r69691 MediaWiki - Code Review archive

Repository:	MediaWiki
Revision:	< r69690‎ \| r69691 \| r69692 >
Date:	16:57, 21 July 2010
Author:	btongminh
Status:	ok
Tags:
Comment:	stylize.php on SearchEngine.php
Modified paths:	/trunk/phase3/includes/search/SearchEngine.php (modified) (history)

Diff [purge]

Index: trunk/phase3/includes/search/SearchEngine.php
—	—	@@ -101,21 +101,21 @@
102	102	private static function getNearMatchInternal( $searchterm ) {
103	103	global $wgContLang;
104	104
105		~~- $allSearchTerms = array($searchterm);~~
	105	+ $allSearchTerms = array( $searchterm );
106	106
107	107	if ( $wgContLang->hasVariants() ) {
108		~~- $allSearchTerms = array_merge($allSearchTerms,$wgContLang->convertLinkToAllVariants($searchterm));~~
	108	+ $allSearchTerms = array_merge( $allSearchTerms, $wgContLang->convertLinkToAllVariants( $searchterm ) );
109	109	}
110	110
111		~~- if( !wfRunHooks( 'SearchGetNearMatchBefore', array( $allSearchTerms, &$titleResult ) ) ) {~~
	111	+ if ( !wfRunHooks( 'SearchGetNearMatchBefore', array( $allSearchTerms, &$titleResult ) ) ) {
112	112	return $titleResult;
113	113	}
114	114
115		~~- foreach($allSearchTerms as $term) {~~
	115	+ foreach ( $allSearchTerms as $term ) {
116	116
117	117	# Exact match? No need to look further.
118	118	$title = Title::newFromText( $term );
119		~~- if (is_null($title))~~
	119	+ if ( is_null( $title ) )
120	120	return null;
121	121
122	122	if ( $title->getNamespace() == NS_SPECIAL \|\| $title->isExternal() \|\| $title->exists() ) {
—	—	@@ -124,7 +124,7 @@
125	125
126	126	# See if it still otherwise has content is some sane sense
127	127	$article = MediaWiki::articleFromTitle( $title );
128		~~- if( $article->hasViewableContent() ) {~~
	128	+ if ( $article->hasViewableContent() ) {
129	129	return $title;
130	130	}
131	131
—	—	@@ -150,14 +150,14 @@
151	151	}
152	152
153	153	# Now try Word-Caps-Breaking-At-Word-Breaks, for hyphenated names etc
154		~~- $title = Title::newFromText( $wgContLang->ucwordbreaks($term) );~~
	154	+ $title = Title::newFromText( $wgContLang->ucwordbreaks( $term ) );
155	155	if ( $title && $title->exists() ) {
156	156	return $title;
157	157	}
158	158
159	159	// Give hooks a chance at better match variants
160	160	$title = null;
161		~~- if( !wfRunHooks( 'SearchGetNearMatch', array( $term, &$title ) ) ) {~~
	161	+ if ( !wfRunHooks( 'SearchGetNearMatch', array( $term, &$title ) ) ) {
162	162	return $title;
163	163	}
164	164	}
—	—	@@ -165,7 +165,7 @@
166	166	$title = Title::newFromText( $searchterm );
167	167
168	168	# Entering an IP address goes to the contributions page
169		~~- if ( ( $title->getNamespace() == NS_USER && User::isIP($title->getText() ) )~~
	169	+ if ( ( $title->getNamespace() == NS_USER && User::isIP( $title->getText() ) )
170	170	\|\| User::isIP( trim( $searchterm ) ) ) {
171	171	return SpecialPage::getTitleFor( 'Contributions', $title->getDBkey() );
172	172	}
—	—	@@ -179,22 +179,22 @@
180	180	# Go to images that exist even if there's no local page.
181	181	# There may have been a funny upload, or it may be on a shared
182	182	# file repository such as Wikimedia Commons.
183		~~- if( $title->getNamespace() == NS_FILE ) {~~
	183	+ if ( $title->getNamespace() == NS_FILE ) {
184	184	$image = wfFindFile( $title );
185		~~- if( $image ) {~~
	185	+ if ( $image ) {
186	186	return $title;
187	187	}
188	188	}
189	189
190	190	# MediaWiki namespace? Page may be "implied" if not customized.
191	191	# Just return it, with caps forced as the message system likes it.
192		~~- if( $title->getNamespace() == NS_MEDIAWIKI ) {~~
	192	+ if ( $title->getNamespace() == NS_MEDIAWIKI ) {
193	193	return Title::makeTitle( NS_MEDIAWIKI, $wgContLang->ucfirst( $title->getText() ) );
194	194	}
195	195
196	196	# Quoted term? Try without the quotes...
