includes/MagicWord.php

   1 <?php
   2 /**
   3  * File for magic words
   4  * @package MediaWiki
   5  * @subpackage Parser
   6  */
   7
   8 /**
   9  * private
  10  */
  11 $wgMagicFound = false;
  12
  13 /** Actual keyword to be used is set in Language.php */
  14
  15 $magicWords = array(
  16         'MAG_REDIRECT',
  17         'MAG_NOTOC',
  18         'MAG_START',
  19         'MAG_CURRENTMONTH',
  20         'MAG_CURRENTMONTHNAME',
  21         'MAG_CURRENTMONTHNAMEGEN',
  22         'MAG_CURRENTMONTHABBREV',
  23         'MAG_CURRENTDAY',
  24         'MAG_CURRENTDAY2',
  25         'MAG_CURRENTDAYNAME',
  26         'MAG_CURRENTYEAR',
  27         'MAG_CURRENTTIME',
  28         'MAG_NUMBEROFARTICLES',
  29         'MAG_SUBST',
  30         'MAG_MSG',
  31         'MAG_MSGNW',
  32         'MAG_NOEDITSECTION',
  33         'MAG_END',
  34         'MAG_IMG_THUMBNAIL',
  35         'MAG_IMG_RIGHT',
  36         'MAG_IMG_LEFT',
  37         'MAG_IMG_NONE',
  38         'MAG_IMG_WIDTH',
  39         'MAG_IMG_CENTER',
  40         'MAG_INT',
  41         'MAG_FORCETOC',
  42         'MAG_SITENAME',
  43         'MAG_NS',
  44         'MAG_LOCALURL',
  45         'MAG_LOCALURLE',
  46         'MAG_SERVER',
  47         'MAG_IMG_FRAMED',
  48         'MAG_PAGENAME',
  49         'MAG_PAGENAMEE',
  50         'MAG_NAMESPACE',
  51         'MAG_NAMESPACEE',
  52         'MAG_TOC',
  53         'MAG_GRAMMAR',
  54         'MAG_NOTITLECONVERT',
  55         'MAG_NOCONTENTCONVERT',
  56         'MAG_CURRENTWEEK',
  57         'MAG_CURRENTDOW',
  58         'MAG_REVISIONID',
  59         'MAG_SCRIPTPATH',
  60         'MAG_SERVERNAME',
  61         'MAG_NUMBEROFFILES',
  62         'MAG_IMG_MANUALTHUMB',
  63         'MAG_PLURAL',
  64         'MAG_FULLURL',
  65         'MAG_FULLURLE',
  66         'MAG_LCFIRST',
  67         'MAG_UCFIRST',
  68         'MAG_LC',
  69         'MAG_UC',
  70         'MAG_FULLPAGENAME',
  71         'MAG_FULLPAGENAMEE',
  72         'MAG_RAW',
  73         'MAG_SUBPAGENAME',
  74 );
  75 if ( ! defined( 'MEDIAWIKI_INSTALL' ) )
  76         wfRunHooks( 'MagicWordMagicWords', array( &$magicWords ) );
  77
  78 for ( $i = 0; $i < count( $magicWords ); ++$i )
  79         define( $magicWords[$i], $i );
  80
  81 $wgVariableIDs = array(
  82         MAG_CURRENTMONTH,
  83         MAG_CURRENTMONTHNAME,
  84         MAG_CURRENTMONTHNAMEGEN,
  85         MAG_CURRENTMONTHABBREV,
  86         MAG_CURRENTDAY,
  87         MAG_CURRENTDAY2,
  88         MAG_CURRENTDAYNAME,
  89         MAG_CURRENTYEAR,
  90         MAG_CURRENTTIME,
  91         MAG_NUMBEROFARTICLES,
  92         MAG_NUMBEROFFILES,
  93         MAG_SITENAME,
  94         MAG_SERVER,
  95         MAG_SERVERNAME,
  96         MAG_SCRIPTPATH,
  97         MAG_PAGENAME,
  98         MAG_PAGENAMEE,
  99         MAG_FULLPAGENAME,
 100         MAG_FULLPAGENAMEE,
 101         MAG_NAMESPACE,
 102         MAG_NAMESPACEE,
 103         MAG_CURRENTWEEK,
 104         MAG_CURRENTDOW,
 105         MAG_REVISIONID,
 106         MAG_SUBPAGENAME
 107 );
 108 if ( ! defined( 'MEDIAWIKI_INSTALL' ) )
 109         wfRunHooks( 'MagicWordwgVariableIDs', array( &$wgVariableIDs ) );
 110
 111 /**
 112  * This class encapsulates "magic words" such as #redirect, __NOTOC__, etc.
