lhc/web/wiklou.git
22 months agoLocalisation updates from https://translatewiki.net.
Translation updater bot [Tue, 1 Oct 2019 06:24:41 +0000 (08:24 +0200)]
Localisation updates from https://translatewiki.net.

Change-Id: I8c399d24532a18870561e054cca2d9651e0908ec

22 months agoAvoid dynamic call to static method in ResourceLoaderModule::buildContent()
Aaron Schulz [Mon, 30 Sep 2019 23:08:05 +0000 (16:08 -0700)]
Avoid dynamic call to static method in ResourceLoaderModule::buildContent()

Change-Id: I4e3d944b6468ad2f4bfc39532498d8ca38617689

22 months agoMerge "Selenium: Hide VE welcome and beta dialogs"
jenkins-bot [Mon, 30 Sep 2019 20:30:42 +0000 (20:30 +0000)]
Merge "Selenium: Hide VE welcome and beta dialogs"

22 months agoSelenium: Hide VE welcome and beta dialogs
Kosta Harlan [Mon, 30 Sep 2019 19:07:57 +0000 (21:07 +0200)]
Selenium: Hide VE welcome and beta dialogs

Remove uses of setLocalStorage (incompatible with chromedriver >= 75)
and also requires slightly fewer requests since BlankPage.open()
doesn't need to be invoked.

Bug: T234002
Change-Id: I4cfb4fe17a3c99bfcb7ff39ac6db60f5c5e8a687

22 months agoMerge "RevisionStore::newRevisionFromBatch should use Title::newFromRow"
jenkins-bot [Mon, 30 Sep 2019 19:47:47 +0000 (19:47 +0000)]
Merge "RevisionStore::newRevisionFromBatch should use Title::newFromRow"

22 months agoMerge "Parser: Add Title type hints"
jenkins-bot [Mon, 30 Sep 2019 19:47:40 +0000 (19:47 +0000)]
Merge "Parser: Add Title type hints"

22 months agoMerge "filebackend: Convert trigger_error to PSR log warning"
jenkins-bot [Mon, 30 Sep 2019 19:23:38 +0000 (19:23 +0000)]
Merge "filebackend: Convert trigger_error to PSR log warning"

22 months agoMerge "selenium: Fix inefficient use of MWBot in specs/page.js"
jenkins-bot [Mon, 30 Sep 2019 18:57:01 +0000 (18:57 +0000)]
Merge "selenium: Fix inefficient use of MWBot in specs/page.js"

22 months agoMerge "selenium: Use async-await for wdio-mediawiki/Api internals"
jenkins-bot [Mon, 30 Sep 2019 18:56:52 +0000 (18:56 +0000)]
Merge "selenium: Use async-await for wdio-mediawiki/Api internals"

22 months agoRevisionStore::newRevisionFromBatch should use Title::newFromRow
Petr Pchelko [Wed, 25 Sep 2019 20:17:38 +0000 (13:17 -0700)]
RevisionStore::newRevisionFromBatch should use Title::newFromRow

If the rows were obtained using RevisionStore::getQueryInfo with
'page' flags, the revision row already contains the fields needed
to construct the Title without an additional database query.

Change-Id: Ie36c85871a8996a5706c80d286854a9c8363905f

22 months agoMerge "selenium: Remove "RunJobs" wait from specialrecentchanges test"
jenkins-bot [Mon, 30 Sep 2019 18:34:21 +0000 (18:34 +0000)]
Merge "selenium: Remove "RunJobs" wait from specialrecentchanges test"

22 months agoMerge "selenium: Remove deprecated legacy methods"
jenkins-bot [Mon, 30 Sep 2019 18:11:33 +0000 (18:11 +0000)]
Merge "selenium: Remove deprecated legacy methods"

22 months agoMerge "selenium: Upgrade from webdriver v4 to v5"
jenkins-bot [Mon, 30 Sep 2019 18:09:50 +0000 (18:09 +0000)]
Merge "selenium: Upgrade from webdriver v4 to v5"

22 months agoMerge "RevisionStore: Introduce getContentBlobsForBatch"
jenkins-bot [Mon, 30 Sep 2019 17:46:08 +0000 (17:46 +0000)]
Merge "RevisionStore: Introduce getContentBlobsForBatch"

22 months agoMerge "tests: Replace assertions on count() == 0 with strict === []"
jenkins-bot [Mon, 30 Sep 2019 17:43:35 +0000 (17:43 +0000)]
Merge "tests: Replace assertions on count() == 0 with strict === []"

22 months agoMerge "Split down patch-actor-table.sql"
jenkins-bot [Mon, 30 Sep 2019 17:40:25 +0000 (17:40 +0000)]
Merge "Split down patch-actor-table.sql"

22 months agoSplit down patch-actor-table.sql
Reedy [Sat, 28 Sep 2019 14:48:19 +0000 (15:48 +0100)]
Split down patch-actor-table.sql