197	197	$matches = array();
198		~~- if( preg_match( '/^"([^"]+)"$/', $searchterm, $matches ) ) {~~
	198	+ if ( preg_match( '/^"([^"]+)"$/', $searchterm, $matches ) ) {
199	199	return SearchEngine::getNearMatch( $matches[1] );
200	200	}
201	201
—	—	@@ -233,28 +233,28 @@
234	234	*
235	235	* @param $query String
236	236	*/
237		~~- function replacePrefixes( $query ){~~
	237	+ function replacePrefixes( $query ) {
238	238	global $wgContLang;
239	239
240	240	$parsed = $query;
241		~~- if( strpos($query,':') === false ) { // nothing to do~~
	241	+ if ( strpos( $query, ':' ) === false ) { // nothing to do
242	242	wfRunHooks( 'SearchEngineReplacePrefixesComplete', array( $this, $query, &$parsed ) );
243	243	return $parsed;
244	244	}
245	245
246		~~- $allkeyword = wfMsgForContent('searchall').":";~~
247		~~- if( strncmp($query, $allkeyword, strlen($allkeyword)) == 0 ){~~
	246	+ $allkeyword = wfMsgForContent( 'searchall' ) . ":";
	247	+ if ( strncmp( $query, $allkeyword, strlen( $allkeyword ) ) == 0 ) {
248	248	$this->namespaces = null;
249		~~- $parsed = substr($query,strlen($allkeyword));~~
250		~~- } else if( strpos($query,':') !== false ) {~~
251		~~- $prefix = substr($query,0,strpos($query,':'));~~
252		~~- $index = $wgContLang->getNsIndex($prefix);~~
253		~~- if($index !== false){~~
254		~~- $this->namespaces = array($index);~~
255		~~- $parsed = substr($query,strlen($prefix)+1);~~
	249	+ $parsed = substr( $query, strlen( $allkeyword ) );
	250	+ } else if ( strpos( $query, ':' ) !== false ) {
	251	+ $prefix = substr( $query, 0, strpos( $query, ':' ) );
	252	+ $index = $wgContLang->getNsIndex( $prefix );
	253	+ if ( $index !== false ) {
	254	+ $this->namespaces = array( $index );
	255	+ $parsed = substr( $query, strlen( $prefix ) + 1 );
256	256	}
257	257	}
258		~~- if(trim($parsed) == '')~~
	258	+ if ( trim( $parsed ) == '' )
259	259	$parsed = $query; // prefix was the whole query
260	260
261	261	wfRunHooks( 'SearchEngineReplacePrefixesComplete', array( $this, $query, &$parsed ) );
—	—	@@ -269,8 +269,8 @@
270	270	public static function searchableNamespaces() {
271	271	global $wgContLang;
272	272	$arr = array();
273		~~- foreach( $wgContLang->getNamespaces() as $ns => $name ) {~~
274		~~- if( $ns >= NS_MAIN ) {~~
	273	+ foreach ( $wgContLang->getNamespaces() as $ns => $name ) {
	274	+ if ( $ns >= NS_MAIN ) {
275	275	$arr[$ns] = $name;
276	276	}
277	277	}
—	—	@@ -291,18 +291,18 @@
292	292
293	293	// get search everything preference, that can be set to be read for logged-in users
294	294	$searcheverything = false;
295		~~- if( ( $wgSearchEverythingOnlyLoggedIn && $user->isLoggedIn() )~~
	295	+ if ( ( $wgSearchEverythingOnlyLoggedIn && $user->isLoggedIn() )
296	296	\|\| !$wgSearchEverythingOnlyLoggedIn )
297		~~- $searcheverything = $user->getOption('searcheverything');~~
	297	+ $searcheverything = $user->getOption( 'searcheverything' );
298	298
299	299	// searcheverything overrides other options
300		~~- if( $searcheverything )~~
301		~~- return array_keys(SearchEngine::searchableNamespaces());~~
	300	+ if ( $searcheverything )
	301	+ return array_keys( SearchEngine::searchableNamespaces() );
302	302
303	303	$arr = Preferences::loadOldSearchNs( $user );
304	304	$searchableNamespaces = SearchEngine::searchableNamespaces();
305	305
306		~~- $arr = array_intersect( $arr, array_keys($searchableNamespaces) ); // Filter~~
	306	+ $arr = array_intersect( $arr, array_keys( $searchableNamespaces ) ); // Filter
307	307
308	308	return $arr;
309	309	}
—	—	@@ -313,12 +313,12 @@
314	314	* @param $user User
315	315	* @return Array contextlines, contextchars
316	316	*/
317		~~- public static function userHighlightPrefs( &$user ){~~
318		~~- //$contextlines = $user->getOption( 'contextlines', 5 );~~
319		~~- //$contextchars = $user->getOption( 'contextchars', 50 );~~
	317	+ public static function userHighlightPrefs( &$user ) {
	318	+ // $contextlines = $user->getOption( 'contextlines', 5 );
	319	+ // $contextchars = $user->getOption( 'contextchars', 50 );
320	320	$contextlines = 2; // Hardcode this. Old defaults sucked. :)
321	321	$contextchars = 75; // same as above.... :P
322		~~- return array($contextlines, $contextchars);~~
	322	+ return array( $contextlines, $contextchars );
323	323	}
324	324
325	325	/**
—	—	@@ -326,10 +326,10 @@
327	327	*
328	328	* @return Array
329	329	*/
330		~~- public static function defaultNamespaces(){~~
	330	+ public static function defaultNamespaces() {
331	331	global $wgNamespacesToBeSearchedDefault;
332	332
333		~~- return array_keys($wgNamespacesToBeSearchedDefault, true);~~
	333	+ return array_keys( $wgNamespacesToBeSearchedDefault, true );
334	334	}
335	335
336	336	/**
—	—	@@ -338,12 +338,12 @@
339	339	*
340	340	* @param $namespaces Array
341	341	*/
342		~~- public static function namespacesAsText( $namespaces ){~~
	342	+ public static function namespacesAsText( $namespaces ) {
343	343	global $wgContLang;
