includes/SearchPostgres.php

   1 <?php
   2 # Copyright (C) 2006 Greg Sabino Mullane <greg@turnstep.com>
   3 # http://www.mediawiki.org/
   4 #
   5 # This program is free software; you can redistribute it and/or modify
   6 # it under the terms of the GNU General Public License as published by
   7 # the Free Software Foundation; either version 2 of the License, or
   8 # (at your option) any later version.
   9 #
  10 # This program is distributed in the hope that it will be useful,
  11 # but WITHOUT ANY WARRANTY; without even the implied warranty of
  12 # MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
  13 # GNU General Public License for more details.
  14 #
  15 # You should have received a copy of the GNU General Public License along
  16 # with this program; if not, write to the Free Software Foundation, Inc.,
  17 # 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA.
  18 # http://www.gnu.org/copyleft/gpl.html
  19
  20 ## XXX Better catching of SELECT to_tsquery('the')
  21
  22 /**
  23  * Search engine hook base class for Postgres
  24  * @package MediaWiki
  25  * @subpackage Search
  26  */
  27
  28 /** @package MediaWiki */
  29 class SearchPostgres extends SearchEngine {
  30
  31         function SearchPostgres( &$db ) {
  32                 $this->db =& $db;
  33         }
  34
  35         /**
  36          * Perform a full text search query via tsearch2 and return a result set.
  37          * Currently searches a page's current title (p.page_title) and text (t.old_text)
  38          *
  39          * @param string $term - Raw search term
  40          * @return PostgresSearchResultSet
  41          * @access public
  42          */
  43         function searchText( $term ) {
  44                 $resultSet = $this->db->resultObject( $this->db->query( $this->searchQuery( $term, 'textvector' ) ) );
  45                 return new PostgresSearchResultSet( $resultSet, $this->searchTerms );
  46         }
  47         function searchTitle( $term ) {
  48                 $resultSet = $this->db->resultObject( $this->db->query( $this->searchQuery( $term , 'titlevector' ) ) );
  49                 return new PostgresSearchResultSet( $resultSet, $this->searchTerms );
  50         }
  51
  52
  53         /*
  54          * Transform the user's search string into a better form for tsearch2
  55         */
  56         function parseQuery( $filteredText, $fulltext ) {
  57                 global $wgContLang;
  58                 $lc = SearchEngine::legalSearchChars();
  59                 $searchon = '';
  60                 $this->searchTerms = array();
  61
  62                 # FIXME: This doesn't handle parenthetical expressions.
  63                 $m = array();
  64                 if( preg_match_all( '/([-+<>~]?)(([' . $lc . ']+)(\*?)|"[^"]*")/',
  65                           $filteredText, $m, PREG_SET_ORDER ) ) {
  66                         foreach( $m as $terms ) {
  67                                 if( $searchon !== '' ) $searchon .= ' ';
  68                                 if($terms[1] == '') {
  69                                         $terms[1] = '+';
  70                                 }
  71                                 $searchon .= $terms[1] . $wgContLang->stripForSearch( $terms[2] );
  72                                 if( !empty( $terms[3] ) ) {
  73                                         $regexp = preg_quote( $terms[3], '/' );
  74                                         if( $terms[4] ) $regexp .= "[0-9A-Za-z_]+";
  75                                 } else {
  76                                         $regexp = preg_quote( str_replace( '"', '', $terms[2] ), '/' );
  77                                 }
  78                                 $this->searchTerms[] = $regexp;
  79                         }
  80                         wfDebug( "Would search with '$searchon'\n" );
  81                         wfDebug( 'Match with /\b' . implode( '\b|\b', $this->searchTerms ) . "\b/\n" );
  82                 } else {
  83                         wfDebug( "Can't understand search query '{$this->filteredText}'\n" );
  84                 }
  85
  86                 $searchon = preg_replace('/(\s+)/','&',$searchon);
  87                 $searchon = $this->db->strencode( $searchon );
  88                 return $searchon;
  89         }
  90
  91         /**
  92          * Construct the full SQL query to do the search.
  93          * @param string $filteredTerm
  94          * @param string $fulltext
  95          * @private
  96          */
  97         function searchQuery( $filteredTerm, $fulltext ) {
  98
  99                 $match = $this->parseQuery( $filteredTerm, $fulltext );
 100
 101                 $query = "SELECT page_id, page_namespace, page_title, old_text AS page_text, ".
 102                         "rank(titlevector, to_tsquery('default','$match')) AS rnk ".
 103                         "FROM page p, revision r, pagecontent c WHERE p.page_latest = r.rev_id " .
 104                         "AND r.rev_text_id = c.old_id AND $fulltext @@ to_tsquery('default','$match')";
 105
 106                 ## Redirects
 107                 if (! $this->showRedirects)
 108                         $query .= ' AND page_is_redirect = 0'; ## IS FALSE
 109
 110                 ## Namespaces - defaults to 0
 111                 if ( count($this->namespaces) < 1)
 112                         $query .= ' AND page_namespace = 0';
 113                 else {
 114                         $namespaces = implode( ',', $this->namespaces );
 115                         $query .=  " AND page_namespace IN ($namespaces)";
 116                 }
 117
 118                 $query .= " ORDER BY rnk DESC, page_id DESC";
 119
 120                 $query .= $this->db->limitResult( '', $this->limit, $this->offset );
 121
 122                 return $query;
 123         }
 124
 125         ## These two functions are done automatically via triggers
 126
 127         function update( $pageid, $title, $text ) {
 128                 $dbw =& wfGetDB( DB_MASTER );
 129                 ## We don't want to index older revisions
 130                 $SQL = "UPDATE pagecontent SET textvector = NULL WHERE old_id = ".
 131                                 "(SELECT rev_text_id FROM revision WHERE rev_page = $pageid ".
 132                                 "ORDER BY rev_text_id DESC LIMIT 1 OFFSET 1)";
 133                 $dbw->doQuery($SQL);
 134                 return true;
 135         }
 136     function updateTitle( $id, $title )   { return true; }
 137
 138 } ## end of the SearchPostgres class
 139
 140
 141 /** @package MediaWiki */
 142 class PostgresSearchResultSet extends SearchResultSet {
 143         function PostgresSearchResultSet( $resultSet, $terms ) {
 144                 $this->mResultSet = $resultSet;
 145                 $this->mTerms = $terms;
 146         }
 147
 148         function termMatches() {
 149                 return $this->mTerms;
 150         }
 151
 152         function numRows() {
 153                 return $this->mResultSet->numRows();
 154         }
 155
 156         function next() {
 157                 $row = $this->mResultSet->fetchObject();
 158                 if( $row === false ) {
 159                         return false;
 160                 } else {
 161                         return new SearchResult( $row );
 162                 }
 163         }
 164 }
 165
 166 ?>