStartToEndRegex = '';
39 var $mModified = false;
42 static public $mVariableIDsInitialised = false;
43 static public $mVariableIDs = array(
47 'currentmonthnamegen',
104 'numberofactiveusers',
117 /* Array of caching hints for ParserCache */
118 static public $mCacheTTLs = array (
119 'currentmonth' => 86400,
120 'currentmonth1' => 86400,
121 'currentmonthname' => 86400,
122 'currentmonthnamegen' => 86400,
123 'currentmonthabbrev' => 86400,
124 'currentday' => 3600,
125 'currentday2' => 3600,
126 'currentdayname' => 3600,
127 'currentyear' => 86400,
128 'currenttime' => 3600,
129 'currenthour' => 3600,
130 'localmonth' => 86400,
131 'localmonth1' => 86400,
132 'localmonthname' => 86400,
133 'localmonthnamegen' => 86400,
134 'localmonthabbrev' => 86400,
137 'localdayname' => 3600,
138 'localyear' => 86400,
141 'numberofarticles' => 3600,
142 'numberoffiles' => 3600,
143 'numberofedits' => 3600,
144 'currentweek' => 3600,
145 'currentdow' => 3600,
148 'numberofusers' => 3600,
149 'numberofactiveusers' => 3600,
150 'numberofpages' => 3600,
151 'currentversion' => 86400,
152 'currenttimestamp' => 3600,
153 'localtimestamp' => 3600,
154 'pagesinnamespace' => 3600,
155 'numberofadmins' => 3600,
156 'numberofviews' => 3600,
157 'numberingroup' => 3600,
160 static public $mDoubleUnderscoreIDs = array(
176 static public $mSubstIDs = array(
181 static public $mObjects = array();
182 static public $mDoubleUnderscoreArray = null;
186 function __construct($id = 0, $syn = array(), $cs = false) {
188 $this->mSynonyms
= (array)$syn;
189 $this->mCaseSensitive
= $cs;
193 * Factory: creates an object representing an ID
199 static function &get( $id ) {
200 if ( !isset( self
::$mObjects[$id] ) ) {
201 $mw = new MagicWord();
203 self
::$mObjects[$id] = $mw;
205 return self
::$mObjects[$id];
209 * Get an array of parser variable IDs
213 static function getVariableIDs() {
214 if ( !self
::$mVariableIDsInitialised ) {
215 # Deprecated constant definition hook, available for extensions that need it
216 $magicWords = array();
217 wfRunHooks( 'MagicWordMagicWords', array( &$magicWords ) );
218 foreach ( $magicWords as $word ) {
219 define( $word, $word );
223 wfRunHooks( 'MagicWordwgVariableIDs', array( &self
::$mVariableIDs ) );
224 self
::$mVariableIDsInitialised = true;
226 return self
::$mVariableIDs;
230 * Get an array of parser substitution modifier IDs
233 static function getSubstIDs() {
234 return self
::$mSubstIDs;
238 * Allow external reads of TTL array
243 static function getCacheTTL( $id ) {
244 if ( array_key_exists( $id, self
::$mCacheTTLs ) ) {
245 return self
::$mCacheTTLs[$id];
252 * Get a MagicWordArray of double-underscore entities
254 * @return MagicWordArray
256 static function getDoubleUnderscoreArray() {
257 if ( is_null( self
::$mDoubleUnderscoreArray ) ) {
258 self
::$mDoubleUnderscoreArray = new MagicWordArray( self
::$mDoubleUnderscoreIDs );
260 return self
::$mDoubleUnderscoreArray;
264 * Clear the self::$mObjects variable
265 * For use in parser tests
267 public static function clearCache() {
268 self
::$mObjects = array();
272 * Initialises this object with an ID
276 function load( $id ) {
278 wfProfileIn( __METHOD__
);
280 $wgContLang->getMagic( $this );
281 if ( !$this->mSynonyms
) {
282 $this->mSynonyms
= array( 'dkjsagfjsgashfajsh' );
283 #throw new MWException( "Error: invalid magic word '$id'" );
284 wfDebugLog( 'exception', "Error: invalid magic word '$id'\n" );
286 wfProfileOut( __METHOD__
);
290 * Preliminary initialisation
293 function initRegex() {
294 // Sort the synonyms by length, descending, so that the longest synonym
295 // matches in precedence to the shortest
296 $synonyms = $this->mSynonyms
;
297 usort( $synonyms, array( $this, 'compareStringLength' ) );
300 foreach ( $synonyms as $synonym )
301 // In case a magic word contains /, like that's going to happen;)
302 $escSyn[] = preg_quote( $synonym, '/' );
303 $this->mBaseRegex
= implode( '|', $escSyn );
305 $case = $this->mCaseSensitive ?
