includes/MagicWord.php

   1 <?php
   2 /**
   3  * File for magic words.
   4  *
   5  * See docs/magicword.txt.
   6  *
   7  * This program is free software; you can redistribute it and/or modify
   8  * it under the terms of the GNU General Public License as published by
   9  * the Free Software Foundation; either version 2 of the License, or
  10  * (at your option) any later version.
  11  *
  12  * This program is distributed in the hope that it will be useful,
  13  * but WITHOUT ANY WARRANTY; without even the implied warranty of
  14  * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
  15  * GNU General Public License for more details.
  16  *
  17  * You should have received a copy of the GNU General Public License along
  18  * with this program; if not, write to the Free Software Foundation, Inc.,
  19  * 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA.
  20  * http://www.gnu.org/copyleft/gpl.html
  21  *
  22  * @file
  23  * @ingroup Parser
  24  */
  25
  26 /**
  27  * This class encapsulates "magic words" such as #redirect, __NOTOC__, etc.
  28  *
  29  * @par Usage:
  30  * @code
  31  *     if (MagicWord::get( 'redirect' )->match( $text ) ) {
  32  *       // some code
  33  *     }
  34  * @endcode
  35  *
  36  * Possible future improvements:
  37  *   * Simultaneous searching for a number of magic words
  38  *   * MagicWord::$mObjects in shared memory
  39  *
  40  * Please avoid reading the data out of one of these objects and then writing
  41  * special case code. If possible, add another match()-like function here.
  42  *
  43  * To add magic words in an extension, use $magicWords in a file listed in
  44  * $wgExtensionMessagesFiles[].
  45  *
  46  * @par Example:
  47  * @code
  48  * $magicWords = array();
  49  *
  50  * $magicWords['en'] = array(
  51  *      'magicwordkey' => array( 0, 'case_insensitive_magic_word' ),
  52  *      'magicwordkey2' => array( 1, 'CASE_sensitive_magic_word2' ),
  53  * );
  54  * @endcode
  55  *
  56  * For magic words which are also Parser variables, add a MagicWordwgVariableIDs
  57  * hook. Use string keys.
  58  *
  59  * @ingroup Parser
  60  */
  61 class MagicWord {
  62         /**#@+
  63          * @private
  64          */
  65         var $mId, $mSynonyms, $mCaseSensitive;
  66         var $mRegex = '';
  67         var $mRegexStart = '';
  68         var $mBaseRegex = '';
  69         var $mVariableRegex = '';
  70         var $mVariableStartToEndRegex = '';
  71         var $mModified = false;
  72         var $mFound = false;
  73
  74         static public $mVariableIDsInitialised = false;
  75         static public $mVariableIDs = array(
  76                 'currentmonth',
  77                 'currentmonth1',
  78                 'currentmonthname',
  79                 'currentmonthnamegen',
  80                 'currentmonthabbrev',
  81                 'currentday',
  82                 'currentday2',
  83                 'currentdayname',
  84                 'currentyear',
  85                 'currenttime',
  86                 'currenthour',
  87                 'localmonth',
  88                 'localmonth1',
  89                 'localmonthname',
  90                 'localmonthnamegen',
  91                 'localmonthabbrev',
  92                 'localday',
  93                 'localday2',
  94                 'localdayname',
  95                 'localyear',
  96                 'localtime',
  97                 'localhour',
  98                 'numberofarticles',
  99                 'numberoffiles',
 100                 'numberofedits',
 101                 'articlepath',
 102                 'pageid',
 103                 'sitename',
 104                 'server',
 105                 'servername',
 106                 'scriptpath',
 107                 'stylepath',
 108                 'pagename',
 109                 'pagenamee',
 110                 'fullpagename',
 111                 'fullpagenamee',
 112                 'namespace',
 113                 'namespacee',
 114                 'namespacenumber',
 115                 'currentweek',
 116                 'currentdow',
 117                 'localweek',
 118                 'localdow',
 119                 'revisionid',
 120                 'revisionday',
 121                 'revisionday2',
 122                 'revisionmonth',
