Timo Tijhof [Sat, 18 Aug 2018 03:29:53 +0000 (04:29 +0100)]
benchmarks: Minor cleanup for strtr benchmark
Only bench with variable input ("indirect"). The static string is
too unrealistic and not worth testing for imho, if it even makes
a difference at all.
Also bump the default count from 100 to 10,000 given it's so tiny.
Change-Id: Iccc35af4bd3c5b0967983ddfacd1d4ead235e4a4
jenkins-bot [Sat, 18 Aug 2018 01:18:38 +0000 (01:18 +0000)]
Merge "TitleFormatter: Remove indirection from getText()"
jenkins-bot [Sat, 18 Aug 2018 01:14:51 +0000 (01:14 +0000)]
Merge "TitleFormatter: Don't look up namespace name for ns 0"
jenkins-bot [Sat, 18 Aug 2018 01:05:28 +0000 (01:05 +0000)]
Merge "TitleValue: Avoid function call for internal property"
jenkins-bot [Sat, 18 Aug 2018 01:04:31 +0000 (01:04 +0000)]
Merge "test: Assert that API generators have unique prefixes"
jenkins-bot [Sat, 18 Aug 2018 00:55:18 +0000 (00:55 +0000)]
Merge "Turn ApiPrefixUniquenessTest into a structure test"
Antoine Musso [Tue, 19 Jun 2018 07:50:12 +0000 (09:50 +0200)]
test: Assert that API generators have unique prefixes
Signed-off-by: Brad Jorsch <bjorsch@wikimedia.org>
Change-Id: I30758f2ac95fd8ae361ef8079abdfb0a82d92e34
Kunal Mehta [Fri, 17 Aug 2018 23:41:02 +0000 (16:41 -0700)]
TitleFormatter: Remove indirection from getText()
At the end of the day, the output is identical to LinkTarget::getText().
Change-Id: Ifddafb79d7602ba5e368cc40f99c56e22a56d506
Kunal Mehta [Fri, 17 Aug 2018 23:21:27 +0000 (16:21 -0700)]
TitleFormatter: Don't look up namespace name for ns 0
It's hardcoded to empty string, so we can skip some code.
Profiling (after switching the benchmark script to use NS_MAIN) showed
no noticable improvements in getPrefixedText().
Change-Id: Id6b044a51648d0a3f58331ac0427f9d5cd9d8f0f
Kunal Mehta [Fri, 17 Aug 2018 23:15:20 +0000 (16:15 -0700)]
TitleValue: Avoid function call for internal property
Change-Id: I16342469a8122bd46cada11fd1f1d3f88d2e8d1e
jenkins-bot [Fri, 17 Aug 2018 21:42:39 +0000 (21:42 +0000)]
Merge "Inject SpecialPageFactory into Parser"
jenkins-bot [Fri, 17 Aug 2018 21:22:58 +0000 (21:22 +0000)]
Merge "Pass through 'helpInline' to OOUI FieldLayout and make true by default"
Kunal Mehta [Mon, 11 Jun 2018 23:51:42 +0000 (16:51 -0700)]
Turn ApiPrefixUniquenessTest into a structure test
This way it gets run during extension test runs as well.
Temporarily add a hack allowing 'wbeu' duplicates until T196962 is fixed.
Change-Id: Ic89a22a2ff4525585de9e290a1d47d22cfaaac5e
jenkins-bot [Fri, 17 Aug 2018 21:18:22 +0000 (21:18 +0000)]
Merge "mediawiki.util: Only call $.fn.updateTooltipAccessKeys if accesskey is set"
Translation updater bot [Fri, 17 Aug 2018 21:12:20 +0000 (23:12 +0200)]
Localisation updates from https://translatewiki.net.