Bug: T227662
Change-Id: I024ff1d6f4c2726242138ba7e7f19480d9d2b948

22 months agoMerge "maintenance: Enable gzip in router.php for static files"
jenkins-bot [Mon, 30 Sep 2019 14:38:53 +0000 (14:38 +0000)]
Merge "maintenance: Enable gzip in router.php for static files"

22 months agotests: Replace assertions on count() == 0 with strict === []
Thiemo Kreuz [Mon, 30 Sep 2019 14:20:34 +0000 (16:20 +0200)]
tests: Replace assertions on count() == 0 with strict === []

The benefit of using count() is that the test would still succeed if
the return vfalue is not an array, but an iterable object. It seems
this is not needed.

Change-Id: I23529f6990aebe0cce86e236a21820fe74993204

22 months agoMerge "selenium: Restore sauce user/key configuration for wdio"
jenkins-bot [Mon, 30 Sep 2019 14:11:52 +0000 (14:11 +0000)]
Merge "selenium: Restore sauce user/key configuration for wdio"

22 months agoselenium: Fix inefficient use of MWBot in specs/page.js
Timo Tijhof [Sun, 29 Sep 2019 03:24:47 +0000 (04:24 +0100)]
selenium: Fix inefficient use of MWBot in specs/page.js

This test previously took 4 minutes to run locally and ended up
running the MWBot.login process 9 times.

After this, the specs/page tests only log-in once.

Bug: T234002
Change-Id: I374620a01f49d4da559070d0982bdbe4c1269e2e

22 months agoselenium: Use async-await for wdio-mediawiki/Api internals
Timo Tijhof [Sun, 29 Sep 2019 02:37:09 +0000 (03:37 +0100)]
selenium: Use async-await for wdio-mediawiki/Api internals

Bug: T234002
Change-Id: Ie4ad5248f832dd57f35710e0e9a91250e061b0bb

22 months agomaintenance: Enable gzip in router.php for static files
Timo Tijhof [Thu, 26 Sep 2019 21:57:50 +0000 (22:57 +0100)]
maintenance: Enable gzip in router.php for static files

Bug: T233992
Change-Id: Ie401180ac968210c9f923ad920bf15955c8551d7

22 months agoLocalisation updates from https://translatewiki.net.
Translation updater bot [Mon, 30 Sep 2019 06:31:33 +0000 (08:31 +0200)]
Localisation updates from https://translatewiki.net.

Change-Id: I2ff71393f57b3c397dae4e63d9607dcd1807343e

22 months agoMerge "resourceloader: Hard deprecate ResourceLoader::getLessVars"
jenkins-bot [Mon, 30 Sep 2019 00:34:28 +0000 (00:34 +0000)]
Merge "resourceloader: Hard deprecate ResourceLoader::getLessVars"

22 months agoresourceloader: Hard deprecate ResourceLoader::getLessVars
Fomafix [Sun, 29 Sep 2019 09:02:32 +0000 (11:02 +0200)]
resourceloader: Hard deprecate ResourceLoader::getLessVars

ResourceLoader::getLessVars is already deprecated since MW 1.32.

Change-Id: If2fbe7828d8cde73575c0306486835fa1ea06b84

22 months agoMerge "Added option to specify "Various authors" as author in extension credits using...
jenkins-bot [Sun, 29 Sep 2019 21:16:23 +0000 (21:16 +0000)]
Merge "Added option to specify "Various authors" as author in extension credits using "..." as the _only_ author name"

22 months agoExpand testDoQuickOperations() tests for FileBackend
Aaron Schulz [Wed, 18 Sep 2019 09:52:38 +0000 (02:52 -0700)]
Expand testDoQuickOperations() tests for FileBackend

Change-Id: I0633f13f07b5a3234b9bbae4840b064972f68fd9

22 months agoAdded option to specify "Various authors" as author in extension credits using "...
Jack Phoenix [Sun, 15 Sep 2019 20:34:27 +0000 (23:34 +0300)]
Added option to specify "Various authors" as author in extension credits using "..." as the _only_ author name

If the "author" array contains more than one entry and "..." is one of the entries in the array, "..." will be parsed as "others" (version-poweredby-others i18n message) like previously.

Change-Id: Ibebd3c38a674abd9f066b38e92c7f712f7b3ba8d

22 months agoselenium: Remove "RunJobs" wait from specialrecentchanges test
Timo Tijhof [Sun, 29 Sep 2019 01:32:47 +0000 (02:32 +0100)]
selenium: Remove "RunJobs" wait from specialrecentchanges test

This test was consistently reaching the RunJobs timeout of
10 tries and then giving up. This happens in CI, where it wasted
about ~ 5 seconds, and locally where it takes about 30s for mw.