 123                 'revisionmonth1',
 124                 'revisionyear',
 125                 'revisiontimestamp',
 126                 'revisionuser',
 127                 'subpagename',
 128                 'subpagenamee',
 129                 'talkspace',
 130                 'talkspacee',
 131                 'subjectspace',
 132                 'subjectspacee',
 133                 'talkpagename',
 134                 'talkpagenamee',
 135                 'subjectpagename',
 136                 'subjectpagenamee',
 137                 'numberofusers',
 138                 'numberofactiveusers',
 139                 'numberofpages',
 140                 'currentversion',
 141                 'basepagename',
 142                 'basepagenamee',
 143                 'currenttimestamp',
 144                 'localtimestamp',
 145                 'directionmark',
 146                 'contentlanguage',
 147                 'numberofadmins',
 148                 'numberofviews',
 149         );
 150
 151         /* Array of caching hints for ParserCache */
 152         static public $mCacheTTLs = array (
 153                 'currentmonth' => 86400,
 154                 'currentmonth1' => 86400,
 155                 'currentmonthname' => 86400,
 156                 'currentmonthnamegen' => 86400,
 157                 'currentmonthabbrev' => 86400,
 158                 'currentday' => 3600,
 159                 'currentday2' => 3600,
 160                 'currentdayname' => 3600,
 161                 'currentyear' => 86400,
 162                 'currenttime' => 3600,
 163                 'currenthour' => 3600,
 164                 'localmonth' => 86400,
 165                 'localmonth1' => 86400,
 166                 'localmonthname' => 86400,
 167                 'localmonthnamegen' => 86400,
 168                 'localmonthabbrev' => 86400,
 169                 'localday' => 3600,
 170                 'localday2' => 3600,
 171                 'localdayname' => 3600,
 172                 'localyear' => 86400,
 173                 'localtime' => 3600,
 174                 'localhour' => 3600,
 175                 'numberofarticles' => 3600,
 176                 'numberoffiles' => 3600,
 177                 'numberofedits' => 3600,
 178                 'currentweek' => 3600,
 179                 'currentdow' => 3600,
 180                 'localweek' => 3600,
 181                 'localdow' => 3600,
 182                 'numberofusers' => 3600,
 183                 'numberofactiveusers' => 3600,
 184                 'numberofpages' => 3600,
 185                 'currentversion' => 86400,
 186                 'currenttimestamp' => 3600,
 187                 'localtimestamp' => 3600,
 188                 'pagesinnamespace' => 3600,
 189                 'numberofadmins' => 3600,
 190                 'numberofviews' => 3600,
 191                 'numberingroup' => 3600,
 192                 );
 193
 194         static public $mDoubleUnderscoreIDs = array(
 195                 'notoc',
 196                 'nogallery',
 197                 'forcetoc',
 198                 'toc',
 199                 'noeditsection',
 200                 'newsectionlink',
 201                 'nonewsectionlink',
 202                 'hiddencat',
 203                 'index',
 204                 'noindex',
 205                 'staticredirect',
 206                 'notitleconvert',
 207                 'nocontentconvert',
 208         );
 209
 210         static public $mSubstIDs = array(
 211                 'subst',
 212                 'safesubst',
 213         );
 214
 215         static public $mObjects = array();
 216         static public $mDoubleUnderscoreArray = null;
 217
 218         /**#@-*/
 219
 220         function __construct($id = 0, $syn = array(), $cs = false) {
 221                 $this->mId = $id;
 222                 $this->mSynonyms = (array)$syn;
 223                 $this->mCaseSensitive = $cs;
 224         }
 225
 226         /**
 227          * Factory: creates an object representing an ID
 228          *
 229          * @param $id
 230          *
 231          * @return MagicWord
 232          */
 233         static function &get( $id ) {
 234                 if ( !isset( self::$mObjects[$id] ) ) {
 235                         $mw = new MagicWord();
