includes/MagicWord.php

   1 <?php
   2
   3 # This class encapsulates "magic words" such as #redirect, __NOTOC__, etc.
   4 # Usage:
   5 #     if (MagicWord::get( MAG_REDIRECT )->match( $text ) )
   6 #
   7 # Possible future improvements:
   8 #   * Simultaneous searching for a number of magic words
   9 #   * $wgMagicWords in shared memory
  10 #
  11 # Please avoid reading the data out of one of these objects and then writing
  12 # special case code. If possible, add another match()-like function here.
  13
  14 /*private*/ $wgMagicFound = false;
  15
  16 class MagicWord {
  17         /*private*/ var $mId, $mSynonyms, $mCaseSensitive, $mRegex;
  18         /*private*/ var $mRegexStart, $mBaseRegex, $mVariableRegex;
  19         /*private*/ var $mModified;
  20
  21         function MagicWord($id = 0, $syn = "", $cs = false)
  22         {
  23                 $this->mId = $id;
  24                 $this->mSynonyms = (array)$syn;
  25                 $this->mCaseSensitive = $cs;
  26                 $this->mRegex = "";
  27                 $this->mRegexStart = "";
  28                 $this->mVariableRegex = "";
  29                 $this->mVariableStartToEndRegex = "";
  30                 $this->mModified = false;
  31         }
  32
  33         # Factory: creates an object representing an ID
  34         /*static*/ function &get( $id )
  35         {
  36                 global $wgMagicWords;
  37
  38                 if (!array_key_exists( $id, $wgMagicWords ) ) {
  39                         $mw = new MagicWord();
  40                         $mw->load( $id );
  41                         $wgMagicWords[$id] = $mw;
  42                 }
  43                 return $wgMagicWords[$id];
  44         }
  45
  46         # Initialises this object with an ID
  47         function load( $id )
  48         {
  49                 global $wgLang;
  50                 $this->mId = $id;
  51                 $wgLang->getMagic( $this );
  52         }
  53
  54         # Preliminary initialisation
  55         /* private */ function initRegex()
  56         {
  57                 $variableClass = "A-Za-z0-9_\-\x80-\xff";
  58                 $escSyn = array_map( "preg_quote", $this->mSynonyms );
  59                 $this->mBaseRegex = implode( "|", $escSyn );
  60                 $case = $this->mCaseSensitive ? "" : "i";
  61                 $this->mRegex = "/{$this->mBaseRegex}/{$case}";
  62                 $this->mRegexStart = "/^{$this->mBaseRegex}/{$case}";
  63                 $this->mVariableRegex = str_replace( "\\$1", "([$variableClass]*)", $this->mRegex );
  64                 $this->mVariableStartToEndRegex = str_replace( "\\$1", "([$variableClass]*)",
  65                         "/^{$this->mBaseRegex}$/{$case}" );
  66         }
  67
  68         # Gets a regex representing matching the word
  69         function getRegex()
  70         {
  71                 if ($this->mRegex == "" ) {
  72                         $this->initRegex();
  73                 }
  74                 return $this->mRegex;
  75         }
  76
  77         # Gets a regex matching the word, if it is at the
  78         # string start
  79         function getRegexStart()
  80         {
  81                 if ($this->mRegex == "" ) {
  82                         $this->initRegex();
  83                 }
  84                 return $this->mRegexStart;
  85         }
  86
  87         # regex without the slashes and what not
  88         function getBaseRegex()
  89         {
  90                 if ($this->mRegex == "") {
  91                         $this->initRegex();
  92                 }
  93                 return $this->mBaseRegex;
  94         }
  95
  96         # Returns true if the text contains the word
  97         function match( $text ) {
  98                 return preg_match( $this->getRegex(), $text );
  99         }
 100
 101         # Returns true if the text starts with the word
 102         function matchStart( $text )
 103         {
 104                 return preg_match( $this->getRegexStart(), $text );
 105         }
 106
 107         # Returns NULL if there's no match, the value of $1 otherwise
 108         # The return code is the matched string, if there's no variable
 109         # part in the regex and the matched variable part ($1) if there
 110         # is one.
 111         function matchVariableStartToEnd( $text ) {
 112                 $matchcount = preg_match( $this->getVariableStartToEndRegex(), $text, $matches );
 113                 if ( $matchcount == 0 ) {
 114                         return NULL;
 115                 } elseif ( count($matches) == 1 ) {
 116                         return $matches[0];
 117                 } else {
 118                         return $matches[1];
 119                 }
 120         }
 121
 122
 123         # Returns true if the text matches the word, and alters the
 124         # input string, removing all instances of the word
 125         function matchAndRemove( &$text )
 126         {
 127                 global $wgMagicFound;
 128                 $wgMagicFound = false;
 129                 $text = preg_replace_callback( $this->getRegex(), "pregRemoveAndRecord", $text );
 130                 return $wgMagicFound;
 131         }
 132
 133         # Replaces the word with something else
 134         function replace( $replacement, $subject )
 135         {
 136                 $res = preg_replace( $this->getRegex(), $replacement, $subject );
 137                 $this->mModified = !($res === $subject);
 138                 return $res;
 139         }
 140
 141         # Variable handling: {{SUBST:xxx}} style words
 142         # Calls back a function to determine what to replace xxx with
 143         # Input word must contain $1
 144         function substituteCallback( $text, $callback ) {
 145                 $regex = $this->getVariableRegex();
 146                 $res = preg_replace_callback( $this->getVariableRegex(), $callback, $text );
 147                 $this->mModified = !($res === $text);
 148                 return $res;
 149         }
 150
 151         # Matches the word, where $1 is a wildcard
 152         function getVariableRegex()
 153         {
 154                 if ( $this->mVariableRegex == "" ) {
 155                         $this->initRegex();
 156                 }
 157                 return $this->mVariableRegex;
 158         }
 159
 160         # Matches the entire string, where $1 is a wildcard
 161         function getVariableStartToEndRegex()
 162         {
 163                 if ( $this->mVariableStartToEndRegex == "" ) {
 164                         $this->initRegex();
 165                 }
 166                 return $this->mVariableStartToEndRegex;
 167         }
 168
 169         # Accesses the synonym list directly
 170         function getSynonym( $i ) {
 171                 return $this->mSynonyms[$i];
 172         }
 173
 174         # Returns true if the last call to replace() or substituteCallback()
 175         # returned a modified text, otherwise false.
 176         function getWasModified(){
 177                 return $this->mModified;
 178         }
 179
 180         # $magicarr is an associative array of (magic word ID => replacement)
 181         # This method uses the php feature to do several replacements at the same time,
 182         # thereby gaining some efficiency. The result is placed in the out variable
 183         # $result. The return value is true if something was replaced.
 184
 185         /* static */ function replaceMultiple( $magicarr, $subject, &$result ){
 186                 $search = array();
 187                 $replace = array();
 188                 foreach( $magicarr as $id => $replacement ){
 189                         $mw = MagicWord::get( $id );
 190                         $search[] = $mw->getRegex();
 191                         $replace[] = $replacement;
 192                 }
 193
 194                 $result = preg_replace( $search, $replace, $subject );
 195                 return !($result === $subject);
 196         }
 197 }
 198
 199 # Used in matchAndRemove()
 200 /*private*/ function pregRemoveAndRecord( $match )
 201 {
 202         global $wgMagicFound;
 203         $wgMagicFound = true;
 204         return "";
 205 }
 206
 207 ?>