Change-Id: Ie7ab020ed7722038335b1a27c6faaaa7e07f8f3f
jenkins-bot [Fri, 17 Aug 2018 21:02:58 +0000 (21:02 +0000)]
Merge "Mass conversion to SpecialPageFactory service"
jenkins-bot [Fri, 17 Aug 2018 20:45:43 +0000 (20:45 +0000)]
Merge "Add index on rc_this_oldid"
jenkins-bot [Fri, 17 Aug 2018 20:40:47 +0000 (20:40 +0000)]
Merge "Improve profileinfo.php documentation"
Ed Sanders [Tue, 12 Jun 2018 14:46:48 +0000 (15:46 +0100)]
Pass through 'helpInline' to OOUI FieldLayout and make true by default
This change is meant to improve the layout of Special:Preferences in
particular, but it will affect any OOUI form using help messages. It
should be a harmless or beneficial change for most of them.
Previous behavior can be restored by passing `'help-inline' => false`
to the HTMLForm factory after 'help-message'/'help-messages'/'help'.
For example:
* Special:Preferences?ooui=1#mw-prefsection-watchlist
* Before: https://phabricator.wikimedia.org/
F25025213
* After: https://phabricator.wikimedia.org/
F25025181
* Special:ChangeEmail
* Before: https://phabricator.wikimedia.org/
F25073327
* After: https://phabricator.wikimedia.org/
F25073328
* Special:BotPasswords/foo
* Before: https://phabricator.wikimedia.org/
F25073324
* After: https://phabricator.wikimedia.org/
F25073326
Bug: T181854
Change-Id: Ica67fe4081dfaa8eb9e8f56fdb93530750e47012
jenkins-bot [Fri, 17 Aug 2018 19:32:28 +0000 (19:32 +0000)]
Merge "jquery.expandableField: Remove module unused since 1.22"
jenkins-bot [Fri, 17 Aug 2018 19:26:42 +0000 (19:26 +0000)]
Merge "resourceloader: Increase minification cache version"
jenkins-bot [Fri, 17 Aug 2018 19:10:10 +0000 (19:10 +0000)]
Merge "JavaScriptMinifier: Fix bad state after ternary in object literal"
jenkins-bot [Fri, 17 Aug 2018 19:08:37 +0000 (19:08 +0000)]
Merge "jquery.suggestions: Remove dead code for returning options"
Timo Tijhof [Fri, 17 Aug 2018 18:24:01 +0000 (19:24 +0100)]
jquery.expandableField: Remove module unused since 1.22
Was originally created for Vector, which was originally part
of MediaWiki core. It hasn't been used for a while, and Vector
is in its own repository now.
Found 0 uses of this anywhere in Wikimedia Git, nor elsewhere
indexed by MediaWiki Codesearch.
Bug: T202154
Change-Id: I72ab5ad7f25be87b2f6d2f5f41a672b6a48b8005
Kunal Mehta [Wed, 15 Aug 2018 01:11:59 +0000 (18:11 -0700)]
Inject SpecialPageFactory into Parser
Change-Id: I6a6a94cbdafdc724ce02408cd9e744e7b3eda92b
Aryeh Gregor [Sun, 12 Aug 2018 09:08:58 +0000 (12:08 +0300)]
Mass conversion to SpecialPageFactory service
Change-Id: Ia6e1e819ec6cbe8bf75b820109f51d47863e31fc
jenkins-bot [Fri, 17 Aug 2018 18:53:27 +0000 (18:53 +0000)]
Merge "Make SpecialPageFactory a service"
jenkins-bot [Fri, 17 Aug 2018 18:41:58 +0000 (18:41 +0000)]
Merge "Deprecate wfGetCache() and wfGetMainCache()"
jenkins-bot [Fri, 17 Aug 2018 18:28:38 +0000 (18:28 +0000)]
Merge "Deprecate $wgParser"
jenkins-bot [Fri, 17 Aug 2018 18:19:48 +0000 (18:19 +0000)]
Merge "Add release notes for $wgContLang deprecation"
Aryeh Gregor [Tue, 7 Aug 2018 10:58:31 +0000 (13:58 +0300)]
Make SpecialPageFactory a service
Calling SpecialPageFactory methods statically is now soft-deprecated.
SpecialPageFactory::resetList() is a no-op, and I changed tests
in core to use overrideMwServices() instead.
Methods that fell back to $wgUser now require a User object being passed.
