wiki.techinc.nl/includes/search/SearchUpdate.php

145 lines
4.5 KiB
PHP
Raw Normal View History

<?php
/**
* Search index updater
*
* See deferred.txt
*
* This program is free software; you can redistribute it and/or modify
* it under the terms of the GNU General Public License as published by
* the Free Software Foundation; either version 2 of the License, or
* (at your option) any later version.
*
* This program is distributed in the hope that it will be useful,
* but WITHOUT ANY WARRANTY; without even the implied warranty of
* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
* GNU General Public License for more details.
*
* You should have received a copy of the GNU General Public License along
* with this program; if not, write to the Free Software Foundation, Inc.,
* 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA.
* http://www.gnu.org/copyleft/gpl.html
*
* @file
* @ingroup Search
*/
/**
* Database independant search index updater
*
* @ingroup Search
*/
class SearchUpdate implements DeferrableUpdate {
2003-04-14 23:10:40 +00:00
private $mId = 0, $mNamespace, $mTitle, $mText;
private $mTitleWords;
2003-04-14 23:10:40 +00:00
2010-08-30 16:52:51 +00:00
function __construct( $id, $title, $text = false ) {
if ( is_string( $title ) ) {
$nt = Title::newFromText( $title );
} else {
$nt = $title;
}
2003-12-03 00:42:31 +00:00
if( $nt ) {
$this->mId = $id;
$this->mText = $text;
$this->mNamespace = $nt->getNamespace();
$this->mTitle = $nt->getText(); # Discard namespace
2003-04-14 23:10:40 +00:00
2003-12-03 00:42:31 +00:00
$this->mTitleWords = $this->mTextWords = array();
} else {
wfDebug( "SearchUpdate object created with invalid title '$title'\n" );
}
2003-04-14 23:10:40 +00:00
}
function doUpdate() {
2005-12-04 18:27:59 +00:00
global $wgContLang, $wgDisableSearchUpdate;
2003-12-03 00:42:31 +00:00
if( $wgDisableSearchUpdate || !$this->mId ) {
return;
}
2010-12-07 10:51:57 +00:00
wfProfileIn( __METHOD__ );
$search = SearchEngine::create();
$lc = SearchEngine::legalSearchChars() . '&#;';
2006-01-07 13:31:29 +00:00
if( $this->mText === false ) {
$search->updateTitle($this->mId,
$search->normalizeText( Title::indexTitle( $this->mNamespace, $this->mTitle ) ) );
2010-12-07 10:51:57 +00:00
wfProfileOut( __METHOD__ );
2003-04-14 23:10:40 +00:00
return;
}
2003-04-14 23:10:40 +00:00
# Language-specific strip/conversion
$text = $wgContLang->normalizeForSearch( $this->mText );
2003-04-14 23:10:40 +00:00
2010-12-07 10:51:57 +00:00
wfProfileIn( __METHOD__ . '-regexps' );
$text = preg_replace( "/<\\/?\\s*[A-Za-z][^>]*?>/",
' ', $wgContLang->lc( " " . $text . " " ) ); # Strip HTML markup
$text = preg_replace( "/(^|\\n)==\\s*([^\\n]+)\\s*==(\\s)/sD",
"\\1\\2 \\2 \\2\\3", $text ); # Emphasize headings
2003-04-14 23:10:40 +00:00
# Strip external URLs
2011-11-30 00:37:35 +00:00
$uc = "A-Za-z0-9_\\/:.,~%\\-+&;#?!=()@\\x80-\\xFF";
2003-04-14 23:10:40 +00:00
$protos = "http|https|ftp|mailto|news|gopher";
$pat = "/(^|[^\\[])({$protos}):[{$uc}]+([^{$uc}]|$)/";
$text = preg_replace( $pat, "\\1 \\3", $text );
$p1 = "/([^\\[])\\[({$protos}):[{$uc}]+]/";
$p2 = "/([^\\[])\\[({$protos}):[{$uc}]+\\s+([^\\]]+)]/";
$text = preg_replace( $p1, "\\1 ", $text );
$text = preg_replace( $p2, "\\1 \\3 ", $text );
# Internal image links
$pat2 = "/\\[\\[image:([{$uc}]+)\\.(gif|png|jpg|jpeg)([^{$uc}])/i";
$text = preg_replace( $pat2, " \\1 \\3", $text );
$text = preg_replace( "/([^{$lc}])([{$lc}]+)]]([a-z]+)/",
"\\1\\2 \\2\\3", $text ); # Handle [[game]]s
# Strip all remaining non-search characters
$text = preg_replace( "/[^{$lc}]+/", " ", $text );
# Handle 's, s'
#
# $text = preg_replace( "/([{$lc}]+)'s /", "\\1 \\1's ", $text );
# $text = preg_replace( "/([{$lc}]+)s' /", "\\1s ", $text );
#
# These tail-anchored regexps are insanely slow. The worst case comes
# when Japanese or Chinese text (ie, no word spacing) is written on
# a wiki configured for Western UTF-8 mode. The Unicode characters are
# expanded to hex codes and the "words" are very long paragraph-length
# monstrosities. On a large page the above regexps may take over 20
# seconds *each* on a 1GHz-level processor.
#
# Following are reversed versions which are consistently fast
# (about 3 milliseconds on 1GHz-level processor).
#
$text = strrev( preg_replace( "/ s'([{$lc}]+)/", " s'\\1 \\1", strrev( $text ) ) );
$text = strrev( preg_replace( "/ 's([{$lc}]+)/", " s\\1", strrev( $text ) ) );
2003-04-14 23:10:40 +00:00
# Strip wiki '' and '''
$text = preg_replace( "/''[']*/", " ", $text );
2010-12-07 10:51:57 +00:00
wfProfileOut( __METHOD__ . '-regexps' );
wfRunHooks( 'SearchUpdate', array( $this->mId, $this->mNamespace, $this->mTitle, &$text ) );
# Perform the actual update
$search->update($this->mId, $search->normalizeText( Title::indexTitle( $this->mNamespace, $this->mTitle ) ),
$search->normalizeText( $text ) );
2010-12-07 10:51:57 +00:00
wfProfileOut( __METHOD__ );
2003-04-14 23:10:40 +00:00
}
}
2005-01-27 19:51:47 +00:00
/**
* Placeholder class
*
* @ingroup Search
2005-01-27 19:51:47 +00:00
*/
class SearchUpdateMyISAM extends SearchUpdate {
# Inherits everything
}