includes/MagicWord.php

   1 <?php
   2 /**
   3  * File for magic words
   4  * @package MediaWiki
   5  * @subpackage Parser
   6  */
   7
   8 /**
   9  * This class encapsulates "magic words" such as #redirect, __NOTOC__, etc.
  10  * Usage:
  11  *     if (MagicWord::get( 'redirect' )->match( $text ) )
  12  *
  13  * Possible future improvements:
  14  *   * Simultaneous searching for a number of magic words
  15  *   * MagicWord::$mObjects in shared memory
  16  *
  17  * Please avoid reading the data out of one of these objects and then writing
  18  * special case code. If possible, add another match()-like function here.
  19  *
  20  * To add magic words in an extension, use the LanguageGetMagic hook. For
  21  * magic words which are also Parser variables, add a MagicWordwgVariableIDs
  22  * hook. Use string keys.
  23  *
  24  * @package MediaWiki
  25  */
  26 class MagicWord {
  27         /**#@+
  28          * @private
  29          */
  30         var $mId, $mSynonyms, $mCaseSensitive, $mRegex;
  31         var $mRegexStart, $mBaseRegex, $mVariableRegex;
  32         var $mModified, $mFound;
  33
  34         static public $mVariableIDsInitialised = false;
  35         static public $mVariableIDs = array(
  36                 'currentmonth',
  37                 'currentmonthname',
  38                 'currentmonthnamegen',
  39                 'currentmonthabbrev',
  40                 'currentday',
  41                 'currentday2',
  42                 'currentdayname',
  43                 'currentyear',
  44                 'currenttime',
  45                 'numberofarticles',
  46                 'numberoffiles',
  47                 'sitename',
  48                 'server',
  49                 'servername',
  50                 'scriptpath',
  51                 'pagename',
  52                 'pagenamee',
  53                 'fullpagename',
  54                 'fullpagenamee',
  55                 'namespace',
  56                 'namespacee',
  57                 'currentweek',
  58                 'currentdow',
  59                 'revisionid',
  60                 'subpagename',
  61                 'subpagenamee',
  62                 'displaytitle',
  63                 'talkspace',
  64                 'talkspacee',
  65                 'subjectspace',
  66                 'subjectspacee',
  67                 'talkpagename',
  68                 'talkpagenamee',
  69                 'subjectpagename',
  70                 'subjectpagenamee',
  71                 'numberofusers',
  72                 'rawsuffix',
  73                 'newsectionlink',
  74                 'numberofpages',
  75                 'currentversion',
  76                 'basepagename',
  77                 'basepagenamee',
  78                 'urlencode',
  79                 'currenttimestamp',
  80                 'directionmark',
  81                 'language',
  82                 'contentlanguage',
  83                 'pagesinnamespace',
  84                 'numberofadmins',
  85         );
  86
  87         static public $mObjects = array();
  88
  89         /**#@-*/
  90
  91         function MagicWord($id = 0, $syn = '', $cs = false) {
  92                 $this->mId = $id;
  93                 $this->mSynonyms = (array)$syn;
  94                 $this->mCaseSensitive = $cs;
  95                 $this->mRegex = '';
  96                 $this->mRegexStart = '';
  97                 $this->mVariableRegex = '';
  98                 $this->mVariableStartToEndRegex = '';
  99                 $this->mModified = false;
 100         }
 101
 102         /**
 103          * Factory: creates an object representing an ID
 104          * @static
 105          */
 106         static function &get( $id ) {
 107                 if (!array_key_exists( $id, self::$mObjects ) ) {
 108                         $mw = new MagicWord();
 109                         $mw->load( $id );
 110                         self::$mObjects[$id] = $mw;
 111                 }
 112                 return self::$mObjects[$id];
 113         }
 114
 115         /**
 116          * Get an array of parser variable IDs
 117          */
 118         static function getVariableIDs() {
 119                 if ( !self::$mVariableIDsInitialised ) {
 120                         # Deprecated constant definition hook, available for extensions that need it
 121                         $magicWords = array();
 122                         wfRunHooks( 'MagicWordMagicWords', array( &$magicWords ) );
 123                         foreach ( $magicWords as $word ) {
 124                                 define( $word, $word );