Depends-On: Ie1f80315871085b9fd4763a265b588849d94414d
Change-Id: Id8a92d57743f790b7d8c377c033cef38d1bb24de
Kunal Mehta [Tue, 14 Aug 2018 23:43:25 +0000 (16:43 -0700)]
Deprecate wfGetCache() and wfGetMainCache()
The former is already a wrapper around ObjectCache::getInstance().
The latter was identical to ObjectCache::getLocalClusterInstance().
Bug: T115890
Change-Id: Ib4e43bc8d3f4ac9f7a453e36dcce9b3d962666ba
Kunal Mehta [Tue, 14 Aug 2018 23:22:15 +0000 (16:22 -0700)]
Deprecate $wgParser
Bug: T160811
Change-Id: I7bcf4bbd14268ec069d97848c016e0098eaf0be4
Timo Tijhof [Fri, 17 Aug 2018 17:50:36 +0000 (18:50 +0100)]
jquery.suggestions: Remove dead code for returning options
returnValue = (key in obj) ? undefined : obj[key];
By standard, this cannot return anything other undefined.
This code has been broken since the introduction of the module
in 2012 (
32377424b, r72349).
The functionality also didn't have test (naturally), and isn't
used anywhere in Wikimedia Git. Simply remove it.
Also remove the confusing $() wrapping of the non-getter return
value that provides chainability.
The 'this' inside a jQuery method is already an instance of jQuery,
pulling it through $() again achieves very little.
Change-Id: Id13861d42b65a759d74670d8c5850aecee4daf7d
Kunal Mehta [Tue, 14 Aug 2018 23:21:20 +0000 (16:21 -0700)]
Add release notes for $wgContLang deprecation
Bug: T160815
Change-Id: I4dfdd8516b243f3eeccb69fe518a932178534aa2
Timo Tijhof [Fri, 17 Aug 2018 17:23:00 +0000 (18:23 +0100)]
mediawiki.util: Only call $.fn.updateTooltipAccessKeys if accesskey is set
Avoids overhead of mw.Message#parse, jqueryMsg.parser, and various
other DOM operations.
Bug: T202151
Change-Id: Id6304c9223806af6d469d9b4c67e60263f4b17ac
Fomafix [Fri, 17 Aug 2018 16:06:47 +0000 (18:06 +0200)]
Message: Fix error on getTitle() without language
Steps to reproduce:
Old:
> $msg = new Message( 'brackets' )
> print $msg->getTitle()
Error from line 478 of /.../includes/Message.php: Call to a member function equals() on boolean
New:
> $msg = new Message( 'brackets' )
> print $msg->getTitle()
MediaWiki:Brackets
This got broken in
796d62d0.
Change-Id: I52cde36787e348378c4df18d90aa577a5600e06b
Timo Tijhof [Fri, 17 Aug 2018 14:58:57 +0000 (15:58 +0100)]
Improve profileinfo.php documentation
* Move $wgEnableProfileInfo to DefaultSettings.php
* The configuration variable to enable the entry point was added
together with the entry point itself in
9af3c09e5c (r9846).
* Change references to StartProfiler.php to refer to LocalSettings.php,
given the former is deprecated since 1.31 (I4e8dd9558132).