344	344
345		~~- $formatted = array_map( array($wgContLang,'getFormattedNsText'), $namespaces );~~
346		~~- foreach( $formatted as $key => $ns ){~~
347		~~- if ( empty($ns) )~~
	345	+ $formatted = array_map( array( $wgContLang, 'getFormattedNsText' ), $namespaces );
	346	+ foreach ( $formatted as $key => $ns ) {
	347	+ if ( empty( $ns ) )
348	348	$formatted[$key] = wfMsg( 'blanknamespace' );
349	349	}
350	350	return $formatted;
—	—	@@ -379,13 +379,13 @@
380	380	public static function create() {
381	381	global $wgSearchType;
382	382	$dbr = wfGetDB( DB_SLAVE );
383		~~- if( $wgSearchType ) {~~
	383	+ if ( $wgSearchType ) {
384	384	$class = $wgSearchType;
385	385	} else {
386	386	$class = $dbr->getSearchEngine();
387	387	}
388	388	$search = new $class( $dbr );
389		~~- $search->setLimitOffset(0,0);~~
	389	+ $search->setLimitOffset( 0, 0 );
390	390	return $search;
391	391	}
392	392
—	—	@@ -421,12 +421,12 @@
422	422	*/
423	423	public static function getOpenSearchTemplate() {
424	424	global $wgOpenSearchTemplate, $wgServer, $wgScriptPath;
425		~~- if( $wgOpenSearchTemplate ) {~~
	425	+ if ( $wgOpenSearchTemplate ) {
426	426	return $wgOpenSearchTemplate;
427	427	} else {
428	428	$ns = implode( '\|', SearchEngine::defaultNamespaces() );
429		~~- if( !$ns ) $ns = "0";~~
430		~~- return $wgServer . $wgScriptPath . '/api.php?action=opensearch&search={searchTerms}&namespace='.$ns;~~
	429	+ if ( !$ns ) $ns = "0";
	430	+ return $wgServer . $wgScriptPath . '/api.php?action=opensearch&search={searchTerms}&namespace=' . $ns;
431	431	}
432	432	}
433	433
—	—	@@ -437,7 +437,7 @@
438	438	*/
439	439	public static function getMWSuggestTemplate() {
440	440	global $wgMWSuggestTemplate, $wgServer, $wgScriptPath;
441		~~- if($wgMWSuggestTemplate)~~
	441	+ if ( $wgMWSuggestTemplate )
442	442	return $wgMWSuggestTemplate;
443	443	else
444	444	return $wgServer . $wgScriptPath . '/api.php?action=opensearch&search={searchTerms}&namespace={namespaces}&suggest';
—	—	@@ -500,14 +500,14 @@
501	501	/**
502	502	* @return String: suggested query, null if none
503	503	*/
504		~~- function getSuggestionQuery(){~~
	504	+ function getSuggestionQuery() {
505	505	return null;
506	506	}
507	507
508	508	/**
509	509	* @return String: HTML highlighted suggested query, '' if none
510	510	*/
511		~~- function getSuggestionSnippet(){~~
	511	+ function getSuggestionSnippet() {
512	512	return '';
513	513	}
514	514
—	—	@@ -571,25 +571,25 @@
572	572	}
573	573
574	574	function numRows() {
575		~~- if ($this->mResultSet === false )~~
	575	+ if ( $this->mResultSet === false )
576	576	return false;
577	577
578	578	return $this->mResultSet->numRows();
579	579	}
580	580
581	581	function next() {
582		~~- if ($this->mResultSet === false )~~
	582	+ if ( $this->mResultSet === false )
583	583	return false;
584	584
585	585	$row = $this->mResultSet->fetchObject();
586		~~- if ($row === false)~~
	586	+ if ( $row === false )
587	587	return false;
588	588
589	589	return SearchResult::newFromRow( $row );
590	590	}
591	591
592	592	function free() {
593		~~- if ($this->mResultSet === false )~~
	593	+ if ( $this->mResultSet === false )
594	594	return false;
595	595
596	596	$this->mResultSet->free();
—	—	@@ -600,7 +600,7 @@
601	601	* @ingroup Search
602	602	*/
603	603	class SearchResultTooMany {
604		~~- ## Some search engines may bail out if too many matches are found~~
	604	+ # # Some search engines may bail out if too many matches are found
605	605	}
606	606
607	607
—	—	@@ -662,11 +662,11 @@
663	663	*/
664	664	protected function initFromTitle( $title ) {
665	665	$this->mTitle = $title;
666		~~- if( !is_null( $this->mTitle ) ){~~
	666	+ if ( !is_null( $this->mTitle ) ) {
667	667	$this->mRevision = Revision::newFromTitle( $this->mTitle );
668		~~- if( $this->mTitle->getNamespace() === NS_FILE )~~
	668	+ if ( $this->mTitle->getNamespace() === NS_FILE )
669	669	$this->mImage = wfFindFile( $this->mTitle );
670		~~- }~~
	670	+ }
671	671	}
672	672
673	673	/**
—	—	@@ -674,8 +674,8 @@
675	675	*
676	676	* @return Boolean
677	677	*/
678		~~- function isBrokenTitle(){~~
679		~~- if( is_null($this->mTitle) )~~
	678	+ function isBrokenTitle() {
	679	+ if ( is_null( $this->mTitle ) )
680	680	return true;
681	681	return false;
682	682	}
—	—	@@ -685,7 +685,7 @@
686	686	*
687	687	* @return Boolean
688	688	*/
689		~~- function isMissingRevision(){~~
	689	+ function isMissingRevision() {
690	690	return !$this->mRevision && !$this->mImage;
691	691	}
692	692
—	—	@@ -706,9 +706,9 @@
707	707	/**
708	708	* Lazy initialization of article text from DB
709	709	*/
710		~~- protected function initText(){~~
711		~~- if( !isset($this->mText) ){~~
712		~~- if($this->mRevision != null)~~
	710	+ protected function initText() {
	711	+ if ( !isset( $this->mText ) ) {
	712	+ if ( $this->mRevision != null )
713	713	$this->mText = $this->mRevision->getText();
714	714	else // TODO: can we fetch raw wikitext for commons images?