'' : 'iu';
306 $this->mRegex
= "/{$this->mBaseRegex}/{$case}";
307 $this->mRegexStart
= "/^(?:{$this->mBaseRegex})/{$case}";
308 $this->mVariableRegex
= str_replace( "\\$1", "(.*?)", $this->mRegex
);
309 $this->mVariableStartToEndRegex
= str_replace( "\\$1", "(.*?)",
310 "/^(?:{$this->mBaseRegex})$/{$case}" );
314 * A comparison function that returns -1, 0 or 1 depending on whether the
315 * first string is longer, the same length or shorter than the second
323 function compareStringLength( $s1, $s2 ) {
328 } elseif ( $l1 > $l2 ) {
336 * Gets a regex representing matching the word
340 function getRegex() {
341 if ($this->mRegex
== '' ) {
344 return $this->mRegex
;
348 * Gets the regexp case modifier to use, i.e. i or nothing, to be used if
349 * one is using MagicWord::getBaseRegex(), otherwise it'll be included in
350 * the complete expression
354 function getRegexCase() {
355 if ( $this->mRegex
=== '' )
358 return $this->mCaseSensitive ?
'' : 'iu';
362 * Gets a regex matching the word, if it is at the string start
366 function getRegexStart() {
367 if ($this->mRegex
== '' ) {
370 return $this->mRegexStart
;
374 * regex without the slashes and what not
378 function getBaseRegex() {
379 if ($this->mRegex
== '') {
382 return $this->mBaseRegex
;
386 * Returns true if the text contains the word
388 * @param $text string
392 function match( $text ) {
393 return (bool)preg_match( $this->getRegex(), $text );
397 * Returns true if the text starts with the word
399 * @param $text string
403 function matchStart( $text ) {
404 return (bool)preg_match( $this->getRegexStart(), $text );
408 * Returns NULL if there's no match, the value of $1 otherwise
409 * The return code is the matched string, if there's no variable
410 * part in the regex and the matched variable part ($1) if there
413 * @param $text string
417 function matchVariableStartToEnd( $text ) {
419 $matchcount = preg_match( $this->getVariableStartToEndRegex(), $text, $matches );
420 if ( $matchcount == 0 ) {
423 # multiple matched parts (variable match); some will be empty because of
424 # synonyms. The variable will be the second non-empty one so remove any
425 # blank elements and re-sort the indices.
428 $matches = array_values(array_filter($matches));
430 if ( count($matches) == 1 ) {
440 * Returns true if the text matches the word, and alters the
441 * input string, removing all instances of the word
443 * @param $text string
447 function matchAndRemove( &$text ) {
448 $this->mFound
= false;
449 $text = preg_replace_callback( $this->getRegex(), array( &$this, 'pregRemoveAndRecord' ), $text );
450 return $this->mFound
;
457 function matchStartAndRemove( &$text ) {
458 $this->mFound
= false;
459 $text = preg_replace_callback( $this->getRegexStart(), array( &$this, 'pregRemoveAndRecord' ), $text );
460 return $this->mFound
;
464 * Used in matchAndRemove()
468 function pregRemoveAndRecord() {
469 $this->mFound
= true;
474 * Replaces the word with something else
476 * @param $replacement
482 function replace( $replacement, $subject, $limit = -1 ) {
483 $res = preg_replace( $this->getRegex(), StringUtils
::escapeRegexReplacement( $replacement ), $subject, $limit );
484 $this->mModified
= !($res === $subject);
489 * Variable handling: {{SUBST:xxx}} style words
490 * Calls back a function to determine what to replace xxx with
491 * Input word must contain $1
493 * @param $text string
498 function substituteCallback( $text, $callback ) {
499 $res = preg_replace_callback( $this->getVariableRegex(), $callback, $text );
500 $this->mModified
= !($res === $text);
505 * Matches the word, where $1 is a wildcard
509 function getVariableRegex() {
510 if ( $this->mVariableRegex
== '' ) {
513 return $this->mVariableRegex
;
517 * Matches the entire string, where $1 is a wildcard
521 function getVariableStartToEndRegex() {
522 if ( $this->mVariableStartToEndRegex
== '' ) {
525 return $this->mVariableStartToEndRegex
;
529 * Accesses the synonym list directly
535 function getSynonym( $i ) {
536 return $this->mSynonyms
[$i];
542 function getSynonyms() {
543 return $this->mSynonyms
;
547 * Returns true if the last call to replace() or substituteCallback()
548 * returned a modified text, otherwise false.