Change-Id: I7ca5f2deace8645f06bebd915630c1de05c84bc5
jenkins-bot [Fri, 17 Aug 2018 01:08:26 +0000 (01:08 +0000)]
Merge "Html: Add test coverage for inlineScript()"
jenkins-bot [Fri, 17 Aug 2018 00:27:13 +0000 (00:27 +0000)]
Merge "languages: Add coverage for 'ar' and 'ml' normalize()"
jenkins-bot [Fri, 17 Aug 2018 00:09:10 +0000 (00:09 +0000)]
Merge "mediawiki.jqueryMsg: Combine the two generated data scripts"
jenkins-bot [Thu, 16 Aug 2018 23:40:39 +0000 (23:40 +0000)]
Merge "mw.ForeignStructuredUpload: Use 'helpInline' instead of custom styles"
jenkins-bot [Thu, 16 Aug 2018 22:31:26 +0000 (22:31 +0000)]
Merge "Fix eslint warnings and switch to error code"
jenkins-bot [Thu, 16 Aug 2018 22:21:36 +0000 (22:21 +0000)]
Merge "Warn against certain jQuery methods"
jenkins-bot [Thu, 16 Aug 2018 22:10:54 +0000 (22:10 +0000)]
Merge "benchmarks: Create $normBenchs before the run instead of during"
Bartosz Dziewoński [Thu, 16 Aug 2018 20:51:18 +0000 (22:51 +0200)]
mw.ForeignStructuredUpload: Use 'helpInline' instead of custom styles
Before: https://phabricator.wikimedia.org/
F25025765
After: https://phabricator.wikimedia.org/
F25025764
Change-Id: I423c786e30570b8febc888b61450e8fce9185ba6
jenkins-bot [Thu, 16 Aug 2018 20:28:10 +0000 (20:28 +0000)]
Merge "Remove incorrect timezone conversion from date parameters"
Timo Tijhof [Thu, 16 Aug 2018 20:09:59 +0000 (21:09 +0100)]
mediawiki.jqueryMsg: Combine the two generated data scripts
No need to output two mw.jqueryMsg.setParserDefaults() calls.
Follows-up
7e65b6b3a766c1.
Also:
* Fix an inline comment that referred to a non-existent method.
* Document why we pass deep=true.
Change-Id: I7414d4188a4d933e2132f80135cbda87241580c0
Translation updater bot [Thu, 16 Aug 2018 20:06:27 +0000 (22:06 +0200)]
Localisation updates from https://translatewiki.net.
Change-Id: Icc72d5230b3c3c09c5095b52cc3d2191751a109f
jenkins-bot [Thu, 16 Aug 2018 17:22:03 +0000 (17:22 +0000)]
Merge "rdbms: remove some dead code in LoadBalancer::setDomainPrefix"
Timo Tijhof [Thu, 16 Aug 2018 17:11:14 +0000 (18:11 +0100)]
benchmarks: Create $normBenchs before the run instead of during
Separates the concerns a bit better, and also makes the code easier
to debug with less distracting steps during the running of the
benchmark.
Change-Id: Ia5a18216cb77d39fd60cd76b23ebceee7324a250
Timo Tijhof [Sat, 11 Aug 2018 16:31:45 +0000 (17:31 +0100)]
resourceloader: Increase minification cache version
Follows-up
1f5f6fc204.
Bug: T201606
Change-Id: I0b5af067d1d44880c8122343f8e4a1b47e998619
Timo Tijhof [Mon, 13 Aug 2018 17:38:14 +0000 (18:38 +0100)]
JavaScriptMinifier: Fix bad state after ternary in object literal
The following pattern of input (found in jquery.js) triggered
this bug:
call( {
key: 1 ? 0 : function () {
return this;
}
} );
The open brace changes state to PROPERTY_ASSIGNMENT (for object literals).
The colon after 'key' sets state to PROPERTY_EXPRESSION.
Each individual parts of an expression (identifiers and literal values)
is recognised with state *_EXPRESSION_OP, such as PROPERTY_EXPRESSION_OP.
The '1' after 'key:' correctly sets the state to PROPERTY_EXPRESSION_OP.
Upto there it goes well, but after that it goes wrong.
The question mark (TYPE_HOOK) in this context was wrongly switching
back to PROPERTY_EXPRESSION. That is a problem because that does not
handle TYPE_COLON, which meant '0: function' was seen together as a
sequence of continuous PROPERTY_EXPRESSION_OP where TYPE_FUNC may
not be handled.
Fixed by changing handling of TYPE_HOOK in PROPERTY_EXPRESSION to
switch states to EXPRESSION_TERNARY, and also performing a push
so that ternary handling can pop back to the property expression.
This mirrors the handling that already exists for ternaries in
the regular handling of EXPRESSION/EXPRESSION_OP (as opposed to
the variant for object literal properties).