715	715	$this->mText = '';
—	—	@@ -720,12 +720,12 @@
721	721	* @param $terms Array: terms to highlight
722	722	* @return String: highlighted text snippet, null (and not '') if not supported
723	723	*/
724		~~- function getTextSnippet($terms){~~
	724	+ function getTextSnippet( $terms ) {
725	725	global $wgUser, $wgAdvancedSearchHighlighting;
726	726	$this->initText();
727		~~- list($contextlines,$contextchars) = SearchEngine::userHighlightPrefs($wgUser);~~
	727	+ list( $contextlines, $contextchars ) = SearchEngine::userHighlightPrefs( $wgUser );
728	728	$h = new SearchHighlighter();
729		~~- if( $wgAdvancedSearchHighlighting )~~
	729	+ if ( $wgAdvancedSearchHighlighting )
730	730	return $h->highlightText( $this->mText, $terms, $contextlines, $contextchars );
731	731	else
732	732	return $h->highlightSimple( $this->mText, $terms, $contextlines, $contextchars );
—	—	@@ -735,7 +735,7 @@
736	736	* @param $terms Array: terms to highlight
737	737	* @return String: highlighted title, '' if not supported
738	738	*/
739		~~- function getTitleSnippet($terms){~~
	739	+ function getTitleSnippet( $terms ) {
740	740	return '';
741	741	}
742	742
—	—	@@ -743,38 +743,38 @@
744	744	* @param $terms Array: terms to highlight
745	745	* @return String: highlighted redirect name (redirect to this page), '' if none or not supported
746	746	*/
747		~~- function getRedirectSnippet($terms){~~
	747	+ function getRedirectSnippet( $terms ) {
748	748	return '';
749	749	}
750	750
751	751	/**
752	752	* @return Title object for the redirect to this page, null if none or not supported
753	753	*/
754		~~- function getRedirectTitle(){~~
	754	+ function getRedirectTitle() {
755	755	return null;
756	756	}
757	757
758	758	/**
759	759	* @return string highlighted relevant section name, null if none or not supported
760	760	*/
761		~~- function getSectionSnippet(){~~
	761	+ function getSectionSnippet() {
762	762	return '';
763	763	}
764	764
765	765	/**
766	766	* @return Title object (pagename+fragment) for the section, null if none or not supported
767	767	*/
768		~~- function getSectionTitle(){~~
	768	+ function getSectionTitle() {
769	769	return null;
770	770	}
771	771
772	772	/**
773	773	* @return String: timestamp
774	774	*/
775		~~- function getTimestamp(){~~
776		~~- if( $this->mRevision )~~
	775	+ function getTimestamp() {
	776	+ if ( $this->mRevision )
777	777	return $this->mRevision->getTimestamp();
778		~~- else if( $this->mImage )~~
	778	+ else if ( $this->mImage )
779	779	return $this->mImage->getTimestamp();
780	780	return '';
781	781	}
—	—	@@ -782,7 +782,7 @@
783	783	/**
784	784	* @return Integer: number of words
785	785	*/
786		~~- function getWordCount(){~~
	786	+ function getWordCount() {
787	787	$this->initText();
788	788	return str_word_count( $this->mText );
789	789	}
—	—	@@ -790,7 +790,7 @@
791	791	/**
792	792	* @return Integer: size in bytes
793	793	*/
794		~~- function getByteSize(){~~
	794	+ function getByteSize() {
795	795	$this->initText();
796	796	return strlen( $this->mText );
797	797	}
—	—	@@ -798,14 +798,14 @@
799	799	/**
800	800	* @return Boolean if hit has related articles
801	801	*/
802		~~- function hasRelated(){~~
	802	+ function hasRelated() {
803	803	return false;
804	804	}
805	805
806	806	/**
807	807	* @return String: interwiki prefix of the title (return iw even if title is broken)
808	808	*/
809		~~- function getInterwikiPrefix(){~~
	809	+ function getInterwikiPrefix() {
810	810	return '';
811	811	}
812	812	}
—	—	@@ -843,7 +843,7 @@
844	844	class SearchHighlighter {
845	845	var $mCleanWikitext = true;
846	846
847		~~- function SearchHighlighter($cleanupWikitext = true){~~
	847	+ function SearchHighlighter( $cleanupWikitext = true ) {
848	848	$this->mCleanWikitext = $cleanupWikitext;
849	849	}
850	850
—	—	@@ -861,7 +861,7 @@
862	862	global $wgSearchHighlightBoundaries;
863	863	$fname = __METHOD__;
864	864
865		~~- if($text == '')~~
	865	+ if ( $text == '' )
866	866	return '';
867	867
868	868	// spli text into text + templates/links/tables
—	—	@@ -870,10 +870,10 @@
871	871	$endPatterns = array(
872	872	1 => '/(\{\{)\|(\}\})/', // template
873	873	2 => '/(\[\[)\|(\]\])/', // image
874		~~- 3 => "/(\n\\{\\\|)\|(\n\\\|\\})/"); // table~~
	874	+ 3 => "/(\n\\{\\\|)\|(\n\\\|\\})/" ); // table
875	875
876	876	// FIXME: this should prolly be a hook or something
877		~~- if(function_exists('wfCite')){~~
	877	+ if ( function_exists( 'wfCite' ) ) {
878	878	$spat .= '\|(<ref>)'; // references via cite extension
879	879	$endPatterns[4] = '/(<ref>)\|(<\/ref>)/';
880	880	}
—	—	@@ -882,18 +882,18 @@
883	883	$otherExt = array(); // other extracts
884	884	wfProfileIn( "$fname-split" );
885	885	$start = 0;
886		~~- $textLen = strlen($text);~~
	886	+ $textLen = strlen( $text );
887	887	$count = 0; // sequence number to maintain ordering
888		~~- while( $start < $textLen ){~~
	888	+ while ( $start < $textLen ) {
889	889	// find start of template/image/table
890		~~- if( preg_match( $spat, $text, $matches, PREG_OFFSET_CAPTURE, $start ) ){~~
	890	+ if ( preg_match( $spat, $text, $matches, PREG_OFFSET_CAPTURE, $start ) ) {
891	891	$epat = '';
892		~~- foreach($matches as $key => $val){~~
893		~~- if($key > 0 && $val[1] != -1){~~
894		~~- if($key == 2){~~
	892	+ foreach ( $matches as $key => $val ) {
	893	+ if ( $key > 0 && $val[1] != - 1 ) {
	894	+ if ( $key == 2 ) {
895	895	// see if this is an image link
896		~~- $ns = substr($val[0],2,-1);~~
897		~~- if( $wgContLang->getNsIndex($ns) != NS_FILE )~~
	896	+ $ns = substr( $val[0], 2, - 1 );
	897	+ if ( $wgContLang->getNsIndex( $ns ) != NS_FILE )
898	898	break;
899	899
900	900	}
—	—	@@ -903,42 +903,42 @@
904	904	break;
905	905	}
906	906	}
907		~~- if( $epat ){~~
	907	+ if ( $epat ) {
908	908	// find end (and detect any nested elements)
909	909	$level = 0;
910	910	$offset = $start + 1;
911	911	$found = false;
912		~~- while( preg_match( $epat, $text, $endMatches, PREG_OFFSET_CAPTURE, $offset ) ){~~