 113  * Usage:
 114  *     if (MagicWord::get( MAG_REDIRECT )->match( $text ) )
 115  *
 116  * Possible future improvements:
 117  *   * Simultaneous searching for a number of magic words
 118  *   * $wgMagicWords in shared memory
 119  *
 120  * Please avoid reading the data out of one of these objects and then writing
 121  * special case code. If possible, add another match()-like function here.
 122  *
 123  * @package MediaWiki
 124  */
 125 class MagicWord {
 126         /**#@+
 127          * @access private
 128          */
 129         var $mId, $mSynonyms, $mCaseSensitive, $mRegex;
 130         var $mRegexStart, $mBaseRegex, $mVariableRegex;
 131         var $mModified;
 132         /**#@-*/
 133
 134         function MagicWord($id = 0, $syn = '', $cs = false) {
 135                 $this->mId = $id;
 136                 $this->mSynonyms = (array)$syn;
 137                 $this->mCaseSensitive = $cs;
 138                 $this->mRegex = '';
 139                 $this->mRegexStart = '';
 140                 $this->mVariableRegex = '';
 141                 $this->mVariableStartToEndRegex = '';
 142                 $this->mModified = false;
 143         }
 144
 145         /**
 146          * Factory: creates an object representing an ID
 147          * @static
 148          */
 149         function &get( $id ) {
 150                 global $wgMagicWords;
 151
 152                 if ( !is_array( $wgMagicWords ) ) {
 153                         wfDebugDieBacktrace( "Incorrect initialisation order, \$wgMagicWords does not exist\n" );
 154                 }
 155                 if (!array_key_exists( $id, $wgMagicWords ) ) {
 156                         $mw = new MagicWord();
 157                         $mw->load( $id );
 158                         $wgMagicWords[$id] = $mw;
 159                 }
 160                 return $wgMagicWords[$id];
 161         }
 162
 163         # Initialises this object with an ID
 164         function load( $id ) {
 165                 global $wgContLang;
 166                 $this->mId = $id;
 167                 $wgContLang->getMagic( $this );
 168         }
 169
 170         /**
 171          * Preliminary initialisation
 172          * @access private
 173          */
 174         function initRegex() {
 175                 #$variableClass = Title::legalChars();
 176                 # This was used for matching "$1" variables, but different uses of the feature will have
 177                 # different restrictions, which should be checked *after* the MagicWord has been matched,
 178                 # not here. - IMSoP
 179
 180                 $escSyn = array();
 181                 foreach ( $this->mSynonyms as $synonym )
 182                         // In case a magic word contains /, like that's going to happen;)
 183                         $escSyn[] = preg_quote( $synonym, '/' );
 184                 $this->mBaseRegex = implode( '|', $escSyn );
 185
 186                 $case = $this->mCaseSensitive ? '' : 'i';
 187                 $this->mRegex = "/{$this->mBaseRegex}/{$case}";
 188                 $this->mRegexStart = "/^(?:{$this->mBaseRegex})/{$case}";
 189                 $this->mVariableRegex = str_replace( "\\$1", "(.*?)", $this->mRegex );
 190                 $this->mVariableStartToEndRegex = str_replace( "\\$1", "(.*?)",
 191                         "/^(?:{$this->mBaseRegex})$/{$case}" );
 192         }
 193
 194         /**
 195          * Gets a regex representing matching the word
 196          */
 197         function getRegex() {
 198                 if ($this->mRegex == '' ) {
 199                         $this->initRegex();
 200                 }
 201                 return $this->mRegex;
 202         }
 203
 204         /**
 205          * Gets the regexp case modifier to use, i.e. i or nothing, to be used if
 206          * one is using MagicWord::getBaseRegex(), otherwise it'll be included in
 207          * the complete expression
 208          */
 209         function getRegexCase() {
 210                 if ( $this->mRegex === '' )
 211                         $this->initRegex();
 212
 213                 return $this->mCaseSensitive ? '' : 'i';
 214         }
 215
 216         /**
 217          * Gets a regex matching the word, if it is at the string start
 218          */
 219         function getRegexStart() {
 220                 if ($this->mRegex == '' ) {
 221                         $this->initRegex();
 222                 }
 223                 return $this->mRegexStart;
 224         }
 225
 226         /**
 227          * regex without the slashes and what not
 228          */
 229         function getBaseRegex() {
 230                 if ($this->mRegex == '') {
 231                         $this->initRegex();
 232                 }
 233                 return $this->mBaseRegex;
 234         }
 235
 236         /**
 237          * Returns true if the text contains the word
 238          * @return bool
 239          */
 240         function match( $text ) {
 241                 return preg_match( $this->getRegex(), $text );
 242         }
 243
 244         /**
 245          * Returns true if the text starts with the word
 246          * @return bool
 247          */
 248         function matchStart( $text ) {
 249                 return preg_match( $this->getRegexStart(), $text );
 250         }
 251
 252         /**
 253          * Returns NULL if there's no match, the value of $1 otherwise
 254          * The return code is the matched string, if there's no variable
 255          * part in the regex and the matched variable part ($1) if there
 256          * is one.
