r103056 MediaWiki - Code Review archive

Repository:	MediaWiki
Revision:	< r103055‎ \| r103056 \| r103057 >
Date:	22:40, 14 November 2011
Author:	reedy
Status:	ok (Comments)
Tags:
Comment:	MFT r102951, r102954
Modified paths:	/branches/wmf/1.18wmf1 (modified) (history) /branches/wmf/1.18wmf1/includes (modified) (history) /branches/wmf/1.18wmf1/includes/AutoLoader.php (modified) (history) /branches/wmf/1.18wmf1/includes/GlobalFunctions.php (modified) (history) /branches/wmf/1.18wmf1/includes/LinksUpdate.php (modified) (history) /branches/wmf/1.18wmf1/includes/api (modified) (history) /branches/wmf/1.18wmf1/includes/api/ApiQueryExternalLinks.php (modified) (history) /branches/wmf/1.18wmf1/includes/installer/DatabaseUpdater.php (modified) (history) /branches/wmf/1.18wmf1/maintenance/fixExtLinksProtocolRelative.php (added) (history) /branches/wmf/1.18wmf1/tests/phpunit/includes/GlobalFunctions/GlobalTest.php (modified) (history)

Diff [purge]

Index: branches/wmf/1.18wmf1/maintenance/fixExtLinksProtocolRelative.php
—	—	@@ -0,0 +1,81 @@
	2	+<?php
	3	+/**
	4	+ * Fixes any entries for protocol-relative URLs in the externallinks table,
	5	+ * replacing each protocol-relative entry with two entries, one for http
	6	+ * and one for https.
	7	+ *
	8	+ * This program is free software; you can redistribute it and/or modify
	9	+ * it under the terms of the GNU General Public License as published by
	10	+ * the Free Software Foundation; either version 2 of the License, or
	11	+ * (at your option) any later version.
	12	+ *
	13	+ * This program is distributed in the hope that it will be useful,
	14	+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
	15	+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
	16	+ * GNU General Public License for more details.
	17	+ *
	18	+ * You should have received a copy of the GNU General Public License along
	19	+ * with this program; if not, write to the Free Software Foundation, Inc.,
	20	+ * 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA.
	21	+ * http://www.gnu.org/copyleft/gpl.html
	22	+ *
	23	+ * @ingroup Maintenance
	24	+ */
	25	+
	26	+require_once( dirname( __FILE__ ) . '/Maintenance.php' );
	27	+
	28	+class FixExtLinksProtocolRelative extends LoggedUpdateMaintenance {
	29	+ public function __construct() {
	30	+ parent::__construct();
	31	+ $this->mDescription = "Fixes any entries in the externallinks table containing protocol-relative URLs";
	32	+ }
	33	+
	34	+ protected function getUpdateKey() {
	35	+ return 'fix protocol-relative URLs in externallinks';
	36	+ }
	37	+
	38	+ protected function updateSkippedMessage() {
	39	+ return 'protocol-relative URLs in externallinks table already fixed.';
	40	+ }
	41	+
	42	+ protected function doDBUpdates() {
	43	+ $db = wfGetDB( DB_MASTER );
	44	+ if ( !$db->tableExists( 'externallinks' ) ) {
	45	+ $this->error( "externallinks table does not exist" );
	46	+ return false;
	47	+ }
	48	+ $this->output( "Fixing protocol-relative entries in the externallinks table...\n" );
	49	+ $res = $db->select( 'externallinks', array( 'el_from', 'el_to', 'el_index' ),
	50	+ array( 'el_index' . $db->buildLike( '//', $db->anyString() ) ),
	51	+ __METHOD__
	52	+ );
	53	+ $count = 0;
	54	+ foreach ( $res as $row ) {
	55	+ $count++;
	56	+ if ( $count % 100 == 0 ) {
	57	+ $this->output( $count );
	58	+ wfWaitForSlaves();
	59	+ }
	60	+ $db->insert( 'externallinks',
	61	+ array(
	62	+ array(
	63	+ 'el_from' => $row->el_from,
	64	+ 'el_to' => $row->el_to,
	65	+ 'el_index' => "http:{$row->el_index}",
	66	+ ),
	67	+ array(
	68	+ 'el_from' => $row->el_from,
	69	+ 'el_to' => $row->el_to,
	70	+ 'el_index' => "https:{$row->el_index}",
	71	+ )
	72	+ ), __METHOD__, array( 'IGNORE' )
	73	+ );
	74	+ $db->delete( 'externallinks', array( 'el_index' => $row->el_index, 'el_from' => $row->el_from, 'el_to' => $row->el_to ), __METHOD__ );
	75	+ }
	76	+ $this->output( "Done, $count rows updated.\n" );
	77	+ return true;
	78	+ }