 125                         }
 126
 127                         # Get variable IDs
 128                         wfRunHooks( 'MagicWordwgVariableIDs', array( &self::$mVariableIDs ) );
 129                         self::$mVariableIDsInitialised = true;
 130                 }
 131                 return self::$mVariableIDs;
 132         }
 133
 134         # Initialises this object with an ID
 135         function load( $id ) {
 136                 global $wgContLang;
 137                 $this->mId = $id;
 138                 $wgContLang->getMagic( $this );
 139         }
 140
 141         /**
 142          * Preliminary initialisation
 143          * @private
 144          */
 145         function initRegex() {
 146                 #$variableClass = Title::legalChars();
 147                 # This was used for matching "$1" variables, but different uses of the feature will have
 148                 # different restrictions, which should be checked *after* the MagicWord has been matched,
 149                 # not here. - IMSoP
 150
 151                 $escSyn = array();
 152                 foreach ( $this->mSynonyms as $synonym )
 153                         // In case a magic word contains /, like that's going to happen;)
 154                         $escSyn[] = preg_quote( $synonym, '/' );
 155                 $this->mBaseRegex = implode( '|', $escSyn );
 156
 157                 $case = $this->mCaseSensitive ? '' : 'i';
 158                 $this->mRegex = "/{$this->mBaseRegex}/{$case}";
 159                 $this->mRegexStart = "/^(?:{$this->mBaseRegex})/{$case}";
 160                 $this->mVariableRegex = str_replace( "\\$1", "(.*?)", $this->mRegex );
 161                 $this->mVariableStartToEndRegex = str_replace( "\\$1", "(.*?)",
 162                         "/^(?:{$this->mBaseRegex})$/{$case}" );
 163         }
 164
 165         /**
 166          * Gets a regex representing matching the word
 167          */
 168         function getRegex() {
 169                 if ($this->mRegex == '' ) {
 170                         $this->initRegex();
 171                 }
 172                 return $this->mRegex;
 173         }
 174
 175         /**
 176          * Gets the regexp case modifier to use, i.e. i or nothing, to be used if
 177          * one is using MagicWord::getBaseRegex(), otherwise it'll be included in
 178          * the complete expression
 179          */
 180         function getRegexCase() {
 181                 if ( $this->mRegex === '' )
 182                         $this->initRegex();
 183
 184                 return $this->mCaseSensitive ? '' : 'i';
 185         }
 186
 187         /**
 188          * Gets a regex matching the word, if it is at the string start
 189          */
 190         function getRegexStart() {
 191                 if ($this->mRegex == '' ) {
 192                         $this->initRegex();
 193                 }
 194                 return $this->mRegexStart;
 195         }
 196
 197         /**
 198          * regex without the slashes and what not
 199          */
 200         function getBaseRegex() {
 201                 if ($this->mRegex == '') {
 202                         $this->initRegex();
 203                 }
 204                 return $this->mBaseRegex;
 205         }
 206
 207         /**
 208          * Returns true if the text contains the word
 209          * @return bool
 210          */
 211         function match( $text ) {
 212                 return preg_match( $this->getRegex(), $text );
 213         }
 214
 215         /**
 216          * Returns true if the text starts with the word
 217          * @return bool
 218          */
 219         function matchStart( $text ) {
 220                 return preg_match( $this->getRegexStart(), $text );
 221         }
 222
 223         /**
 224          * Returns NULL if there's no match, the value of $1 otherwise
 225          * The return code is the matched string, if there's no variable
 226          * part in the regex and the matched variable part ($1) if there
 227          * is one.
 228          */
 229         function matchVariableStartToEnd( $text ) {
 230                 $matches = array();
 231                 $matchcount = preg_match( $this->getVariableStartToEndRegex(), $text, $matches );
 232                 if ( $matchcount == 0 ) {
 233                         return NULL;
 234                 } else {
 235                         # multiple matched parts (variable match); some will be empty because of
 236                         # synonyms. The variable will be the second non-empty one so remove any
 237                         # blank elements and re-sort the indices.