 257          */
 258         function matchVariableStartToEnd( $text ) {
 259                 $matches = array();
 260                 $matchcount = preg_match( $this->getVariableStartToEndRegex(), $text, $matches );
 261                 if ( $matchcount == 0 ) {
 262                         return NULL;
 263                 } elseif ( count($matches) == 1 ) {
 264                         return $matches[0];
 265                 } else {
 266                         # multiple matched parts (variable match); some will be empty because of synonyms
 267                         # the variable will be the second non-empty one so remove any blank elements and re-sort the indices
 268                         $matches = array_values(array_filter($matches));
 269                         return $matches[1];
 270                 }
 271         }
 272
 273
 274         /**
 275          * Returns true if the text matches the word, and alters the
 276          * input string, removing all instances of the word
 277          */
 278         function matchAndRemove( &$text ) {
 279                 global $wgMagicFound;
 280                 $wgMagicFound = false;
 281                 $text = preg_replace_callback( $this->getRegex(), 'pregRemoveAndRecord', $text );
 282                 return $wgMagicFound;
 283         }
 284
 285         function matchStartAndRemove( &$text ) {
 286                 global $wgMagicFound;
 287                 $wgMagicFound = false;
 288                 $text = preg_replace_callback( $this->getRegexStart(), 'pregRemoveAndRecord', $text );
 289                 return $wgMagicFound;
 290         }
 291
 292
 293         /**
 294          * Replaces the word with something else
 295          */
 296         function replace( $replacement, $subject ) {
 297                 $res = preg_replace( $this->getRegex(), wfRegexReplacement( $replacement ), $subject );
 298                 $this->mModified = !($res === $subject);
 299                 return $res;
 300         }
 301
 302         /**
 303          * Variable handling: {{SUBST:xxx}} style words
 304          * Calls back a function to determine what to replace xxx with
 305          * Input word must contain $1
 306          */
 307         function substituteCallback( $text, $callback ) {
 308                 $res = preg_replace_callback( $this->getVariableRegex(), $callback, $text );
 309                 $this->mModified = !($res === $text);
 310                 return $res;
 311         }
 312
 313         /**
 314          * Matches the word, where $1 is a wildcard
 315          */
 316         function getVariableRegex()     {
 317                 if ( $this->mVariableRegex == '' ) {
 318                         $this->initRegex();
 319                 }
 320                 return $this->mVariableRegex;
 321         }
 322
 323         /**
 324          * Matches the entire string, where $1 is a wildcard
 325          */
 326         function getVariableStartToEndRegex() {
 327                 if ( $this->mVariableStartToEndRegex == '' ) {
 328                         $this->initRegex();
 329                 }
 330                 return $this->mVariableStartToEndRegex;
 331         }
 332
 333         /**
 334          * Accesses the synonym list directly
 335          */
 336         function getSynonym( $i ) {
 337                 return $this->mSynonyms[$i];
 338         }
 339
 340         /**
 341          * Returns true if the last call to replace() or substituteCallback()
 342          * returned a modified text, otherwise false.
 343          */
 344         function getWasModified(){
 345                 return $this->mModified;
 346         }
 347
 348         /**
 349          * $magicarr is an associative array of (magic word ID => replacement)
 350          * This method uses the php feature to do several replacements at the same time,
 351          * thereby gaining some efficiency. The result is placed in the out variable
 352          * $result. The return value is true if something was replaced.
 353          * @static
 354          **/
 355         function replaceMultiple( $magicarr, $subject, &$result ){
 356                 $search = array();
 357                 $replace = array();
 358                 foreach( $magicarr as $id => $replacement ){
 359                         $mw = MagicWord::get( $id );
 360                         $search[] = $mw->getRegex();
 361                         $replace[] = $replacement;
 362                 }
 363
 364                 $result = preg_replace( $search, $replace, $subject );
 365                 return !($result === $subject);
 366         }
 367
 368         /**
 369          * Adds all the synonyms of this MagicWord to an array, to allow quick
 370          * lookup in a list of magic words
 371          */
 372         function addToArray( &$array, $value ) {
 373                 foreach ( $this->mSynonyms as $syn ) {
 374                         $array[$syn] = $value;
 375                 }
 376         }
 377 }
 378
 379 /**
 380  * Used in matchAndRemove()
 381  * @access private
 382  **/
 383 function pregRemoveAndRecord( $match ) {
 384         global $wgMagicFound;
 385         $wgMagicFound = true;
 386         return '';
 387 }
 388
 389 ?>