MediaWiki  REL1_31
SearchPostgres.php
Go to the documentation of this file.
1 <?php
40  function searchTitle( $term ) {
41  $q = $this->searchQuery( $term, 'titlevector', 'page_title' );
42  $olderror = error_reporting( E_ERROR );
43  $resultSet = $this->db->query( $q, 'SearchPostgres', true );
44  error_reporting( $olderror );
45  return new SqlSearchResultSet( $resultSet, $this->searchTerms );
46  }
47 
48  function searchText( $term ) {
49  $q = $this->searchQuery( $term, 'textvector', 'old_text' );
50  $olderror = error_reporting( E_ERROR );
51  $resultSet = $this->db->query( $q, 'SearchPostgres', true );
52  error_reporting( $olderror );
53  return new SqlSearchResultSet( $resultSet, $this->searchTerms );
54  }
55 
64  function parseQuery( $term ) {
65  wfDebug( "parseQuery received: $term \n" );
66 
67  # # No backslashes allowed
68  $term = preg_replace( '/\\\/', '', $term );
69 
70  # # Collapse parens into nearby words:
71  $term = preg_replace( '/\s*\‍(\s*/', ' (', $term );
72  $term = preg_replace( '/\s*\‍)\s*/', ') ', $term );
73 
74  # # Treat colons as word separators:
75  $term = preg_replace( '/:/', ' ', $term );
76 
77  $searchstring = '';
78  $m = [];
79  if ( preg_match_all( '/([-!]?)(\S+)\s*/', $term, $m, PREG_SET_ORDER ) ) {
80  foreach ( $m as $terms ) {
81  if ( strlen( $terms[1] ) ) {
82  $searchstring .= ' & !';
83  }
84  if ( strtolower( $terms[2] ) === 'and' ) {
85  $searchstring .= ' & ';
86  } elseif ( strtolower( $terms[2] ) === 'or' || $terms[2] === '|' ) {
87  $searchstring .= ' | ';
88  } elseif ( strtolower( $terms[2] ) === 'not' ) {
89  $searchstring .= ' & !';
90  } else {
91  $searchstring .= " & $terms[2]";
92  }
93  }
94  }
95 
96  # # Strip out leading junk
97  $searchstring = preg_replace( '/^[\s\&\|]+/', '', $searchstring );
98 
99  # # Remove any doubled-up operators
100  $searchstring = preg_replace( '/([\!\&\|]) +(?:[\&\|] +)+/', "$1 ", $searchstring );
101 
102  # # Remove any non-spaced operators (e.g. "Zounds!")
103  $searchstring = preg_replace( '/([^ ])[\!\&\|]/', "$1", $searchstring );
104 
105  # # Remove any trailing whitespace or operators
106  $searchstring = preg_replace( '/[\s\!\&\|]+$/', '', $searchstring );
107 
108  # # Remove unnecessary quotes around everything
109  $searchstring = preg_replace( '/^[\'"](.*)[\'"]$/', "$1", $searchstring );
110 
111  # # Quote the whole thing
112  $searchstring = $this->db->addQuotes( $searchstring );
113 
114  wfDebug( "parseQuery returned: $searchstring \n" );
115 
116  return $searchstring;
117  }
118 
126  function searchQuery( $term, $fulltext, $colname ) {
127  # Get the SQL fragment for the given term
128  $searchstring = $this->parseQuery( $term );
129 
130  # # We need a separate query here so gin does not complain about empty searches
131  $sql = "SELECT to_tsquery($searchstring)";
132  $res = $this->db->query( $sql );
133  if ( !$res ) {
134  # # TODO: Better output (example to catch: one 'two)
135  die( "Sorry, that was not a valid search string. Please go back and try again" );
136  }
137  $top = $res->fetchRow()[0];
138 
139  $this->searchTerms = [];
140  if ( $top === "" ) { # # e.g. if only stopwords are used XXX return something better
141  $query = "SELECT page_id, page_namespace, page_title, 0 AS score " .
142  "FROM page p, revision r, pagecontent c WHERE p.page_latest = r.rev_id " .
143  "AND r.rev_text_id = c.old_id AND 1=0";
144  } else {
145  $m = [];
146  if ( preg_match_all( "/'([^']+)'/", $top, $m, PREG_SET_ORDER ) ) {
147  foreach ( $m as $terms ) {
148  $this->searchTerms[$terms[1]] = $terms[1];
149  }
150  }
151 
152  $query = "SELECT page_id, page_namespace, page_title, " .
153  "ts_rank($fulltext, to_tsquery($searchstring), 5) AS score " .
154  "FROM page p, revision r, pagecontent c WHERE p.page_latest = r.rev_id " .