Bug: T201606
Change-Id: I6104c839cfc3416257543b54a91b74cb4aa4193b
Aaron Schulz [Thu, 16 Aug 2018 13:31:11 +0000 (06:31 -0700)]
rdbms: remove some dead code in LoadBalancer::setDomainPrefix
Change-Id: I594a390b446c00eb604b0be63f67b2c8cbd45758
Timo Tijhof [Wed, 15 Aug 2018 23:03:19 +0000 (00:03 +0100)]
title: Disable the failing tests from TitlePermissionTest
Bug: T201776
Change-Id: I088bd797225e0c60c66de4d4d1aa12d0b5bf67d8
jenkins-bot [Thu, 16 Aug 2018 10:02:18 +0000 (10:02 +0000)]
Merge "Set getEditConflictMainTextBox parameter type to array"
jenkins-bot [Thu, 16 Aug 2018 09:56:57 +0000 (09:56 +0000)]
Merge "selenium: undo page"
WMDE-Fisch [Thu, 16 Aug 2018 09:00:06 +0000 (11:00 +0200)]
Set getEditConflictMainTextBox parameter type to array
This is only called at one place in core and used in the TwoColConf
extention. The passed argument is always an array.
Change-Id: I4ca94442b79dc8871ed72943be733d100e093e44
Niklas Laxström [Thu, 16 Aug 2018 06:28:17 +0000 (08:28 +0200)]
Unbreak Special:JavaScriptTest
Follow-up to
dec8009
Change-Id: I812253cc1a504a8bb876700026dbfd35a727458b
jenkins-bot [Thu, 16 Aug 2018 05:54:35 +0000 (05:54 +0000)]
Merge "rdbms: use Database::getDomainId for transaction logging calls"
jenkins-bot [Thu, 16 Aug 2018 05:37:02 +0000 (05:37 +0000)]
Merge "Fix spelling of word "necessary""
jenkins-bot [Thu, 16 Aug 2018 04:33:58 +0000 (04:33 +0000)]
Merge "rdbms: Disable DBConnRef::selectDB() for sanity"
petarpetkovic [Wed, 15 Aug 2018 23:33:12 +0000 (01:33 +0200)]
Fix spelling of word "necessary"
Bug: T201491
Change-Id: Ic983889a5702b8b8552c88ebc4d676f35e9fe61e
petarpetkovic [Mon, 13 Aug 2018 17:07:14 +0000 (19:07 +0200)]
Fix some of the common typos
* supress -> suppress (Except in backup_LogTest.php)
* recomend -> recommend
* becuase -> because
* accross -> across
Bug: T201491
Change-Id: I8faa4e6cc688b3ee204b3f79ab55eb7b65cc1fdd
jenkins-bot [Wed, 15 Aug 2018 22:34:18 +0000 (22:34 +0000)]
Merge "Have WANCacheReapUpdate::getEventAffectedKeys() accept LinkTarget"
jenkins-bot [Wed, 15 Aug 2018 20:32:21 +0000 (20:32 +0000)]
Merge "Actually inject InterwikiLookup into MediaWikiTitleCodec"
jenkins-bot [Wed, 15 Aug 2018 20:32:16 +0000 (20:32 +0000)]
Merge "Have LinkCache::getMutableCacheKeys() accept LinkTarget"
jenkins-bot [Wed, 15 Aug 2018 20:13:08 +0000 (20:13 +0000)]
Merge "Title: Access properties directly instead of wrapper functions"
Translation updater bot [Wed, 15 Aug 2018 20:03:09 +0000 (22:03 +0200)]
Localisation updates from https://translatewiki.net.