913		~~- if( array_key_exists(2,$endMatches) ){~~
	912	+ while ( preg_match( $epat, $text, $endMatches, PREG_OFFSET_CAPTURE, $offset ) ) {
	913	+ if ( array_key_exists( 2, $endMatches ) ) {
914	914	// found end
915		~~- if($level == 0){~~
916		~~- $len = strlen($endMatches[2][0]);~~
	915	+ if ( $level == 0 ) {
	916	+ $len = strlen( $endMatches[2][0] );
917	917	$off = $endMatches[2][1];
918	918	$this->splitAndAdd( $otherExt, $count,
919	919	substr( $text, $start, $off + $len - $start ) );
920	920	$start = $off + $len;
921	921	$found = true;
922	922	break;
923		~~- } else{~~
	923	+ } else {
924	924	// end of nested element
925	925	$level -= 1;
926	926	}
927		~~- } else{~~
	927	+ } else {
928	928	// nested
929	929	$level += 1;
930	930	}
931		~~- $offset = $endMatches[0][1] + strlen($endMatches[0][0]);~~
	931	+ $offset = $endMatches[0][1] + strlen( $endMatches[0][0] );
932	932	}
933		~~- if( ! $found ){~~
	933	+ if ( ! $found ) {
934	934	// couldn't find appropriate closing tag, skip
935		~~- $this->splitAndAdd( $textExt, $count, substr( $text, $start, strlen($matches[0][0]) ) );~~
936		~~- $start += strlen($matches[0][0]);~~
	935	+ $this->splitAndAdd( $textExt, $count, substr( $text, $start, strlen( $matches[0][0] ) ) );
	936	+ $start += strlen( $matches[0][0] );
937	937	}
938	938	continue;
939	939	}
940	940	}
941	941	// else: add as text extract
942		~~- $this->splitAndAdd( $textExt, $count, substr($text,$start) );~~
	942	+ $this->splitAndAdd( $textExt, $count, substr( $text, $start ) );
943	943	break;
944	944	}
945	945
—	—	@@ -947,29 +947,29 @@
948	948	wfProfileOut( "$fname-split" );
949	949
950	950	// prepare regexps
951		~~- foreach( $terms as $index => $term ) {~~
	951	+ foreach ( $terms as $index => $term ) {
952	952	// manually do upper/lowercase stuff for utf-8 since PHP won't do it
953		~~- if(preg_match('/[\x80-\xff]/', $term) ){~~
954		~~- $terms[$index] = preg_replace_callback('/./us',array($this,'caseCallback'),$terms[$index]);~~
	953	+ if ( preg_match( '/[\x80-\xff]/', $term ) ) {
	954	+ $terms[$index] = preg_replace_callback( '/./us', array( $this, 'caseCallback' ), $terms[$index] );
955	955	} else {
956	956	$terms[$index] = $term;
957	957	}
958	958	}
959	959	$anyterm = implode( '\|', $terms );
960		~~- $phrase = implode("$wgSearchHighlightBoundaries+", $terms );~~
	960	+ $phrase = implode( "$wgSearchHighlightBoundaries+", $terms );
961	961
962	962	// FIXME: a hack to scale contextchars, a correct solution
963	963	// would be to have contextchars actually be char and not byte
964	964	// length, and do proper utf-8 substrings and lengths everywhere,
965	965	// but PHP is making that very hard and unclean to implement :(
966		~~- $scale = strlen($anyterm) / mb_strlen($anyterm);~~
	966	+ $scale = strlen( $anyterm ) / mb_strlen( $anyterm );
967	967	$contextchars = intval( $contextchars * $scale );
968	968
969	969	$patPre = "(^\|$wgSearchHighlightBoundaries)";
970	970	$patPost = "($wgSearchHighlightBoundaries\|$)";
971	971
972		~~- $pat1 = "/(".$phrase.")/ui";~~
973		~~- $pat2 = "/$patPre(".$anyterm.")$patPost/ui";~~
	972	+ $pat1 = "/(" . $phrase . ")/ui";
	973	+ $pat2 = "/$patPre(" . $anyterm . ")$patPost/ui";
974	974
975	975	wfProfileIn( "$fname-extract" );
976	976
—	—	@@ -981,43 +981,43 @@
982	982	// show beginning only if it contains all words
983	983	$first = 0;
984	984	$firstText = '';
985		~~- foreach($textExt as $index => $line){~~
986		~~- if(strlen($line)>0 && $line[0] != ';' && $line[0] != ':'){~~
	985	+ foreach ( $textExt as $index => $line ) {
	986	+ if ( strlen( $line ) > 0 && $line[0] != ';' && $line[0] != ':' ) {
987	987	$firstText = $this->extract( $line, 0, $contextchars * $contextlines );
988	988	$first = $index;
989	989	break;
990	990	}
991	991	}
992		~~- if( $firstText ){~~
	992	+ if ( $firstText ) {
993	993	$succ = true;
994	994	// check if first text contains all terms
995		~~- foreach($terms as $term){~~
996		~~- if( ! preg_match("/$patPre".$term."$patPost/ui", $firstText) ){~~
	995	+ foreach ( $terms as $term ) {
	996	+ if ( ! preg_match( "/$patPre" . $term . "$patPost/ui", $firstText ) ) {
997	997	$succ = false;
998	998	break;
999	999	}
1000	1000	}
1001		~~- if( $succ ){~~
	1001	+ if ( $succ ) {
1002	1002	$snippets[$first] = $firstText;
1003	1003	$offsets[$first] = 0;
1004	1004	}
1005	1005	}
1006		~~- if( ! $snippets ) {~~
	1006	+ if ( ! $snippets ) {
1007	1007	// match whole query on text
1008		~~- $this->process($pat1, $textExt, $left, $contextchars, $snippets, $offsets);~~
	1008	+ $this->process( $pat1, $textExt, $left, $contextchars, $snippets, $offsets );
1009	1009	// match whole query on templates/tables/images
1010		~~- $this->process($pat1, $otherExt, $left, $contextchars, $snippets, $offsets);~~
	1010	+ $this->process( $pat1, $otherExt, $left, $contextchars, $snippets, $offsets );
1011	1011	// match any words on text
1012		~~- $this->process($pat2, $textExt, $left, $contextchars, $snippets, $offsets);~~
	1012	+ $this->process( $pat2, $textExt, $left, $contextchars, $snippets, $offsets );
1013	1013	// match any words on templates/tables/images
1014		~~- $this->process($pat2, $otherExt, $left, $contextchars, $snippets, $offsets);~~
	1014	+ $this->process( $pat2, $otherExt, $left, $contextchars, $snippets, $offsets );
1015	1015
1016		~~- ksort($snippets);~~
	1016	+ ksort( $snippets );
1017	1017	}
1018	1018
1019	1019	// add extra chars to each snippet to make snippets constant size
1020	1020	$extended = array();
1021		~~- if( count( $snippets ) == 0 ){~~
	1021	+ if ( count( $snippets ) == 0 ) {
1022	1022	// couldn't find the target words, just show beginning of article
1023	1023	if ( array_key_exists( $first, $all ) ) {
1024	1024	$targetchars = $contextchars * $contextlines;
—	—	@@ -1026,32 +1026,32 @@
1027	1027	}
1028	1028	} else {
1029	1029	// if begin of the article contains the whole phrase, show only that !!