 236                         $mw->load( $id );
 237                         self::$mObjects[$id] = $mw;
 238                 }
 239                 return self::$mObjects[$id];
 240         }
 241
 242         /**
 243          * Get an array of parser variable IDs
 244          *
 245          * @return array
 246          */
 247         static function getVariableIDs() {
 248                 if ( !self::$mVariableIDsInitialised ) {
 249                         # Get variable IDs
 250                         wfRunHooks( 'MagicWordwgVariableIDs', array( &self::$mVariableIDs ) );
 251                         self::$mVariableIDsInitialised = true;
 252                 }
 253                 return self::$mVariableIDs;
 254         }
 255
 256         /**
 257          * Get an array of parser substitution modifier IDs
 258          * @return array
 259          */
 260         static function getSubstIDs() {
 261                 return self::$mSubstIDs;
 262         }
 263
 264         /**
 265          * Allow external reads of TTL array
 266          *
 267          * @param $id int
 268          * @return array
 269          */
 270         static function getCacheTTL( $id ) {
 271                 if ( array_key_exists( $id, self::$mCacheTTLs ) ) {
 272                         return self::$mCacheTTLs[$id];
 273                 } else {
 274                         return -1;
 275                 }
 276         }
 277
 278         /**
 279          * Get a MagicWordArray of double-underscore entities
 280          *
 281          * @return MagicWordArray
 282          */
 283         static function getDoubleUnderscoreArray() {
 284                 if ( is_null( self::$mDoubleUnderscoreArray ) ) {
 285                         self::$mDoubleUnderscoreArray = new MagicWordArray( self::$mDoubleUnderscoreIDs );
 286                 }
 287                 return self::$mDoubleUnderscoreArray;
 288         }
 289
 290         /**
 291          * Clear the self::$mObjects variable
 292          * For use in parser tests
 293          */
 294         public static function clearCache() {
 295                 self::$mObjects = array();
 296         }
 297
 298         /**
 299          * Initialises this object with an ID
 300          *
 301          * @param $id
 302          */
 303         function load( $id ) {
 304                 global $wgContLang;
 305                 wfProfileIn( __METHOD__ );
 306                 $this->mId = $id;
 307                 $wgContLang->getMagic( $this );
 308                 if ( !$this->mSynonyms ) {
 309                         $this->mSynonyms = array( 'dkjsagfjsgashfajsh' );
 310                         throw new MWException( "Error: invalid magic word '$id'" );
 311                         #wfDebugLog( 'exception', "Error: invalid magic word '$id'\n" );
 312                 }
 313                 wfProfileOut( __METHOD__ );
 314         }
 315
 316         /**
 317          * Preliminary initialisation
 318          * @private
 319          */
 320         function initRegex() {
 321                 // Sort the synonyms by length, descending, so that the longest synonym
 322                 // matches in precedence to the shortest
 323                 $synonyms = $this->mSynonyms;
 324                 usort( $synonyms, array( $this, 'compareStringLength' ) );
 325
 326                 $escSyn = array();
 327                 foreach ( $synonyms as $synonym )
 328                         // In case a magic word contains /, like that's going to happen;)
 329                         $escSyn[] = preg_quote( $synonym, '/' );
 330                 $this->mBaseRegex = implode( '|', $escSyn );
 331
 332                 $case = $this->mCaseSensitive ? '' : 'iu';
 333                 $this->mRegex = "/{$this->mBaseRegex}/{$case}";
 334                 $this->mRegexStart = "/^(?:{$this->mBaseRegex})/{$case}";
 335                 $this->mVariableRegex = str_replace( "\\$1", "(.*?)", $this->mRegex );
 336                 $this->mVariableStartToEndRegex = str_replace( "\\$1", "(.*?)",
 337                         "/^(?:{$this->mBaseRegex})$/{$case}" );
 338         }
 339
 340         /**
 341          * A comparison function that returns -1, 0 or 1 depending on whether the
 342          * first string is longer, the same length or shorter than the second
 343          * string.