155  "AND r.rev_text_id = c.old_id AND $fulltext @@ to_tsquery($searchstring)";
156  }
157 
158  # # Namespaces - defaults to 0
159  if ( !is_null( $this->namespaces ) ) { // null -> search all
160  if ( count( $this->namespaces ) < 1 ) {
161  $query .= ' AND page_namespace = 0';
162  } else {
163  $namespaces = $this->db->makeList( $this->namespaces );
164  $query .= " AND page_namespace IN ($namespaces)";
165  }
166  }
167 
168  $query .= " ORDER BY score DESC, page_id DESC";
169 
170  $query .= $this->db->limitResult( '', $this->limit, $this->offset );
171 
172  wfDebug( "searchQuery returned: $query \n" );
173 
174  return $query;
175  }
176 
177  # # Most of the work of these two functions are done automatically via triggers
178 
179  function update( $pageid, $title, $text ) {
180  # # We don't want to index older revisions
181  $sql = "UPDATE pagecontent SET textvector = NULL WHERE textvector IS NOT NULL and old_id IN " .
182  "(SELECT DISTINCT rev_text_id FROM revision WHERE rev_page = " . intval( $pageid ) .
183  " ORDER BY rev_text_id DESC OFFSET 1)";
184  $this->db->query( $sql );
185  return true;
186  }
187 
188  function updateTitle( $id, $title ) {
189  return true;
190  }
191 
192 }
SearchPostgres\update
update( $pageid, $title, $text)
Create or update the search index record for the given page.
Definition: SearchPostgres.php:179
SearchPostgres\searchText
searchText( $term)
Perform a full text search query and return a result set.
Definition: SearchPostgres.php:48
SearchPostgres\parseQuery
parseQuery( $term)
Transform the user's search string into a better form for tsearch2 Returns an SQL fragment consisting...
Definition: SearchPostgres.php:64
SearchEngine\$namespaces
int[] null $namespaces
Definition: SearchEngine.php:39
e
in this case you re responsible for computing and outputting the entire conflict i e
Definition: hooks.txt:1421
used
you don t have to do a grep find to see where the $wgReverseTitle variable is used
Definition: hooks.txt:115
$res
$res
Definition: database.txt:21
SearchPostgres
Search engine hook base class for Postgres.
Definition: SearchPostgres.php:31
php
injection txt This is an overview of how MediaWiki makes use of dependency injection The design described here grew from the discussion of RFC T384 The term dependency this means that anything an object needs to operate should be injected from the the object itself should only know narrow no concrete implementation of the logic it relies on The requirement to inject everything typically results in an architecture that based on two main types of and essentially stateless service objects that use other service objects to operate on the value objects As of the beginning MediaWiki is only starting to use the DI approach Much of the code still relies on global state or direct resulting in a highly cyclical dependency which acts as the top level factory for services in MediaWiki which can be used to gain access to default instances of various services MediaWikiServices however also allows new services to be defined and default services to be redefined Services are defined or redefined by providing a callback the instantiator that will return a new instance of the service When it will create an instance of MediaWikiServices and populate it with the services defined in the files listed by thereby bootstrapping the DI framework Per $wgServiceWiringFiles lists includes ServiceWiring php
Definition: injection.txt:37
SearchPostgres\searchQuery
searchQuery( $term, $fulltext, $colname)
Construct the full SQL query to do the search.
Definition: SearchPostgres.php:126
SearchPostgres\searchTitle
searchTitle( $term)
Perform a full text search query via tsearch2 and return a result set.
Definition: SearchPostgres.php:40
SearchDatabase
Base search engine base class for database-backed searches.
Definition: SearchDatabase.php:31
$term
For QUnit the mediawiki tests qunit testrunner dependency will be added to any module whereas SearchGetNearMatch runs after $term
Definition: hooks.txt:2845
namespaces
to move a page</td >< td > &*You are moving the page across namespaces
Definition: All_system_messages.txt:2670
$title
namespace and then decline to actually register it file or subcat img or subcat $title
Definition: hooks.txt:964
wfDebug
wfDebug( $text, $dest='all', array $context=[])
Sends a line to the debug log if enabled or, optionally, to a comment in output.
Definition: GlobalFunctions.php:994
SqlSearchResultSet
This class is used for different SQL-based search engines shipped with MediaWiki.
Definition: SqlSearchResultSet.php:9
are
The ContentHandler facility adds support for arbitrary content types on wiki instead of relying on wikitext for everything It was introduced in MediaWiki Each kind of and so on Built in content types are
Definition: contenthandler.txt:9
as
This document is intended to provide useful advice for parties seeking to redistribute MediaWiki to end users It s targeted particularly at maintainers for Linux since it s been observed that distribution packages of MediaWiki often break We ve consistently had to recommend that users seeking support use official tarballs instead of their distribution s and this often solves whatever problem the user is having It would be nice if this could such as
Definition: distributors.txt:22
SearchPostgres\updateTitle
updateTitle( $id, $title)
Update a search index record's title only.
Definition: SearchPostgres.php:188
$query
null for the wiki Added should default to null in handler for backwards compatibility add a value to it if you want to add a cookie that have to vary cache options can modify $query
Definition: hooks.txt:1620