1030		~~- if( array_key_exists($first,$snippets) && preg_match($pat1,$snippets[$first])~~
1031		~~- && $offsets[$first] < $contextchars * 2 ){~~
1032		~~- $snippets = array ($first => $snippets[$first]);~~
	1030	+ if ( array_key_exists( $first, $snippets ) && preg_match( $pat1, $snippets[$first] )
	1031	+ && $offsets[$first] < $contextchars * 2 ) {
	1032	+ $snippets = array ( $first => $snippets[$first] );
1033	1033	}
1034	1034
1035	1035	// calc by how much to extend existing snippets
1036		~~- $targetchars = intval( ($contextchars * $contextlines) / count ( $snippets ) );~~
	1036	+ $targetchars = intval( ( $contextchars * $contextlines ) / count ( $snippets ) );
1037	1037	}
1038	1038
1039		~~- foreach($snippets as $index => $line){~~
	1039	+ foreach ( $snippets as $index => $line ) {
1040	1040	$extended[$index] = $line;
1041		~~- $len = strlen($line);~~
1042		~~- if( $len < $targetchars - 20 ){~~
	1041	+ $len = strlen( $line );
	1042	+ if ( $len < $targetchars - 20 ) {
1043	1043	// complete this line
1044		~~- if($len < strlen( $all[$index] )){~~
1045		~~- $extended[$index] = $this->extract( $all[$index], $offsets[$index], $offsets[$index]+$targetchars, $offsets[$index]);~~
	1044	+ if ( $len < strlen( $all[$index] ) ) {
	1045	+ $extended[$index] = $this->extract( $all[$index], $offsets[$index], $offsets[$index] + $targetchars, $offsets[$index] );
1046	1046	$len = strlen( $extended[$index] );
1047	1047	}
1048	1048
1049	1049	// add more lines
1050	1050	$add = $index + 1;
1051		~~- while( $len < $targetchars - 20~~
1052		~~- && array_key_exists($add,$all)~~
1053		~~- && !array_key_exists($add,$snippets) ){~~
	1051	+ while ( $len < $targetchars - 20
	1052	+ && array_key_exists( $add, $all )
	1053	+ && !array_key_exists( $add, $snippets ) ) {
1054	1054	$offsets[$add] = 0;
1055		~~- $tt = "\n".$this->extract( $all[$add], 0, $targetchars - $len, $offsets[$add] );~~
	1055	+ $tt = "\n" . $this->extract( $all[$add], 0, $targetchars - $len, $offsets[$add] );
1056	1056	$extended[$add] = $tt;
1057	1057	$len += strlen( $tt );
1058	1058	$add++;
—	—	@@ -1059,27 +1059,27 @@
1060	1060	}
1061	1061	}
1062	1062
1063		~~- //$snippets = array_map('htmlspecialchars', $extended);~~
	1063	+ // $snippets = array_map('htmlspecialchars', $extended);
1064	1064	$snippets = $extended;
1065		~~- $last = -1;~~
	1065	+ $last = - 1;
1066	1066	$extract = '';
1067		~~- foreach($snippets as $index => $line){~~
1068		~~- if($last == -1)~~
	1067	+ foreach ( $snippets as $index => $line ) {
	1068	+ if ( $last == - 1 )
1069	1069	$extract .= $line; // first line
1070		~~- elseif($last+1 == $index && $offsets[$last]+strlen($snippets[$last]) >= strlen($all[$last]))~~
1071		~~- $extract .= " ".$line; // continous lines~~
	1070	+ elseif ( $last + 1 == $index && $offsets[$last] + strlen( $snippets[$last] ) >= strlen( $all[$last] ) )
	1071	+ $extract .= " " . $line; // continous lines
1072	1072	else
1073	1073	$extract .= '<b> ... </b>' . $line;
1074	1074
1075	1075	$last = $index;
1076	1076	}
1077		~~- if( $extract )~~
	1077	+ if ( $extract )
1078	1078	$extract .= '<b> ... </b>';
1079	1079
1080	1080	$processed = array();
1081		~~- foreach($terms as $term){~~
1082		~~- if( ! isset($processed[$term]) ){~~
1083		~~- $pat3 = "/$patPre(".$term.")$patPost/ui"; // highlight word~~
	1081	+ foreach ( $terms as $term ) {
	1082	+ if ( ! isset( $processed[$term] ) ) {
	1083	+ $pat3 = "/$patPre(" . $term . ")$patPost/ui"; // highlight word
1084	1084	$extract = preg_replace( $pat3,
1085	1085	"\\1<span class='searchmatch'>\\2</span>\\3", $extract );
1086	1086	$processed[$term] = true;
—	—	@@ -1098,11 +1098,11 @@
1099	1099	* @param $count Integer
1100	1100	* @param $text String
1101	1101	*/
1102		~~- function splitAndAdd(&$extracts, &$count, $text){~~
1103		~~- $split = explode( "\n", $this->mCleanWikitext? $this->removeWiki($text) : $text );~~
1104		~~- foreach($split as $line){~~
1105		~~- $tt = trim($line);~~
1106		~~- if( $tt )~~
	1102	+ function splitAndAdd( &$extracts, &$count, $text ) {
	1103	+ $split = explode( "\n", $this->mCleanWikitext ? $this->removeWiki( $text ) : $text );
	1104	+ foreach ( $split as $line ) {
	1105	+ $tt = trim( $line );
	1106	+ if ( $tt )
1107	1107	$extracts[$count++] = $tt;