 344          *
 345          * @param $s1 string
 346          * @param $s2 string
 347          *
 348          * @return int
 349          */
 350         function compareStringLength( $s1, $s2 ) {
 351                 $l1 = strlen( $s1 );
 352                 $l2 = strlen( $s2 );
 353                 if ( $l1 < $l2 ) {
 354                         return 1;
 355                 } elseif ( $l1 > $l2 ) {
 356                         return -1;
 357                 } else {
 358                         return 0;
 359                 }
 360         }
 361
 362         /**
 363          * Gets a regex representing matching the word
 364          *
 365          * @return string
 366          */
 367         function getRegex() {
 368                 if ($this->mRegex == '' ) {
 369                         $this->initRegex();
 370                 }
 371                 return $this->mRegex;
 372         }
 373
 374         /**
 375          * Gets the regexp case modifier to use, i.e. i or nothing, to be used if
 376          * one is using MagicWord::getBaseRegex(), otherwise it'll be included in
 377          * the complete expression
 378          *
 379          * @return string
 380          */
 381         function getRegexCase() {
 382                 if ( $this->mRegex === '' )
 383                         $this->initRegex();
 384
 385                 return $this->mCaseSensitive ? '' : 'iu';
 386         }
 387
 388         /**
 389          * Gets a regex matching the word, if it is at the string start
 390          *
 391          * @return string
 392          */
 393         function getRegexStart() {
 394                 if ($this->mRegex == '' ) {
 395                         $this->initRegex();
 396                 }
 397                 return $this->mRegexStart;
 398         }
 399
 400         /**
 401          * regex without the slashes and what not
 402          *
 403          * @return string
 404          */
 405         function getBaseRegex() {
 406                 if ($this->mRegex == '') {
 407                         $this->initRegex();
 408                 }
 409                 return $this->mBaseRegex;
 410         }
 411
 412         /**
 413          * Returns true if the text contains the word
 414          *
 415          * @param $text string
 416          *
 417          * @return bool
 418          */
 419         function match( $text ) {
 420                 return (bool)preg_match( $this->getRegex(), $text );
 421         }
 422
 423         /**
 424          * Returns true if the text starts with the word
 425          *
 426          * @param $text string
 427          *
 428          * @return bool
 429          */
 430         function matchStart( $text ) {
 431                 return (bool)preg_match( $this->getRegexStart(), $text );
 432         }
 433
 434         /**
 435          * Returns NULL if there's no match, the value of $1 otherwise
 436          * The return code is the matched string, if there's no variable
 437          * part in the regex and the matched variable part ($1) if there
 438          * is one.
 439          *
 440          * @param $text string
 441          *
 442          * @return string
 443          */
 444         function matchVariableStartToEnd( $text ) {
 445                 $matches = array();
 446                 $matchcount = preg_match( $this->getVariableStartToEndRegex(), $text, $matches );
 447                 if ( $matchcount == 0 ) {
 448                         return null;
 449                 } else {
 450                         # multiple matched parts (variable match); some will be empty because of
 451                         # synonyms. The variable will be the second non-empty one so remove any
 452                         # blank elements and re-sort the indices.
 453                         # See also bug 6526
 454
 455                         $matches = array_values(array_filter($matches));
 456
 457                         if ( count($matches) == 1 ) {
 458                                 return $matches[0];
 459                         } else {
 460                                 return $matches[1];
 461                         }
 462                 }
 463         }
 464
 465
 466         /**
 467          * Returns true if the text matches the word, and alters the
 468          * input string, removing all instances of the word
 469          *
 470          * @param $text string
 471          *
 472          * @return bool
 473          */
 474         function matchAndRemove( &$text ) {
 475                 $this->mFound = false;
 476                 $text = preg_replace_callback( $this->getRegex(), array( &$this, 'pregRemoveAndRecord' ), $text );
 477                 return $this->mFound;
 478         }
 479
 480         /**
 481          * @param  $text
 482          * @return bool
 483          */
 484         function matchStartAndRemove( &$text ) {
 485                 $this->mFound = false;
 486                 $text = preg_replace_callback( $this->getRegexStart(), array( &$this, 'pregRemoveAndRecord' ), $text );
 487                 return $this->mFound;
 488         }
 489
 490         /**
 491          * Used in matchAndRemove()
 492          *
 493          * @return string
 494          */
 495         function pregRemoveAndRecord() {