 238                         # See also bug 6526
 239
 240                         $matches = array_values(array_filter($matches));
 241
 242                         if ( count($matches) == 1 ) { return $matches[0]; }
 243                         else { return $matches[1]; }
 244                 }
 245         }
 246
 247
 248         /**
 249          * Returns true if the text matches the word, and alters the
 250          * input string, removing all instances of the word
 251          */
 252         function matchAndRemove( &$text ) {
 253                 $this->mFound = false;
 254                 $text = preg_replace_callback( $this->getRegex(), array( &$this, 'pregRemoveAndRecord' ), $text );
 255                 return $this->mFound;
 256         }
 257
 258         function matchStartAndRemove( &$text ) {
 259                 $this->mFound = false;
 260                 $text = preg_replace_callback( $this->getRegexStart(), array( &$this, 'pregRemoveAndRecord' ), $text );
 261                 return $this->mFound;
 262         }
 263
 264         /**
 265          * Used in matchAndRemove()
 266          * @private
 267          **/
 268         function pregRemoveAndRecord( $match ) {
 269                 $this->mFound = true;
 270                 return '';
 271         }
 272
 273         /**
 274          * Replaces the word with something else
 275          */
 276         function replace( $replacement, $subject, $limit=-1 ) {
 277                 $res = preg_replace( $this->getRegex(), wfRegexReplacement( $replacement ), $subject, $limit );
 278                 $this->mModified = !($res === $subject);
 279                 return $res;
 280         }
 281
 282         /**
 283          * Variable handling: {{SUBST:xxx}} style words
 284          * Calls back a function to determine what to replace xxx with
 285          * Input word must contain $1
 286          */
 287         function substituteCallback( $text, $callback ) {
 288                 $res = preg_replace_callback( $this->getVariableRegex(), $callback, $text );
 289                 $this->mModified = !($res === $text);
 290                 return $res;
 291         }
 292
 293         /**
 294          * Matches the word, where $1 is a wildcard
 295          */
 296         function getVariableRegex()     {
 297                 if ( $this->mVariableRegex == '' ) {
 298                         $this->initRegex();
 299                 }
 300                 return $this->mVariableRegex;
 301         }
 302
 303         /**
 304          * Matches the entire string, where $1 is a wildcard
 305          */
 306         function getVariableStartToEndRegex() {
 307                 if ( $this->mVariableStartToEndRegex == '' ) {
 308                         $this->initRegex();
 309                 }
 310                 return $this->mVariableStartToEndRegex;
 311         }
 312
 313         /**
 314          * Accesses the synonym list directly
 315          */
 316         function getSynonym( $i ) {
 317                 return $this->mSynonyms[$i];
 318         }
 319
 320         function getSynonyms() {
 321                 return $this->mSynonyms;
 322         }
 323
 324         /**
 325          * Returns true if the last call to replace() or substituteCallback()
 326          * returned a modified text, otherwise false.
 327          */
 328         function getWasModified(){
 329                 return $this->mModified;
 330         }
 331
 332         /**
 333          * $magicarr is an associative array of (magic word ID => replacement)
 334          * This method uses the php feature to do several replacements at the same time,
 335          * thereby gaining some efficiency. The result is placed in the out variable
 336          * $result. The return value is true if something was replaced.
 337          * @static
 338          **/
 339         function replaceMultiple( $magicarr, $subject, &$result ){
 340                 $search = array();
 341                 $replace = array();
 342                 foreach( $magicarr as $id => $replacement ){
 343                         $mw = MagicWord::get( $id );
 344                         $search[] = $mw->getRegex();
 345                         $replace[] = $replacement;
 346                 }
 347
 348                 $result = preg_replace( $search, $replace, $subject );
 349                 return !($result === $subject);
 350         }
 351
 352         /**
 353          * Adds all the synonyms of this MagicWord to an array, to allow quick
 354          * lookup in a list of magic words
 355          */
 356         function addToArray( &$array, $value ) {
 357                 foreach ( $this->mSynonyms as $syn ) {
 358                         $array[$syn] = $value;
 359                 }
 360         }
 361
 362         function isCaseSensitive() {
 363                 return $this->mCaseSensitive;
 364         }
 365 }
 366
 367 ?>