1108	1108	}
1109	1109	}
—	—	@@ -1112,10 +1112,10 @@
1113	1113	*
1114	1114	* @param $matches Array
1115	1115	*/
1116		~~- function caseCallback($matches){~~
	1116	+ function caseCallback( $matches ) {
1117	1117	global $wgContLang;
1118		~~- if( strlen($matches[0]) > 1 ){~~
1119		~~- return '['.$wgContLang->lc($matches[0]).$wgContLang->uc($matches[0]).']';~~
	1118	+ if ( strlen( $matches[0] ) > 1 ) {
	1119	+ return '[' . $wgContLang->lc( $matches[0] ) . $wgContLang->uc( $matches[0] ) . ']';
1120	1120	} else
1121	1121	return $matches[0];
1122	1122	}
—	—	@@ -1130,23 +1130,23 @@
1131	1131	* @param $posEnd Integer: (out) actual end position
1132	1132	* @return String
1133	1133	*/
1134		~~- function extract($text, $start, $end, &$posStart = null, &$posEnd = null ){~~
	1134	+ function extract( $text, $start, $end, &$posStart = null, &$posEnd = null ) {
1135	1135	global $wgContLang;
1136	1136
1137		~~- if( $start != 0)~~
	1137	+ if ( $start != 0 )
1138	1138	$start = $this->position( $text, $start, 1 );
1139		~~- if( $end >= strlen($text) )~~
1140		~~- $end = strlen($text);~~
	1139	+ if ( $end >= strlen( $text ) )
	1140	+ $end = strlen( $text );
1141	1141	else
1142	1142	$end = $this->position( $text, $end );
1143	1143
1144		~~- if(!is_null($posStart))~~
	1144	+ if ( !is_null( $posStart ) )
1145	1145	$posStart = $start;
1146		~~- if(!is_null($posEnd))~~
	1146	+ if ( !is_null( $posEnd ) )
1147	1147	$posEnd = $end;
1148	1148
1149		~~- if($end > $start)~~
1150		~~- return substr($text, $start, $end-$start);~~
	1149	+ if ( $end > $start )
	1150	+ return substr( $text, $start, $end - $start );
1151	1151	else
1152	1152	return '';
1153	1153	}
—	—	@@ -1159,21 +1159,21 @@
1160	1160	* @param $offset Integer: offset to found index
1161	1161	* @return Integer: nearest nonletter index, or beginning of utf8 char if none
1162	1162	*/
1163		~~- function position($text, $point, $offset=0 ){~~
	1163	+ function position( $text, $point, $offset = 0 ) {
1164	1164	$tolerance = 10;
1165	1165	$s = max( 0, $point - $tolerance );
1166		~~- $l = min( strlen($text), $point + $tolerance ) - $s;~~
	1166	+ $l = min( strlen( $text ), $point + $tolerance ) - $s;
1167	1167	$m = array();
1168		~~- if( preg_match('/[ ,.!?~!@#$%^&*+=\-\\\\|\[\]"\'<>]/', substr($text,$s,$l), $m, PREG_OFFSET_CAPTURE ) ){~~
	1168	+ if ( preg_match( '/[ ,.!?~!@#$%^&*+=\-\\\\|\[\]"\'<>]/', substr( $text, $s, $l ), $m, PREG_OFFSET_CAPTURE ) ) {
1169	1169	return $m[0][1] + $s + $offset;
1170		~~- } else{~~
	1170	+ } else {
1171	1171	// check if point is on a valid first UTF8 char
1172	1172	$char = ord( $text[$point] );
1173		~~- while( $char >= 0x80 && $char < 0xc0 ) {~~
	1173	+ while ( $char >= 0x80 && $char < 0xc0 ) {
1174	1174	// skip trailing bytes
1175	1175	$point++;
1176		~~- if($point >= strlen($text))~~
1177		~~- return strlen($text);~~
	1176	+ if ( $point >= strlen( $text ) )
	1177	+ return strlen( $text );
1178	1178	$char = ord( $text[$point] );
1179	1179	}
1180	1180	return $point;
—	—	@@ -1192,11 +1192,11 @@
1193	1193	* @param $offsets Array: map of starting points of snippets
1194	1194	* @protected
1195	1195	*/
1196		~~- function process( $pattern, $extracts, &$linesleft, &$contextchars, &$out, &$offsets ){~~
1197		~~- if($linesleft == 0)~~
	1196	+ function process( $pattern, $extracts, &$linesleft, &$contextchars, &$out, &$offsets ) {
	1197	+ if ( $linesleft == 0 )
1198	1198	return; // nothing to do
1199		~~- foreach($extracts as $index => $line){~~
1200		~~- if( array_key_exists($index,$out) )~~
	1199	+ foreach ( $extracts as $index => $line ) {
	1200	+ if ( array_key_exists( $index, $out ) )
1201	1201	continue; // this line already highlighted
1202	1202
1203	1203	$m = array();
—	—	@@ -1204,22 +1204,22 @@
1205	1205	continue;
1206	1206
1207	1207	$offset = $m[0][1];
1208		~~- $len = strlen($m[0][0]);~~
1209		~~- if($offset + $len < $contextchars)~~
	1208	+ $len = strlen( $m[0][0] );
	1209	+ if ( $offset + $len < $contextchars )
1210	1210	$begin = 0;
1211		~~- elseif( $len > $contextchars)~~
	1211	+ elseif ( $len > $contextchars )
1212	1212	$begin = $offset;
1213	1213	else
1214		~~- $begin = $offset + intval( ($len - $contextchars) / 2 );~~
	1214	+ $begin = $offset + intval( ( $len - $contextchars ) / 2 );