 496                 $this->mFound = true;
 497                 return '';
 498         }
 499
 500         /**
 501          * Replaces the word with something else
 502          *
 503          * @param $replacement
 504          * @param $subject
 505          * @param $limit int
 506          *
 507          * @return string
 508          */
 509         function replace( $replacement, $subject, $limit = -1 ) {
 510                 $res = preg_replace( $this->getRegex(), StringUtils::escapeRegexReplacement( $replacement ), $subject, $limit );
 511                 $this->mModified = !($res === $subject);
 512                 return $res;
 513         }
 514
 515         /**
 516          * Variable handling: {{SUBST:xxx}} style words
 517          * Calls back a function to determine what to replace xxx with
 518          * Input word must contain $1
 519          *
 520          * @param $text string
 521          * @param $callback
 522          *
 523          * @return string
 524          */
 525         function substituteCallback( $text, $callback ) {
 526                 $res = preg_replace_callback( $this->getVariableRegex(), $callback, $text );
 527                 $this->mModified = !($res === $text);
 528                 return $res;
 529         }
 530
 531         /**
 532          * Matches the word, where $1 is a wildcard
 533          *
 534          * @return string
 535          */
 536         function getVariableRegex()     {
 537                 if ( $this->mVariableRegex == '' ) {
 538                         $this->initRegex();
 539                 }
 540                 return $this->mVariableRegex;
 541         }
 542
 543         /**
 544          * Matches the entire string, where $1 is a wildcard
 545          *
 546          * @return string
 547          */
 548         function getVariableStartToEndRegex() {
 549                 if ( $this->mVariableStartToEndRegex == '' ) {
 550                         $this->initRegex();
 551                 }
 552                 return $this->mVariableStartToEndRegex;
 553         }
 554
 555         /**
 556          * Accesses the synonym list directly
 557          *
 558          * @param $i int
 559          *
 560          * @return string
 561          */
 562         function getSynonym( $i ) {
 563                 return $this->mSynonyms[$i];
 564         }
 565
 566         /**
 567          * @return array
 568          */
 569         function getSynonyms() {
 570                 return $this->mSynonyms;
 571         }
 572
 573         /**
 574          * Returns true if the last call to replace() or substituteCallback()
 575          * returned a modified text, otherwise false.
 576          *
 577          * @return bool
 578          */
 579         function getWasModified(){
 580                 return $this->mModified;
 581         }
 582
 583         /**
 584          * $magicarr is an associative array of (magic word ID => replacement)
 585          * This method uses the php feature to do several replacements at the same time,
 586          * thereby gaining some efficiency. The result is placed in the out variable
 587          * $result. The return value is true if something was replaced.
 588          * @todo Should this be static? It doesn't seem to be used at all
 589          *
 590          * @param $magicarr
 591          * @param $subject
 592          * @param $result
 593          *
 594          * @return bool
 595          */
 596         function replaceMultiple( $magicarr, $subject, &$result ){
 597                 $search = array();
 598                 $replace = array();
 599                 foreach( $magicarr as $id => $replacement ){
 600                         $mw = MagicWord::get( $id );
 601                         $search[] = $mw->getRegex();
 602                         $replace[] = $replacement;
 603                 }
 604
 605                 $result = preg_replace( $search, $replace, $subject );
 606                 return !($result === $subject);
 607         }
 608
 609         /**
 610          * Adds all the synonyms of this MagicWord to an array, to allow quick
 611          * lookup in a list of magic words
 612          *
 613          * @param $array
 614          * @param $value
 615          */
 616         function addToArray( &$array, $value ) {
 617                 global $wgContLang;
 618                 foreach ( $this->mSynonyms as $syn ) {
 619                         $array[$wgContLang->lc($syn)] = $value;
 620                 }
 621         }
 622
 623         /**
 624          * @return bool
 625          */
 626         function isCaseSensitive() {
 627                 return $this->mCaseSensitive;
 628         }
 629
 630         /**
 631          * @return int
 632          */
 633         function getId() {
 634                 return $this->mId;
 635         }
 636 }
 637
 638 /**
 639  * Class for handling an array of magic words
 640  * @ingroup Parser
 641  */
 642 class MagicWordArray {
 643         var $names = array();
 644         var $hash;
 645         var $baseRegex, $regex;
 646         var $matches;
 647
 648         function __construct( $names = array() ) {
 649                 $this->names = $names;
 650         }
 651
 652         /**
 653          * Add a magic word by name
 654          *
 655          * @param $name string
