diocloid/LinkTitles
1
0
Fork 0
mirror of https://github.com/diocloid/LinkTitles.git synced 2025-07-13 17:59:29 +02:00
Code Issues Packages Projects Releases Wiki Activity

Cache page titles.

Browse Source 
- Improvement: Increase performance of special page and maintenance script.
This commit is contained in:
Daniel Kraus
2016-12-29 07:53:58 +01:00
parent 3990252a57
commit 9b25d404ce
1 changed files with 77 additions and 58 deletions
Download Patch File Download Diff File Expand all files Collapse all files

135
includes/LinkTitles_Extension.php
View File

@ -30,6 +30,15 @@ function dump($var) {
/// Central class of the extension. Sets up parser hooks. /// Central class of the extension. Sets up parser hooks.
/// This class contains only static functions; do not instantiate. /// This class contains only static functions; do not instantiate.
class Extension { class Extension {
/// Caching variable for page titles that are fetched from the DB.
private static $pageTitles;
/// Caching variable for the current namespace.
/// This is needed because the sort order of the page titles that
/// are cached in self::$pageTitles depends on the namespace of
/// the page currently being processed.
private static $currentNamespace;
/// A Title object for the page that is being parsed. /// A Title object for the page that is being parsed.
private static $currentTitle; private static $currentTitle;
@ -113,76 +122,23 @@ class Extension {
private static function parseContent( $title, &$text ) { private static function parseContent( $title, &$text ) {
// Configuration variables need to be defined here as globals. // Configuration variables need to be defined here as globals.
global $wgLinkTitlesPreferShortTitles;
global $wgLinkTitlesMinimumTitleLength;
global $wgLinkTitlesBlackList;
global $wgLinkTitlesFirstOnly; global $wgLinkTitlesFirstOnly;
global $wgLinkTitlesSmartMode; global $wgLinkTitlesSmartMode;
global $wgCapitalLinks; global $wgCapitalLinks;
global $wgLinkTitlesNamespaces;
( $wgLinkTitlesPreferShortTitles ) ? $sort_order = 'ASC' : $sort_order = 'DESC';
( $wgLinkTitlesFirstOnly ) ? $limit = 1 : $limit = -1; ( $wgLinkTitlesFirstOnly ) ? $limit = 1 : $limit = -1;
$limitReached = false; $limitReached = false;
self::$currentTitle = $title; self::$currentTitle = $title;
$currentNamespace = $title->getNamespace();
$newText = $text; $newText = $text;
// Build a blacklist of pages that are not supposed to be link if ( !isset( self::$pageTitles ) || ( $currentNamespace != self::$currentNamespace ) ) {
// targets. This includes the current page. self::$currentNamespace = $currentNamespace;
$blackList = str_replace( ' ', '_', self::$pageTitles = self::fetchPageTitles( $currentNamespace );
'("' . implode( '","',$wgLinkTitlesBlackList ) . '","' .
addslashes( self::$currentTitle->getDbKey() ) . '")' );
$currentNamespace[] = $title->getNamespace();
// Build our weight list. Make sure current namespace is first element
$namespaces = array_diff($wgLinkTitlesNamespaces, $currentNamespace);
array_unshift($namespaces, $currentNamespace[0] );
// No need for sanitiy check. we are sure that we have at least one element in the array
$weightSelect = "CASE page_namespace ";
$currentWeight = 0;
foreach ($namespaces as &$namspacevalue) {
$currentWeight = $currentWeight + 100;
$weightSelect = $weightSelect . " WHEN " . $namspacevalue . " THEN " . $currentWeight . PHP_EOL;
}
$weightSelect = $weightSelect . " END ";
$namespacesClause = '(' . implode( ', ', $namespaces ) . ')';
// Build an SQL query and fetch all page titles ordered by length from
// shortest to longest. Only titles from 'normal' pages (namespace uid
// = 0) are returned. Since the db may be sqlite, we need a try..catch
// structure because sqlite does not support the CHAR_LENGTH function.
