includes/search/SearchUpdate.php

   1 <?php
   2 /**
   3  * Search index updater
   4  *
   5  * See deferred.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 Search
  24  */
  25
  26 /**
  27  * Database independant search index updater
  28  *
  29  * @ingroup Search
  30  */
  31 class SearchUpdate implements DeferrableUpdate {
  32
  33         private $mId = 0, $mNamespace, $mTitle, $mText;
  34         private $mTitleWords;
  35
  36         public function __construct( $id, $title, $text = false ) {
  37                 if ( is_string( $title ) ) {
  38                         $nt = Title::newFromText( $title );
  39                 } else {
  40                         $nt = $title;
  41                 }
  42
  43                 if ( $nt ) {
  44                         $this->mId = $id;
  45                         $this->mText = $text;
  46
  47                         $this->mNamespace = $nt->getNamespace();
  48                         $this->mTitle = $nt->getText(); # Discard namespace
  49
  50                         $this->mTitleWords = $this->mTextWords = array();
  51                 } else {
  52                         wfDebug( "SearchUpdate object created with invalid title '$title'\n" );
  53                 }
  54         }
  55
  56         public function doUpdate() {
  57                 global $wgDisableSearchUpdate;
  58
  59                 if ( $wgDisableSearchUpdate || !$this->mId ) {
  60                         return;
  61                 }
  62
  63                 wfProfileIn( __METHOD__ );
  64
  65                 $search = SearchEngine::create();
  66                 $normalTitle = $search->normalizeText( Title::indexTitle( $this->mNamespace, $this->mTitle ) );
  67
  68                 if ( WikiPage::newFromId( $this->mId ) === null ) {
  69                         $search->delete( $this->mId, $normalTitle );
  70                         wfProfileOut( __METHOD__ );
  71                         return;
  72                 } elseif ( $this->mText === false ) {
  73                         $search->updateTitle( $this->mId, $normalTitle );
  74                         wfProfileOut( __METHOD__ );
  75                         return;
  76                 }
  77
  78                 $text = self::updateText( $this->mText );
  79
  80                 wfRunHooks( 'SearchUpdate', array( $this->mId, $this->mNamespace, $this->mTitle, &$text ) );
  81
  82                 # Perform the actual update
  83                 $search->update( $this->mId, $normalTitle, $search->normalizeText( $text ) );
  84
  85                 wfProfileOut( __METHOD__ );
  86         }
  87
  88         public static function updateText( $text ) {
  89                 global $wgContLang;
  90
  91                 # Language-specific strip/conversion
  92                 $text = $wgContLang->normalizeForSearch( $text );
  93                 $lc = SearchEngine::legalSearchChars() . '&#;';
  94
  95                 wfProfileIn( __METHOD__ . '-regexps' );
  96                 $text = preg_replace( "/<\\/?\\s*[A-Za-z][^>]*?>/",
  97                         ' ', $wgContLang->lc( " " . $text . " " ) ); # Strip HTML markup
  98                 $text = preg_replace( "/(^|\\n)==\\s*([^\\n]+)\\s*==(\\s)/sD",
  99                         "\\1\\2 \\2 \\2\\3", $text ); # Emphasize headings
 100
 101                 # Strip external URLs
 102                 $uc = "A-Za-z0-9_\\/:.,~%\\-+&;#?!=()@\\x80-\\xFF";
 103                 $protos = "http|https|ftp|mailto|news|gopher";
 104                 $pat = "/(^|[^\\[])({$protos}):[{$uc}]+([^{$uc}]|$)/";
 105                 $text = preg_replace( $pat, "\\1 \\3", $text );
 106
 107                 $p1 = "/([^\\[])\\[({$protos}):[{$uc}]+]/";
 108                 $p2 = "/([^\\[])\\[({$protos}):[{$uc}]+\\s+([^\\]]+)]/";
 109                 $text = preg_replace( $p1, "\\1 ", $text );
 110                 $text = preg_replace( $p2, "\\1 \\3 ", $text );
 111
 112                 # Internal image links
 113                 $pat2 = "/\\[\\[image:([{$uc}]+)\\.(gif|png|jpg|jpeg)([^{$uc}])/i";
 114                 $text = preg_replace( $pat2, " \\1 \\3", $text );
 115
 116                 $text = preg_replace( "/([^{$lc}])([{$lc}]+)]]([a-z]+)/",
 117                         "\\1\\2 \\2\\3", $text ); # Handle [[game]]s
 118
 119                 # Strip all remaining non-search characters
 120                 $text = preg_replace( "/[^{$lc}]+/", " ", $text );
 121
 122                 # Handle 's, s'
 123                 #
 124                 #   $text = preg_replace( "/([{$lc}]+)'s /", "\\1 \\1's ", $text );
 125                 #   $text = preg_replace( "/([{$lc}]+)s' /", "\\1s ", $text );
 126                 #
 127                 # These tail-anchored regexps are insanely slow. The worst case comes
 128                 # when Japanese or Chinese text (ie, no word spacing) is written on
 129                 # a wiki configured for Western UTF-8 mode. The Unicode characters are
 130                 # expanded to hex codes and the "words" are very long paragraph-length
 131                 # monstrosities. On a large page the above regexps may take over 20
 132                 # seconds *each* on a 1GHz-level processor.
 133                 #
 134                 # Following are reversed versions which are consistently fast
 135                 # (about 3 milliseconds on 1GHz-level processor).
 136                 #
 137                 $text = strrev( preg_replace( "/ s'([{$lc}]+)/", " s'\\1 \\1", strrev( $text ) ) );
 138                 $text = strrev( preg_replace( "/ 's([{$lc}]+)/", " s\\1", strrev( $text ) ) );
 139
 140                 # Strip wiki '' and '''
 141                 $text = preg_replace( "/''[']*/", " ", $text );
 142                 wfProfileOut( __METHOD__ . '-regexps' );
 143                 return $text;
 144         }
 145 }
 146
 147 /**
 148  * Placeholder class
 149  *
 150  * @ingroup Search
 151  */
 152 class SearchUpdateMyISAM extends SearchUpdate {
 153         # Inherits everything
 154 }