 656          */
 657         public function add( $name ) {
 658                 $this->names[] = $name;
 659                 $this->hash = $this->baseRegex = $this->regex = null;
 660         }
 661
 662         /**
 663          * Add a number of magic words by name
 664          *
 665          * @param $names array
 666          */
 667         public function addArray( $names ) {
 668                 $this->names = array_merge( $this->names, array_values( $names ) );
 669                 $this->hash = $this->baseRegex = $this->regex = null;
 670         }
 671
 672         /**
 673          * Get a 2-d hashtable for this array
 674          */
 675         function getHash() {
 676                 if ( is_null( $this->hash ) ) {
 677                         global $wgContLang;
 678                         $this->hash = array( 0 => array(), 1 => array() );
 679                         foreach ( $this->names as $name ) {
 680                                 $magic = MagicWord::get( $name );
 681                                 $case = intval( $magic->isCaseSensitive() );
 682                                 foreach ( $magic->getSynonyms() as $syn ) {
 683                                         if ( !$case ) {
 684                                                 $syn = $wgContLang->lc( $syn );
 685                                         }
 686                                         $this->hash[$case][$syn] = $name;
 687                                 }
 688                         }
 689                 }
 690                 return $this->hash;
 691         }
 692
 693         /**
 694          * Get the base regex
 695          */
 696         function getBaseRegex() {
 697                 if ( is_null( $this->baseRegex ) ) {
 698                         $this->baseRegex = array( 0 => '', 1 => '' );
 699                         foreach ( $this->names as $name ) {
 700                                 $magic = MagicWord::get( $name );
 701                                 $case = intval( $magic->isCaseSensitive() );
 702                                 foreach ( $magic->getSynonyms() as $i => $syn ) {
 703                                         $group = "(?P<{$i}_{$name}>" . preg_quote( $syn, '/' ) . ')';
 704                                         if ( $this->baseRegex[$case] === '' ) {
 705                                                 $this->baseRegex[$case] = $group;
 706                                         } else {
 707                                                 $this->baseRegex[$case] .= '|' . $group;
 708                                         }
 709                                 }
 710                         }
 711                 }
 712                 return $this->baseRegex;
 713         }
 714
 715         /**
 716          * Get an unanchored regex that does not match parameters
 717          */
 718         function getRegex() {
 719                 if ( is_null( $this->regex ) ) {
 720                         $base = $this->getBaseRegex();
 721                         $this->regex = array( '', '' );
 722                         if ( $this->baseRegex[0] !== '' ) {
 723                                 $this->regex[0] = "/{$base[0]}/iuS";
 724                         }
 725                         if ( $this->baseRegex[1] !== '' ) {
 726                                 $this->regex[1] = "/{$base[1]}/S";
 727                         }
 728                 }
 729                 return $this->regex;
 730         }
 731
 732         /**
 733          * Get a regex for matching variables with parameters
 734          *
 735          * @return string
 736          */
 737         function getVariableRegex() {
 738                 return str_replace( "\\$1", "(.*?)", $this->getRegex() );
 739         }
 740
 741         /**
 742          * Get a regex anchored to the start of the string that does not match parameters
 743          *
 744          * @return array
 745          */
 746         function getRegexStart() {
 747                 $base = $this->getBaseRegex();
 748                 $newRegex = array( '', '' );
 749                 if ( $base[0] !== '' ) {
 750                         $newRegex[0] = "/^(?:{$base[0]})/iuS";
 751                 }
 752                 if ( $base[1] !== '' ) {
 753                         $newRegex[1] = "/^(?:{$base[1]})/S";
 754                 }
 755                 return $newRegex;
 756         }
 757
 758         /**
 759          * Get an anchored regex for matching variables with parameters
 760          *
 761          * @return array
 762          */
 763         function getVariableStartToEndRegex() {
 764                 $base = $this->getBaseRegex();
 765                 $newRegex = array( '', '' );
 766                 if ( $base[0] !== '' ) {
 767                         $newRegex[0] = str_replace( "\\$1", "(.*?)", "/^(?:{$base[0]})$/iuS" );
 768                 }
 769                 if ( $base[1] !== '' ) {
 770                         $newRegex[1] = str_replace( "\\$1", "(.*?)", "/^(?:{$base[1]})$/S" );
 771                 }
 772                 return $newRegex;
 773         }
 774
 775         /**
 776          * Parse a match array from preg_match
 777          * Returns array(magic word ID, parameter value)
 778          * If there is no parameter value, that element will be false.