1215	1215
1216	1216	$end = $begin + $contextchars;
1217	1217
1218	1218	$posBegin = $begin;
1219	1219	// basic snippet from this line
1220		~~- $out[$index] = $this->extract($line,$begin,$end,$posBegin);~~
	1220	+ $out[$index] = $this->extract( $line, $begin, $end, $posBegin );
1221	1221	$offsets[$index] = $posBegin;
1222	1222	$linesleft--;
1223		~~- if($linesleft == 0)~~
	1223	+ if ( $linesleft == 0 )
1224	1224	return;
1225	1225	}
1226	1226	}
—	—	@@ -1228,25 +1228,25 @@
1229	1229	* Basic wikitext removal
1230	1230	* @protected
1231	1231	*/
1232		~~- function removeWiki($text) {~~
	1232	+ function removeWiki( $text ) {
1233	1233	$fname = __METHOD__;
1234	1234	wfProfileIn( $fname );
1235	1235
1236		~~- //$text = preg_replace("/'{2,5}/", "", $text);~~
1237		~~- //$text = preg_replace("/\[[a-z]+:\/\/[^ ]+ ([^]]+)\]/", "\\2", $text);~~
1238		~~- //$text = preg_replace("/\[\[([^]\|]+)\]\]/", "\\1", $text);~~
1239		~~- //$text = preg_replace("/\[\[([^]]+\\|)?([^\|]]+)\]\]/", "\\2", $text);~~
1240		~~- //$text = preg_replace("/\\{\\\|(.*?)\\\|\\}/", "", $text);~~
1241		~~- //$text = preg_replace("/\\[\\[[A-Za-z_-]+:([^\|]+?)\\]\\]/", "", $text);~~
1242		~~- $text = preg_replace("/\\{\\{([^\|]+?)\\}\\}/", "", $text);~~
1243		~~- $text = preg_replace("/\\{\\{([^\|]+\\\|)(.*?)\\}\\}/", "\\2", $text);~~
1244		~~- $text = preg_replace("/\\[\\[([^\|]+?)\\]\\]/", "\\1", $text);~~
1245		~~- $text = preg_replace_callback("/\\[\\[([^\|]+\\\|)(.*?)\\]\\]/", array($this,'linkReplace'), $text);~~
1246		~~- //$text = preg_replace("/\\[\\[([^\|]+\\\|)(.*?)\\]\\]/", "\\2", $text);~~
1247		~~- $text = preg_replace("/<\/?[^>]+>/", "", $text);~~
1248		~~- $text = preg_replace("/'''''/", "", $text);~~
1249		~~- $text = preg_replace("/('''\|<\/?[iIuUbB]>)/", "", $text);~~
1250		~~- $text = preg_replace("/''/", "", $text);~~
	1236	+ // $text = preg_replace("/'{2,5}/", "", $text);
	1237	+ // $text = preg_replace("/\[[a-z]+:\/\/[^ ]+ ([^]]+)\]/", "\\2", $text);
	1238	+ // $text = preg_replace("/\[\[([^]\|]+)\]\]/", "\\1", $text);
	1239	+ // $text = preg_replace("/\[\[([^]]+\\|)?([^\|]]+)\]\]/", "\\2", $text);
	1240	+ // $text = preg_replace("/\\{\\\|(.*?)\\\|\\}/", "", $text);
	1241	+ // $text = preg_replace("/\\[\\[[A-Za-z_-]+:([^\|]+?)\\]\\]/", "", $text);
	1242	+ $text = preg_replace( "/\\{\\{([^\|]+?)\\}\\}/", "", $text );
	1243	+ $text = preg_replace( "/\\{\\{([^\|]+\\\|)(.*?)\\}\\}/", "\\2", $text );
	1244	+ $text = preg_replace( "/\\[\\[([^\|]+?)\\]\\]/", "\\1", $text );
	1245	+ $text = preg_replace_callback( "/\\[\\[([^\|]+\\\|)(.*?)\\]\\]/", array( $this, 'linkReplace' ), $text );
	1246	+ // $text = preg_replace("/\\[\\[([^\|]+\\\|)(.*?)\\]\\]/", "\\2", $text);
	1247	+ $text = preg_replace( "/<\/?[^>]+>/", "", $text );
	1248	+ $text = preg_replace( "/'''''/", "", $text );
	1249	+ $text = preg_replace( "/('''\|<\/?[iIuUbB]>)/", "", $text );
	1250	+ $text = preg_replace( "/''/", "", $text );
1251	1251
1252	1252	wfProfileOut( $fname );
1253	1253	return $text;
—	—	@@ -1258,14 +1258,14 @@
1259	1259	*
1260	1260	* @param $matches Array
1261	1261	*/
1262		~~- function linkReplace($matches){~~
	1262	+ function linkReplace( $matches ) {
1263	1263	$colon = strpos( $matches[1], ':' );
1264		~~- if( $colon === false )~~
	1264	+ if ( $colon === false )
1265	1265	return $matches[2]; // replace with caption
1266	1266	global $wgContLang;
1267	1267	$ns = substr( $matches[1], 0, $colon );
1268		~~- $index = $wgContLang->getNsIndex($ns);~~
1269		~~- if( $index !== false && ($index == NS_FILE \|\| $index == NS_CATEGORY) )~~
	1268	+ $index = $wgContLang->getNsIndex( $ns );
	1269	+ if ( $index !== false && ( $index == NS_FILE \|\| $index == NS_CATEGORY ) )
1270	1270	return $matches[0]; // return the whole thing
1271	1271	else
1272	1272	return $matches[2];
—	—	@@ -1306,7 +1306,7 @@
1307	1307	continue;
1308	1308	}
1309	1309	--$contextlines;
1310		~~- $pre = $wgContLang->truncate( $m[1], -$contextchars );~~
	1310	+ $pre = $wgContLang->truncate( $m[1], - $contextchars );
1311	1311
1312	1312	if ( count( $m ) < 3 ) {
1313	1313	$post = '';

Status & tagging log

12:16, 6 November 2010 Hashar (talk | contribs) changed the status of r69691 [removed: new added: ok]