552 function getWasModified(){
553 return $this->mModified
;
557 * $magicarr is an associative array of (magic word ID => replacement)
558 * This method uses the php feature to do several replacements at the same time,
559 * thereby gaining some efficiency. The result is placed in the out variable
560 * $result. The return value is true if something was replaced.
561 * @todo Should this be static? It doesn't seem to be used at all
569 function replaceMultiple( $magicarr, $subject, &$result ){
572 foreach( $magicarr as $id => $replacement ){
573 $mw = MagicWord
::get( $id );
574 $search[] = $mw->getRegex();
575 $replace[] = $replacement;
578 $result = preg_replace( $search, $replace, $subject );
579 return !($result === $subject);
583 * Adds all the synonyms of this MagicWord to an array, to allow quick
584 * lookup in a list of magic words
589 function addToArray( &$array, $value ) {
591 foreach ( $this->mSynonyms
as $syn ) {
592 $array[$wgContLang->lc($syn)] = $value;
599 function isCaseSensitive() {
600 return $this->mCaseSensitive
;
612 * Class for handling an array of magic words
615 class MagicWordArray
{
616 var $names = array();
618 var $baseRegex, $regex;
621 function __construct( $names = array() ) {
622 $this->names
= $names;
626 * Add a magic word by name
628 * @param $name string
630 public function add( $name ) {
631 $this->names
[] = $name;
632 $this->hash
= $this->baseRegex
= $this->regex
= null;
636 * Add a number of magic words by name
638 * @param $names array
640 public function addArray( $names ) {
641 $this->names
= array_merge( $this->names
, array_values( $names ) );
642 $this->hash
= $this->baseRegex
= $this->regex
= null;
646 * Get a 2-d hashtable for this array
649 if ( is_null( $this->hash
) ) {
651 $this->hash
= array( 0 => array(), 1 => array() );
652 foreach ( $this->names
as $name ) {
653 $magic = MagicWord
::get( $name );
654 $case = intval( $magic->isCaseSensitive() );
655 foreach ( $magic->getSynonyms() as $syn ) {
657 $syn = $wgContLang->lc( $syn );
659 $this->hash
[$case][$syn] = $name;
669 function getBaseRegex() {
670 if ( is_null( $this->baseRegex
) ) {
671 $this->baseRegex
= array( 0 => '', 1 => '' );
672 foreach ( $this->names
as $name ) {
673 $magic = MagicWord
::get( $name );
674 $case = intval( $magic->isCaseSensitive() );
675 foreach ( $magic->getSynonyms() as $i => $syn ) {
676 $group = "(?P<{$i}_{$name}>" . preg_quote( $syn, '/' ) . ')';
677 if ( $this->baseRegex
[$case] === '' ) {
678 $this->baseRegex
[$case] = $group;
680 $this->baseRegex
[$case] .= '|' . $group;
685 return $this->baseRegex
;
689 * Get an unanchored regex that does not match parameters
691 function getRegex() {
692 if ( is_null( $this->regex
) ) {
693 $base = $this->getBaseRegex();
694 $this->regex
= array( '', '' );
695 if ( $this->baseRegex
[0] !== '' ) {
696 $this->regex
[0] = "/{$base[0]}/iuS";
698 if ( $this->baseRegex
[1] !== '' ) {
699 $this->regex
[1] = "/{$base[1]}/S";
706 * Get a regex for matching variables with parameters
710 function getVariableRegex() {
711 return str_replace( "\\$1", "(.*?)", $this->getRegex() );
715 * Get a regex anchored to the start of the string that does not match parameters
719 function getRegexStart() {
720 $base = $this->getBaseRegex();
721 $newRegex = array( '', '' );