Change-Id: I4b07075f2c8b2965f542f08cfbb62d71b23d9099
jenkins-bot [Wed, 15 Aug 2018 19:57:26 +0000 (19:57 +0000)]
Merge "Make LinkBatch::setCaller chainable"
jenkins-bot [Wed, 15 Aug 2018 17:56:16 +0000 (17:56 +0000)]
Merge "Update OOUI to v0.28.0"
Volker E [Tue, 14 Aug 2018 23:55:10 +0000 (16:55 -0700)]
Update OOUI to v0.28.0
Release notes:
https://phabricator.wikimedia.org/diffusion/GOJU/browse/master/History.md;v0.28.0
Bug: T109529
Bug: T183601
Bug: T183645
Bug: T185751
Bug: T189113
Bug: T189953
Bug: T190441
Bug: T190581
Bug: T194104
Bug: T197004
Bug: T198845
Bug: T201329
Bug: T201871
Depends-on: I3c000504661e190ac8752f69419bed28785601b5
Change-Id: I0c101a942011f6095678ffbd90f9bd536132091c
jenkins-bot [Wed, 15 Aug 2018 15:06:03 +0000 (15:06 +0000)]
Merge "Add option to populateChangeTagDef not to update the count"
jenkins-bot [Wed, 15 Aug 2018 15:01:32 +0000 (15:01 +0000)]
Merge "Remove dead code from Title::secureAndSplit()"
jenkins-bot [Wed, 15 Aug 2018 14:48:05 +0000 (14:48 +0000)]
Merge "TitleValue: Don't use Assert for basic type checks"
jenkins-bot [Wed, 15 Aug 2018 14:43:44 +0000 (14:43 +0000)]
Merge "Have TitleParser::parseTitle() default to NS_MAIN to match Title"
Amir Sarabadani [Mon, 13 Aug 2018 10:44:11 +0000 (12:44 +0200)]
Add option to populateChangeTagDef not to update the count
I need this when I'm picking this up to continue populating
Change-Id: I52c74d3c1c59ee966027d5d9cb1a2bd3cf6dfc7a
Kunal Mehta [Wed, 15 Aug 2018 09:28:51 +0000 (02:28 -0700)]
Have WANCacheReapUpdate::getEventAffectedKeys() accept LinkTarget
Nothing specifically requires TitleValue here.
Change-Id: I62ee1995c49071db2871149b941277b29836a94a
Niklas Laxström [Wed, 15 Aug 2018 08:57:21 +0000 (10:57 +0200)]
Make LinkBatch::setCaller chainable
To allow for one-liners such as $lb->setCaller( __METHOD__ )->execute();
Change-Id: I45cd81d9e583f09aacb578ffb87d7c11b0e13d57
Aaron Schulz [Wed, 15 Aug 2018 08:20:30 +0000 (01:20 -0700)]
rdbms: use Database::getDomainId for transaction logging calls
Change-Id: I360ca633b145dc8e510bb00b542ac44933283943
Kunal Mehta [Wed, 15 Aug 2018 07:15:04 +0000 (00:15 -0700)]
Actually inject InterwikiLookup into MediaWikiTitleCodec
Change-Id: Ic469f0fb8068bfb06b741d957b21e66dfa839cd6
jenkins-bot [Wed, 15 Aug 2018 06:06:02 +0000 (06:06 +0000)]
Merge "rdbms: split out private LoadBalancer::openLocalConnection method"
Kunal Mehta [Wed, 15 Aug 2018 05:56:27 +0000 (22:56 -0700)]
Have LinkCache::getMutableCacheKeys() accept LinkTarget
Change-Id: Id84590b3aa285e31a358877167fe8ff502077f3c
Kunal Mehta [Wed, 15 Aug 2018 05:33:56 +0000 (22:33 -0700)]
Title: Access properties directly instead of wrapper functions
Since this is pretty hot code, avoid the performance hit of a function
call when we can directly access the property.