 779          *
 780          * @param $m array
 781          *
 782          * @return array
 783          */
 784         function parseMatch( $m ) {
 785                 reset( $m );
 786                 while ( list( $key, $value ) = each( $m ) ) {
 787                         if ( $key === 0 || $value === '' ) {
 788                                 continue;
 789                         }
 790                         $parts = explode( '_', $key, 2 );
 791                         if ( count( $parts ) != 2 ) {
 792                                 // This shouldn't happen
 793                                 // continue;
 794                                 throw new MWException( __METHOD__ . ': bad parameter name' );
 795                         }
 796                         list( /* $synIndex */, $magicName ) = $parts;
 797                         $paramValue = next( $m );
 798                         return array( $magicName, $paramValue );
 799                 }
 800                 // This shouldn't happen either
 801                 throw new MWException( __METHOD__.': parameter not found' );
 802         }
 803
 804         /**
 805          * Match some text, with parameter capture
 806          * Returns an array with the magic word name in the first element and the
 807          * parameter in the second element.
 808          * Both elements are false if there was no match.
 809          *
 810          * @param $text string
 811          *
 812          * @return array
 813          */
 814         public function matchVariableStartToEnd( $text ) {
 815                 $regexes = $this->getVariableStartToEndRegex();
 816                 foreach ( $regexes as $regex ) {
 817                         if ( $regex !== '' ) {
 818                                 $m = false;
 819                                 if ( preg_match( $regex, $text, $m ) ) {
 820                                         return $this->parseMatch( $m );
 821                                 }
 822                         }
 823                 }
 824                 return array( false, false );
 825         }
 826
 827         /**
 828          * Match some text, without parameter capture
 829          * Returns the magic word name, or false if there was no capture
 830          *
 831          * @param $text string
 832          *
 833          * @return string|bool False on failure
 834          */
 835         public function matchStartToEnd( $text ) {
 836                 $hash = $this->getHash();
 837                 if ( isset( $hash[1][$text] ) ) {
 838                         return $hash[1][$text];
 839                 }
 840                 global $wgContLang;
 841                 $lc = $wgContLang->lc( $text );
 842                 if ( isset( $hash[0][$lc] ) ) {
 843                         return $hash[0][$lc];
 844                 }
 845                 return false;
 846         }
 847
 848         /**
 849          * Returns an associative array, ID => param value, for all items that match
 850          * Removes the matched items from the input string (passed by reference)
 851          *
 852          * @param $text string
 853          *
 854          * @return array
 855          */
 856         public function matchAndRemove( &$text ) {
 857                 $found = array();
 858                 $regexes = $this->getRegex();
 859                 foreach ( $regexes as $regex ) {
 860                         if ( $regex === '' ) {
 861                                 continue;
 862                         }
 863                         preg_match_all( $regex, $text, $matches, PREG_SET_ORDER );
 864                         foreach ( $matches as $m ) {
 865                                 list( $name, $param ) = $this->parseMatch( $m );
 866                                 $found[$name] = $param;
 867                         }
 868                         $text = preg_replace( $regex, '', $text );
 869                 }
 870                 return $found;
 871         }
 872
 873         /**
 874          * Return the ID of the magic word at the start of $text, and remove
 875          * the prefix from $text.
 876          * Return false if no match found and $text is not modified.
 877          * Does not match parameters.
 878          *
 879          * @param $text string
 880          *
 881          * @return int|bool False on failure
 882          */
 883         public function matchStartAndRemove( &$text ) {
 884                 $regexes = $this->getRegexStart();
 885                 foreach ( $regexes as $regex ) {
 886                         if ( $regex === '' ) {
 887                                 continue;
 888                         }
 889                         if ( preg_match( $regex, $text, $m ) ) {
 890                                 list( $id, ) = $this->parseMatch( $m );
 891                                 if ( strlen( $m[0] ) >= strlen( $text ) ) {
 892                                         $text = '';
 893                                 } else {
 894                                         $text = substr( $text, strlen( $m[0] ) );
 895                                 }
 896                                 return $id;
 897                         }
 898                 }
 899                 return false;
 900         }
 901 }