722 if ( $base[0] !== '' ) {
723 $newRegex[0] = "/^(?:{$base[0]})/iuS";
725 if ( $base[1] !== '' ) {
726 $newRegex[1] = "/^(?:{$base[1]})/S";
732 * Get an anchored regex for matching variables with parameters
736 function getVariableStartToEndRegex() {
737 $base = $this->getBaseRegex();
738 $newRegex = array( '', '' );
739 if ( $base[0] !== '' ) {
740 $newRegex[0] = str_replace( "\\$1", "(.*?)", "/^(?:{$base[0]})$/iuS" );
742 if ( $base[1] !== '' ) {
743 $newRegex[1] = str_replace( "\\$1", "(.*?)", "/^(?:{$base[1]})$/S" );
749 * Parse a match array from preg_match
750 * Returns array(magic word ID, parameter value)
751 * If there is no parameter value, that element will be false.
757 function parseMatch( $m ) {
759 while ( list( $key, $value ) = each( $m ) ) {
760 if ( $key === 0 ||
$value === '' ) {
763 $parts = explode( '_', $key, 2 );
764 if ( count( $parts ) != 2 ) {
765 // This shouldn't happen
767 throw new MWException( __METHOD__
. ': bad parameter name' );
769 list( /* $synIndex */, $magicName ) = $parts;
770 $paramValue = next( $m );
771 return array( $magicName, $paramValue );
773 // This shouldn't happen either
774 throw new MWException( __METHOD__
.': parameter not found' );
778 * Match some text, with parameter capture
779 * Returns an array with the magic word name in the first element and the
780 * parameter in the second element.
781 * Both elements are false if there was no match.
783 * @param $text string
787 public function matchVariableStartToEnd( $text ) {
788 $regexes = $this->getVariableStartToEndRegex();
789 foreach ( $regexes as $regex ) {
790 if ( $regex !== '' ) {
792 if ( preg_match( $regex, $text, $m ) ) {
793 return $this->parseMatch( $m );
797 return array( false, false );
801 * Match some text, without parameter capture
802 * Returns the magic word name, or false if there was no capture
804 * @param $text string
806 * @return string|false
808 public function matchStartToEnd( $text ) {
809 $hash = $this->getHash();
810 if ( isset( $hash[1][$text] ) ) {
811 return $hash[1][$text];
814 $lc = $wgContLang->lc( $text );
815 if ( isset( $hash[0][$lc] ) ) {
816 return $hash[0][$lc];
822 * Returns an associative array, ID => param value, for all items that match
823 * Removes the matched items from the input string (passed by reference)
825 * @param $text string
829 public function matchAndRemove( &$text ) {
831 $regexes = $this->getRegex();
832 foreach ( $regexes as $regex ) {
833 if ( $regex === '' ) {
836 preg_match_all( $regex, $text, $matches, PREG_SET_ORDER
);
837 foreach ( $matches as $m ) {
838 list( $name, $param ) = $this->parseMatch( $m );
839 $found[$name] = $param;
841 $text = preg_replace( $regex, '', $text );
847 * Return the ID of the magic word at the start of $text, and remove
848 * the prefix from $text.
849 * Return false if no match found and $text is not modified.
850 * Does not match parameters.
852 * @param $text string
856 public function matchStartAndRemove( &$text ) {
857 $regexes = $this->getRegexStart();
858 foreach ( $regexes as $regex ) {
859 if ( $regex === '' ) {
862 if ( preg_match( $regex, $text, $m ) ) {
863 list( $id, ) = $this->parseMatch( $m );
864 if ( strlen( $m[0] ) >= strlen( $text ) ) {
867 $text = substr( $text, strlen( $m[0] ) );