	79	+}
	80	+
	81	+$maintClass = "FixExtLinksProtocolRelative";
	82	+require_once( RUN_MAINTENANCE_IF_MAIN );
Property changes on: branches/wmf/1.18wmf1/maintenance/fixExtLinksProtocolRelative.php
___________________________________________________________________
Added: svn:eol-style
1	83	+ native
Index: branches/wmf/1.18wmf1/tests/phpunit/includes/GlobalFunctions/GlobalTest.php
—	—	@@ -831,42 +831,42 @@
832	832	}
833	833
834	834	/**
835		~~- * @dataProvider provideMakeUrlIndex()~~
	835	+ * @dataProvider provideMakeUrlIndexes()
836	836	*/
837		~~- function testMakeUrlIndex( $url, $expected ) {~~
838		~~- $index = wfMakeUrlIndex( $url );~~
839		~~- $this->assertEquals( $expected, $index, "wfMakeUrlIndex(\"$url\")" );~~
	837	+ function testMakeUrlIndexes( $url, $expected ) {
	838	+ $index = wfMakeUrlIndexes( $url );
	839	+ $this->assertEquals( $expected, $index, "wfMakeUrlIndexes(\"$url\")" );
840	840	}
841	841
842		~~- function provideMakeUrlIndex() {~~
	842	+ function provideMakeUrlIndexes() {
843	843	return array(
844	844	array(
845	845	// just a regular :)
846	846	'https://bugzilla.wikimedia.org/show_bug.cgi?id=28627',
847		~~- 'https://org.wikimedia.bugzilla./show_bug.cgi?id=28627'~~
	847	+ array( 'https://org.wikimedia.bugzilla./show_bug.cgi?id=28627' )
848	848	),
849	849	array(
850	850	// mailtos are handled special
851	851	// is this really right though? that final . probably belongs earlier?
852	852	'mailto:wiki@wikimedia.org',
853		~~- 'mailto:org.wikimedia@wiki.',~~
	853	+ array( 'mailto:org.wikimedia@wiki.' )
854	854	),
855	855
856	856	// file URL cases per bug 28627...
857	857	array(
858	858	// three slashes: local filesystem path Unix-style
859	859	'file:///whatever/you/like.txt',
860		~~- 'file://./whatever/you/like.txt'~~
	860	+ array( 'file://./whatever/you/like.txt' )
861	861	),
862	862	array(
863	863	// three slashes: local filesystem path Windows-style
864	864	'file:///c:/whatever/you/like.txt',
865		~~- 'file://./c:/whatever/you/like.txt'~~
	865	+ array( 'file://./c:/whatever/you/like.txt' )
866	866	),
867	867	array(
868	868	// two slashes: UNC filesystem path Windows-style
869	869	'file://intranet/whatever/you/like.txt',
870		~~- 'file://intranet./whatever/you/like.txt'~~
	870	+ array( 'file://intranet./whatever/you/like.txt' )
871	871	),
872	872	// Multiple-slash cases that can sorta work on Mozilla
873	873	// if you hack it just right are kinda pathological,
—	—	@@ -875,6 +875,15 @@
876	876	//
877	877	// Those will survive the algorithm but with results that
878	878	// are less consistent.
	879	+
	880	+ // protocol-relative URL cases per bug 29854...
	881	+ array(
	882	+ '//bugzilla.wikimedia.org/show_bug.cgi?id=28627',
	883	+ array(
	884	+ 'http://org.wikimedia.bugzilla./show_bug.cgi?id=28627',
	885	+ 'https://org.wikimedia.bugzilla./show_bug.cgi?id=28627'
	886	+ )
	887	+ ),
879	888	);
880	889	}
881	890
Index: branches/wmf/1.18wmf1/includes/GlobalFunctions.php
—	—	@@ -597,12 +597,12 @@
598	598	}
599	599
600	600	/**
601		~~- * Make a URL index, appropriate for the el_index field of externallinks.~~
	601	+ * Make URL indexes, appropriate for the el_index field of externallinks.
602	602	*
603	603	* @param $url String
604		~~- * @return String~~
	604	+ * @return array
605	605	*/
606		~~-function wfMakeUrlIndex( $url ) {~~
	606	+function wfMakeUrlIndexes( $url ) {
607	607	$bits = wfParseUrl( $url );
608	608
609	609	// Reverse the labels in the hostname, convert to lower case
—	—	@@ -642,7 +642,12 @@
643	643	if ( isset( $bits['fragment'] ) ) {
644	644	$index .= '#' . $bits['fragment'];
645	645	}
646		~~- return $index;~~
	646	+
	647	+ if ( $prot == '' ) {
	648	+ return array( "http:$index", "https:$index" );
	649	+ } else {
	650	+ return array( $index );
	651	+ }
647	652	}
648	653
649	654	/**
Property changes on: branches/wmf/1.18wmf1/includes/GlobalFunctions.php
___________________________________________________________________
Modified: svn:mergeinfo