$dbr = wfGetDB( DB_SLAVE );
try {
$res = $dbr->select(
'page',
array( 'page_title', 'page_namespace' , "weight" => $weightSelect),
array(
'page_namespace IN ' . $namespacesClause,
'CHAR_LENGTH(page_title) >= ' . $wgLinkTitlesMinimumTitleLength,
'page_title NOT IN ' . $blackList,
),
__METHOD__,
array( 'ORDER BY' => 'weight ASC, CHAR_LENGTH(page_title) ' . $sort_order )
);
} catch (Exception $e) {
$res = $dbr->select(
'page',
array( 'page_title', 'page_namespace' , "weight" => $weightSelect ),
array(
'page_namespace IN ' . $namespacesClause,
'LENGTH(page_title) >= ' . $wgLinkTitlesMinimumTitleLength,
'page_title NOT IN ' . $blackList,
),
__METHOD__,
array( 'ORDER BY' => 'weight ASC, LENGTH(page_title) ' . $sort_order )
);
} }
// Iterate through the page titles // Iterate through the page titles
foreach( $res as $row ) { foreach( self::$pageTitles as $row ) {
self::newTarget( $row->page_namespace, $row->page_title ); self::newTarget( $row->page_namespace, $row->page_title );
// split the page content by [[...]] groups // split the page content by [[...]] groups
@ -286,6 +242,69 @@ class Extension {
return true; return true;
} }
// Fetches the page titles from the database.
// @param $currentNamespace String holding the namespace of the page currently being processed.
private static function fetchPageTitles( $currentNamespace ) {
global $wgLinkTitlesPreferShortTitles;
global $wgLinkTitlesMinimumTitleLength;
global $wgLinkTitlesBlackList;
global $wgLinkTitlesNamespaces;
( $wgLinkTitlesPreferShortTitles ) ? $sort_order = 'ASC' : $sort_order = 'DESC';
// Build a blacklist of pages that are not supposed to be link
// targets. This includes the current page.
$blackList = str_replace( ' ', '_',
'("' . implode( '","',$wgLinkTitlesBlackList ) . '","' .
addslashes( self::$currentTitle->getDbKey() ) . '")' );
// Build our weight list. Make sure current namespace is first element
$namespaces = array_diff( $wgLinkTitlesNamespaces, [ $currentNamespace ] );
array_unshift( $namespaces, $currentNamespace );
// No need for sanitiy check. we are sure that we have at least one element in the array
$weightSelect = "CASE page_namespace ";
$currentWeight = 0;
foreach ($namespaces as &$namspacevalue) {
$currentWeight = $currentWeight + 100;
$weightSelect = $weightSelect . " WHEN " . $namspacevalue . " THEN " . $currentWeight . PHP_EOL;
}
$weightSelect = $weightSelect . " END ";
$namespacesClause = '(' . implode( ', ', $namespaces ) . ')';
// Build an SQL query and fetch all page titles ordered by length from
// shortest to longest. Only titles from 'normal' pages (namespace uid
// = 0) are returned. Since the db may be sqlite, we need a try..catch
// structure because sqlite does not support the CHAR_LENGTH function.
$dbr = wfGetDB( DB_SLAVE );
try {
$res = $dbr->select(
'page',
array( 'page_title', 'page_namespace' , "weight" => $weightSelect),
array(
'page_namespace IN ' . $namespacesClause,
'CHAR_LENGTH(page_title) >= ' . $wgLinkTitlesMinimumTitleLength,
'page_title NOT IN ' . $blackList,
),
__METHOD__,
array( 'ORDER BY' => 'weight ASC, CHAR_LENGTH(page_title) ' . $sort_order )
);
} catch (Exception $e) {
$res = $dbr->select(
'page',
array( 'page_title', 'page_namespace' , "weight" => $weightSelect ),
array(
'page_namespace IN ' . $namespacesClause,
'LENGTH(page_title) >= ' . $wgLinkTitlesMinimumTitleLength,
'page_title NOT IN ' . $blackList,
),
__METHOD__,
array( 'ORDER BY' => 'weight ASC, LENGTH(page_title) ' . $sort_order )
);
}
return $res;
}
// Build an anonymous callback function to be used in simple mode. // Build an anonymous callback function to be used in simple mode.
private static function simpleModeCallback( array $matches ) { private static function simpleModeCallback( array $matches ) {
if ( self::checkTargetPage() ) { if ( self::checkTargetPage() ) {