Change-Id: I287de5eb003582b5704e9d93e2fc99d16c0ffb2a
jenkins-bot [Wed, 15 Aug 2018 05:34:15 +0000 (05:34 +0000)]
Merge "Make ExternalStoreDB "wiki" context override the server "dbname" field"
jenkins-bot [Wed, 15 Aug 2018 05:23:46 +0000 (05:23 +0000)]
Merge "JavaScriptMinifier: Merge $push and $pop into $model"
Kunal Mehta [Wed, 15 Aug 2018 05:13:18 +0000 (22:13 -0700)]
Remove dead code from Title::secureAndSplit()
Change-Id: I08810f0ab20fc1608f2d3b2248dcd5106044080b
Aaron Schulz [Wed, 15 Aug 2018 04:19:12 +0000 (21:19 -0700)]
rdbms: split out private LoadBalancer::openLocalConnection method
Change-Id: I2138b3d28f8f1a8ea706165fbe3a7e3f0d858447
Kunal Mehta [Wed, 15 Aug 2018 04:09:44 +0000 (21:09 -0700)]
TitleValue: Don't use Assert for basic type checks
Previously, `new TitleValue()` was roughly twice as slow as
`Title::makeTitle()`. Switching to basic is_int/is_string checks
makes TitleValue roughly twice as fast as Title!
Tested with benchmarkTitleValue.php.
Bug: T201801
Change-Id: Idccf159983145c1c11af9182292e1d5f704ba077
Kunal Mehta [Wed, 15 Aug 2018 03:58:34 +0000 (20:58 -0700)]
Have TitleParser::parseTitle() default to NS_MAIN to match Title
Matches the default parameters of Title::newFromText().
Change-Id: Id08312d424be9ae7e64688bb21687204ae77c560
jenkins-bot [Wed, 15 Aug 2018 03:27:48 +0000 (03:27 +0000)]
Merge "MergeHistory: Fix flaky test due to relative timestamp"
Aaron Schulz [Tue, 14 Aug 2018 07:27:27 +0000 (00:27 -0700)]
rdbms: Disable DBConnRef::selectDB() for sanity
Bug: T193565
Change-Id: I4276d1a7d77a019e0e60dab4b9ec36c93e418037
Aaron Schulz [Wed, 15 Aug 2018 01:44:40 +0000 (18:44 -0700)]
Make ExternalStoreDB "wiki" context override the server "dbname" field
This is all that is needed for b/c and going beyond that can break
foreign wiki (domain) external store access if matching per-wiki DB
names *are* used for external store.
Follow up to
92e4ace7eae61.
Bug: T200471
Change-Id: I877139ff659f542da04d4d8f5ef3297dbfcfd734
jenkins-bot [Wed, 15 Aug 2018 01:52:55 +0000 (01:52 +0000)]
Merge "DateTimeInputWidget: Rename LESS var to align with WikimediaUI Base"
jenkins-bot [Wed, 15 Aug 2018 01:24:36 +0000 (01:24 +0000)]
Merge "resourceloader: Remove $wgResourceLoaderLESSVars support"
Timo Tijhof [Wed, 15 Aug 2018 00:11:43 +0000 (01:11 +0100)]
MergeHistory: Fix flaky test due to relative timestamp
In PHP, `strtotime('tomorrow')` evaluates to the first midnight
after the current time.
Which means on Aug 14 at 23:59:59, strtotime('tomorrow') refers
to a timestamp merely 1 second in the future (Aug 15 00:00:00).
This causes the test to fail because this timestamp is computed
in a data provider, before the test itself starts. In order to
pass, it needs to be far enough in the future that it will also
be after the point in time where the test creates the page and
revisions in question.
Fix this by:
* ... moving it into the test itself, where even plain time()
should probably suffice, but that's for another time.
* ... using +24*3600 instead of strtotime().
Bug: T201976
Change-Id: Ice5d54af4fb7cffa8db9b657796ba6ebecdaffa0
Volker E [Sat, 11 Aug 2018 00:14:49 +0000 (17:14 -0700)]
DateTimeInputWidget: Rename LESS var to align with WikimediaUI Base
Change-Id: I1d3a33cb9911250793dc07758d384d7a625de1c5
jenkins-bot [Wed, 15 Aug 2018 00:49:40 +0000 (00:49 +0000)]
Merge "Parser: Call firstCallInit() in getTags/getFunctionHooks"
jenkins-bot [Wed, 15 Aug 2018 00:05:28 +0000 (00:05 +0000)]
Merge "Fast manual restart for populateContentTables.php"
jenkins-bot [Wed, 15 Aug 2018 00:03:14 +0000 (00:03 +0000)]
Merge "RemexDriver: improved tracing"