includes/SearchPostgres.php

   1 <?php
   2 # Copyright (C) 2006-2007 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  * @addtogroup Search
  25  */
  26
  27 class SearchPostgres extends SearchEngine {
  28
  29         function SearchPostgres( $db ) {
  30                 $this->db = $db;
  31         }
  32
  33         /**
  34          * Perform a full text search query via tsearch2 and return a result set.
  35          * Currently searches a page's current title (p.page_title) and text (t.old_text)
  36          *
  37          * @param string $term - Raw search term
  38          * @return PostgresSearchResultSet
  39          * @access public
  40          */
  41         function searchText( $term ) {
  42                 $resultSet = $this->db->resultObject( $this->db->query( $this->searchQuery( $term, 'textvector' ) ) );
  43                 return new PostgresSearchResultSet( $resultSet, $this->searchTerms );
  44         }
  45         function searchTitle( $term ) {
  46                 $resultSet = $this->db->resultObject( $this->db->query( $this->searchQuery( $term , 'titlevector' ) ) );
  47                 return new PostgresSearchResultSet( $resultSet, $this->searchTerms );
  48         }
  49
  50
  51         /*
  52          * Transform the user's search string into a better form for tsearch2
  53         */
  54         function parseQuery( $filteredText, $fulltext ) {
  55                 global $wgContLang;
  56                 $lc = SearchEngine::legalSearchChars();
  57                 $searchon = '';
  58                 $searchconst = '';
  59                 $this->searchTerms = array();
  60
  61                 # FIXME: This doesn't handle parenthetical expressions.
  62                 $m = array();
  63                 if( preg_match_all( '/([-+<>~]?)(([' . $lc . ']+)(\*?)|"[^"]*")/',
  64                           $filteredText, $m, PREG_SET_ORDER ) ) {
  65                         foreach( $m as $terms ) {
  66                 switch ($terms[1]) {
  67                   case '~':  // negate but do not exclude like "-" for now, simply negate
  68                   case '-': $searchconst.= '&!' . $wgContLang->stripForSearch( $terms[2] );
  69                                         break;
  70                   case '+': $searchconst.= '&'  . $wgContLang->stripForSearch( $terms[2] );
  71                                         break;
  72                   case '<':  // decrease priority of word - not implemented
  73                   case '>':  // increase priority of word - not implemented
  74                   default : $searchon.=    '|'  . $wgContLang->stripForSearch( $terms[2] );
  75                 }
  76                                 if( !empty( $terms[3] ) ) {
  77                                         $regexp = preg_quote( $terms[3], '/' );
  78                                         if( $terms[4] ) $regexp .= "[0-9A-Za-z_]+";
  79                                 } else {
  80                                         $regexp = preg_quote( str_replace( '"', '', $terms[2] ), '/' );
  81                                 }
  82                                 $this->searchTerms[] = $regexp;
  83                         }
  84                         wfDebug( "Would search with '$searchon'\n" );
  85                         wfDebug( 'Match with /\b' . implode( '\b|\b', $this->searchTerms ) . "\b/\n" );
  86                 } else {
  87                         wfDebug( "Can't understand search query '{$this->filteredText}'\n" );
  88                 }
  89
  90                 if (substr_count($searchon,'|')==1) {
  91                         $searchon = str_replace ('|','&',$searchon);
  92                 }
  93                 $searchon = substr($searchconst . $searchon, 1) ;
  94                 $searchon = preg_replace('/(\s+)/','&',$searchon);
  95                 $searchon = $this->db->strencode( $searchon );
  96                 return $searchon;
  97         }
  98
  99         /**
 100          * Construct the full SQL query to do the search.
 101          * @param string $filteredTerm
 102          * @param string $fulltext
 103          * @private
 104          */
 105         function searchQuery( $filteredTerm, $fulltext ) {
 106
 107                 $match = $this->parseQuery( $filteredTerm, $fulltext );
 108
 109                 $query = "SELECT page_id, page_namespace, page_title, old_text AS page_text, ".
 110                         "rank(titlevector, to_tsquery('default','$match')) AS rnk ".
 111                         "FROM page p, revision r, pagecontent c WHERE p.page_latest = r.rev_id " .
 112                         "AND r.rev_text_id = c.old_id AND $fulltext @@ to_tsquery('default','$match')";
 113
 114                 ## Redirects
 115                 if (! $this->showRedirects)
 116                         $query .= ' AND page_is_redirect = 0'; ## IS FALSE
 117
 118                 ## Namespaces - defaults to 0
 119                 if ( count($this->namespaces) < 1)
 120                         $query .= ' AND page_namespace = 0';
 121                 else {
 122                         $namespaces = implode( ',', $this->namespaces );
 123                         $query .=  " AND page_namespace IN ($namespaces)";
 124                 }
 125
 126                 $query .= " ORDER BY rnk DESC, page_id DESC";
 127
 128                 $query .= $this->db->limitResult( '', $this->limit, $this->offset );
 129
 130                 return $query;
 131         }
 132
 133         ## These two functions are done automatically via triggers
 134
 135         function update( $pageid, $title, $text ) {
 136                 $dbw = wfGetDB( DB_MASTER );
 137                 ## We don't want to index older revisions
 138                 $SQL = "UPDATE pagecontent SET textvector = NULL WHERE old_id = ".
 139                                 "(SELECT rev_text_id FROM revision WHERE rev_page = $pageid ".
 140                                 "ORDER BY rev_text_id DESC LIMIT 1 OFFSET 1)";
 141                 $dbw->doQuery($SQL);
 142                 return true;
 143         }
 144     function updateTitle( $id, $title )   { return true; }
 145
 146 } ## end of the SearchPostgres class
 147
 148
 149 class PostgresSearchResultSet extends SearchResultSet {
 150         function PostgresSearchResultSet( $resultSet, $terms ) {
 151                 $this->mResultSet = $resultSet;
 152                 $this->mTerms = $terms;
 153         }
 154
 155         function termMatches() {
 156                 return $this->mTerms;
 157         }
 158
 159         function numRows() {
 160                 return $this->mResultSet->numRows();
 161         }
 162
 163         function next() {
 164                 $row = $this->mResultSet->fetchObject();
 165                 if( $row === false ) {
 166                         return false;
 167                 } else {
 168                         return new SearchResult( $row );
 169                 }
 170         }
 171 }
 172
 173 ?>