650	655	Merged /trunk/phase3/includes/GlobalFunctions.php:r102951,102954
Index: branches/wmf/1.18wmf1/includes/installer/DatabaseUpdater.php
—	—	@@ -40,7 +40,8 @@
41	41	protected $shared = false;
42	42
43	43	protected $postDatabaseUpdateMaintenance = array(
44		~~- 'DeleteDefaultMessages'~~
	44	+ 'DeleteDefaultMessages',
	45	+ 'FixExtLinksProtocolRelative',
45	46	);
46	47
47	48	/**
Index: branches/wmf/1.18wmf1/includes/LinksUpdate.php
—	—	@@ -438,11 +438,13 @@
439	439	$arr = array();
440	440	$diffs = array_diff_key( $this->mExternals, $existing );
441	441	foreach( $diffs as $url => $dummy ) {
442		~~- $arr[] = array(~~
443		~~- 'el_from' => $this->mId,~~
444		~~- 'el_to' => $url,~~
445		~~- 'el_index' => wfMakeUrlIndex( $url ),~~
446		~~- );~~
	442	+ foreach( wfMakeUrlIndexes( $url ) as $index ) {
	443	+ $arr[] = array(
	444	+ 'el_from' => $this->mId,
	445	+ 'el_to' => $url,
	446	+ 'el_index' => $index,
	447	+ );
	448	+ }
447	449	}
448	450	return $arr;
449	451	}
Index: branches/wmf/1.18wmf1/includes/api/ApiQueryExternalLinks.php
—	—	@@ -69,6 +69,11 @@
70	70	$this->addOption( 'ORDER BY', 'el_from' );
71	71	}
72	72
	73	+ // If we're querying all protocols, use DISTINCT to avoid repeating protocol-relative links twice
	74	+ if ( $protocol === null ) {
	75	+ $this->addOption( 'DISTINCT' );
	76	+ }
	77	+
73	78	$this->addOption( 'LIMIT', $params['limit'] + 1 );
74	79	$offset = isset( $params['offset'] ) ? $params['offset'] : 0;
75	80	if ( $offset ) {
Property changes on: branches/wmf/1.18wmf1/includes/api
___________________________________________________________________
Modified: svn:mergeinfo
76	81	Merged /trunk/phase3/includes/api:r102951,102954
Index: branches/wmf/1.18wmf1/includes/AutoLoader.php
—	—	@@ -827,6 +827,7 @@
828	828	'DeleteDefaultMessages' => 'maintenance/deleteDefaultMessages.php',
829	829	'FakeMaintenance' => 'maintenance/Maintenance.php',
830	830	'Maintenance' => 'maintenance/Maintenance.php',
	831	+ 'FixExtLinksProtocolRelative' => 'maintenance/fixExtLinksProtocolRelative.php',
831	832	'PopulateCategory' => 'maintenance/populateCategory.php',
832	833	'PopulateLogSearch' => 'maintenance/populateLogSearch.php',
833	834	'PopulateLogUsertext' => 'maintenance/populateLogUsertext.php',
Property changes on: branches/wmf/1.18wmf1/includes/AutoLoader.php
___________________________________________________________________
Modified: svn:mergeinfo
834	835	Merged /trunk/phase3/includes/AutoLoader.php:r102951,102954
Property changes on: branches/wmf/1.18wmf1/includes
___________________________________________________________________
Modified: svn:mergeinfo
835	836	Merged /trunk/phase3/includes:r102951,102954
Property changes on: branches/wmf/1.18wmf1
___________________________________________________________________
Modified: svn:mergeinfo
836	837	Merged /trunk/phase3:r102951,102954

Follow-up revisions

Revision	Commit summary	Author	Date
r104716	Merge r103033 from REL1_18 to fix fatal on fixExtLinksProtocolRelative from r...	reedy	18:56, 30 November 2011

Past revisions this follows-up on

Revision	Commit summary	Author	Date
r102951	(bug 29854) Store protocol-relative links twice in the externallinks table, o...	catrope	09:13, 14 November 2011
r102954	Fix for r102949: don't delete too much	catrope	09:16, 14 November 2011

Comments

#Comment by Krinkle (talk | contribs) 18:22, 30 November 2011

Fatal error: Class 'LoggedUpdateMaintenance' not found in /srv/org/wikimedia/prototype/wikis/rc/maintenance/fixExtLinksProtocolRelative.php on line 27

LoggedUpdateMaintenance does not exist in 1.18wmf1 Maintenance.php

Status & tagging log

03:07, 12 December 2011 Tim Starling (talk | contribs) changed the status of r103056 [removed: new added: ok]
18:58, 30 November 2011 Reedy (talk | contribs) changed the status of r103056 [removed: fixme added: new]
18:22, 30 November 2011 Krinkle (talk | contribs) changed the status of r103056 [removed: new added: fixme]