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 /**
  21  * Search engine hook base class for Postgres
  22  * @addtogroup Search
  23  */
  24
  25 class SearchPostgres extends SearchEngine {
  26
  27         function SearchPostgres( $db ) {
  28                 $this->db = $db;
  29         }
  30
  31         /**
  32          * Perform a full text search query via tsearch2 and return a result set.
  33          * Currently searches a page's current title (page.page_title) and
  34          * latest revision article text (pagecontent.old_text)
  35          *
  36          * @param string $term - Raw search term
  37          * @return PostgresSearchResultSet
  38          * @access public
  39          */
  40         function searchTitle( $term ) {
  41                 $resultSet = $this->db->resultObject( $this->db->query( $this->searchQuery( $term , 'titlevector', 'page_title' )));
  42                 return new PostgresSearchResultSet( $resultSet, $this->searchTerms );
  43         }
  44         function searchText( $term ) {
  45                 $resultSet = $this->db->resultObject( $this->db->query( $this->searchQuery( $term, 'textvector', 'old_text' )));
  46                 return new PostgresSearchResultSet( $resultSet, $this->searchTerms );
  47         }
  48
  49
  50         /*
  51          * Transform the user's search string into a better form for tsearch2
  52         */
  53         function parseQuery( $term ) {
  54
  55                 wfDebug( "parseQuery received: $term" );
  56
  57                 ## No backslashes allowed
  58                 $term = preg_replace('/\\\/', '', $term);
  59
  60                 ## Collapse parens into nearby words:
  61                 $term = preg_replace('/\s*\(\s*/', ' (', $term);
  62                 $term = preg_replace('/\s*\)\s*/', ') ', $term);
  63
  64                 $this->searchTerms = array();
  65                 $m = array();
  66                 $searchstring = '';
  67                 if( preg_match_all('/([-!]?)(\S+)\s*/', $term, $m, PREG_SET_ORDER ) ) {
  68                         foreach( $m as $terms ) {
  69                                 if (strlen($terms[1])) {
  70                                         $searchstring .= ' & !';
  71                                 }
  72                                 if (strtolower($terms[2]) === 'and') {
  73                                         $searchstring .= ' & ';
  74                                 }
  75                                 else if (strtolower($terms[2]) === 'or' or $terms[2] === '|') {
  76                                         $searchstring .= ' | ';
  77                                 }
  78                                 else if (strtolower($terms[2]) === 'not') {
  79                                         $searchstring .= ' & !';
  80                                 }
  81                                 else {
  82                                         $searchstring .= " & $terms[2]";
  83                                         array_push( $this->searchTerms, $terms[2] );
  84                                 }
  85                         }
  86                 }
  87
  88                 ## Strip out leading junk
  89                 $searchstring = preg_replace('/^[\s\&\|]+/', '', $searchstring);
  90
  91                 ## Remove any doubled-up operators
  92                 $searchstring = preg_replace('/([\!\&\|]) +(?:[\&\|] +)+/', "$1 ", $searchstring);
  93
  94                 ## Quote the whole thing
  95                 $searchstring = $this->db->addQuotes($searchstring);
  96
  97                 wfDebug( "parseQuery returned: $searchstring" );
  98
  99                 return $searchstring;
 100
 101         }
 102
 103         /**
 104          * Construct the full SQL query to do the search.
 105          * @param string $filteredTerm
 106          * @param string $fulltext
 107          * @private
 108          */
 109         function searchQuery( $term, $fulltext, $colname ) {
 110
 111                 $searchstring = $this->parseQuery( $term );
 112
 113                 ## We need a separate query here so gin does not complain about empty searches
 114                 $SQL = "SELECT to_tsquery('default',$searchstring)";
 115                 $res = $this->db->doQuery($SQL);
 116                 if (!$res) {
 117                         ## TODO: Better output (example to catch: one 'two)
 118                         die ("Sorry, that was not a valid search string. Please go back and try again");
 119                 }
 120                 $top = pg_fetch_result($res,0,0);
 121
 122                 if ($top === "") { ## e.g. if only stopwords are used XXX return something better
 123                         $query = "SELECT page_id, page_namespace, page_title, 0 AS score ".
 124                                 "FROM page p, revision r, pagecontent c WHERE p.page_latest = r.rev_id " .
 125                                 "AND r.rev_text_id = c.old_id AND 1=0";
 126                 }
 127                 else {
 128                         $query = "SELECT page_id, page_namespace, page_title, ".
 129                         "rank($fulltext, to_tsquery('default',$searchstring),5) AS score ".
 130                         "FROM page p, revision r, pagecontent c WHERE p.page_latest = r.rev_id " .
 131                         "AND r.rev_text_id = c.old_id AND $fulltext @@ to_tsquery('default',$searchstring)";
 132                 }
 133
 134                 ## Redirects
 135                 if (! $this->showRedirects)
 136                         $query .= ' AND page_is_redirect = 0'; ## IS FALSE
 137
 138                 ## Namespaces - defaults to 0
 139                 if ( count($this->namespaces) < 1)
 140                         $query .= ' AND page_namespace = 0';
 141                 else {
 142                         $namespaces = implode( ',', $this->namespaces );
 143                         $query .= " AND page_namespace IN ($namespaces)";
 144                 }
 145
 146                 $query .= " ORDER BY score DESC, page_id DESC";
 147
 148                 $query .= $this->db->limitResult( '', $this->limit, $this->offset );
 149
 150                 wfDebug( "searchQuery returned: $query" );
 151
 152                 return $query;
 153         }
 154
 155         ## Most of the work of these two functions are done automatically via triggers
 156
 157         function update( $pageid, $title, $text ) {
 158                 ## We don't want to index older revisions
 159                 $SQL = "UPDATE pagecontent SET textvector = NULL WHERE old_id = ".
 160                                 "(SELECT rev_text_id FROM revision WHERE rev_page = $pageid ".
 161                                 "ORDER BY rev_text_id DESC LIMIT 1 OFFSET 1)";
 162                 $this->db->doQuery($SQL);
 163                 return true;
 164         }
 165
 166         function updateTitle( $id, $title ) {
 167                 return true;
 168         }
 169
 170 } ## end of the SearchPostgres class
 171
 172
 173 class PostgresSearchResult extends SearchResult {
 174         function PostgresSearchResult( $row ) {
 175                 $this->mTitle = Title::makeTitle( $row->page_namespace, $row->page_title );
 176                 $this->score = $row->score;
 177         }
 178         function getScore() {
 179                 return $this->score;
 180         }
 181 }
 182
 183 class PostgresSearchResultSet extends SearchResultSet {
 184         function PostgresSearchResultSet( $resultSet, $terms ) {
 185                 $this->mResultSet = $resultSet;
 186                 $this->mTerms = $terms;
 187         }
 188
 189         function termMatches() {
 190                 return $this->mTerms;
 191         }
 192
 193         function numRows() {
 194                 return $this->mResultSet->numRows();
 195         }
 196
 197         function next() {
 198                 $row = $this->mResultSet->fetchObject();
 199                 if( $row === false ) {
 200                         return false;
 201                 } else {
 202                         return new PostgresSearchResult( $row );
 203                 }
 204         }
 205 }
 206
 207
 208 ?>