Like so:
> RunJobs through requests to the main page (run 1).
> RunJobs detected 12 more queued job(s).
> RunJobs through requests to the main page (run 2).
> RunJobs detected 12 more queued job(s).

Etc, until it reaches run 10 and gives up. This also suggests
the whole thing isn't wokring, but that's a separate matter.

First and foremost, RecentChange objects are saved from a PRESEND
deferred update, which means it does not require a job queue wait
in order to appear on Special:RecentChanges.

This was added in e908d347ef2 (T199446) "to keep the number of
unrun jobs low". The idea being that this test makes an edit via
the API, which means it creates jobs without running them. On
the other hand, there's also lots of test steps that view pages
but don't create jobs, which would make this optimisation premature.
If and when we do encounter a test failing due to a job backlog,
I'm confident we can find a better solution at that time.

Bug: T234002
Bug: T225730
Bug: T199446
Change-Id: I9ef8c5b65d4b636572569912f04f48ed40aa0564

22 months agoselenium: Remove deprecated legacy methods
Timo Tijhof [Sun, 29 Sep 2019 01:59:56 +0000 (02:59 +0100)]
selenium: Remove deprecated legacy methods

These are from 2018, prior to introducing wdio-mediawiki in
commit 16a4d9da7bb81. Some extensions referenced these pageobjects
files through escalating "../../" references.

None of these are seen in Codesearch anymore, as well as no
references to these deprecated methods by other means.

Change-Id: Idbc373f13f9767cecd6404922b5c9773d796dda5

22 months agoselenium: Upgrade from webdriver v4 to v5
Timo Tijhof [Fri, 27 Sep 2019 03:08:00 +0000 (04:08 +0100)]
selenium: Upgrade from webdriver v4 to v5

* Options no longer needed or no longer exist in wdio v5:
  - coloredLogs: Now always on. The underlying 'chalk' library can still
    be influenced via the FORCE_COLOR environment variable.
  - screenshotPath: Removed. Was already disabled in our config.
  - deprecationWarnings: Meh.
  - 'sync: true' – On by default when `@wdio/sync` is installed.
    The wdio v5 config generator doesn't recommend setting manually.

* The selenium.sh script was removed. It existed to start and stop
  chromedriver for local use by developers. This is now done by
  the wdio-chromedriver-service. In WMF CI, Quibble starts its own
  chromedriver (as optimisation, reused across gated repos),
  which is why the 'selenium-test' entry points remains and skips
  this.

* The wdio-mediawiki package now requires wdio v5 and Node 10.
  This doesn't affect extension repos because versions are pinned.
  Upgrade may happen at the earliest convenience.

* Several WDIO methods changed names or signature. Full list at:
  <https://github.com/webdriverio/webdriverio/blob/v5.13.2/CHANGELOG.md#v500-2018-12-20>
  Highlights:
  - browser.element() is now browser.findElement() with "$()" as alias.
  - browser.localStorage replaced by browser.setLocalStorage.
  - browser.deleteCookie() requires `name` param. To delete all at once,
    there is a new method browser.deleteAllCookies().
  - Commands that return data no longer wrapped in `{ value: … }`.
    Values are now returned directly.
  - Custom config keys are now under browser.config instead of browser.options.
    Renamed our username/password keys to be mw-prefixed, to avoid clashes
    and reduce confusion with similar config keys.
  - browser.click(selector) and browser.getText(selector) no longer exist.
    Use $(selector).click() or .getText() instead.

* Fix "no such alert" warning from specs/page.js by removing the apparently
  redundant code.

Bug: T234002
Bug: T213268
Change-Id: I908997569ca8457997af30cb29e98ac41fae3b64

22 months agoselenium: Restore sauce user/key configuration for wdio
Timo Tijhof [Fri, 27 Sep 2019 22:09:59 +0000 (23:09 +0100)]
selenium: Restore sauce user/key configuration for wdio

I removed these in e78484109ae7 because they seemed redundant
given that the library used by wdio-sauce-service already reads
these when present. However, our current version of this package
clobbers this unconditionally with config.user/config.key, thus
not letting it reach that default.

Also add sauceConnect:true which means developers can run
localhost tests against sauce labs if they so choose.

Change-Id: I8d31a23374b2b9b6cd6f1405c7a2f1001da08f1a

22 months agoMerge "resourceloader: Add version to ResourceLoaderImage urls for long-cache"
jenkins-bot [Sat, 28 Sep 2019 20:20:32 +0000 (20:20 +0000)]
Merge "resourceloader: Add version to ResourceLoaderImage urls for long-cache"

22 months agoMerge "resourceloader: Simplify makeLoaderStateScript and makeLoaderSourcesScript"
jenkins-bot [Sat, 28 Sep 2019 20:02:18 +0000 (20:02 +0000)]
Merge "resourceloader: Simplify makeLoaderStateScript and makeLoaderSourcesScript"

22 months agoMerge "resourceloader: Remove support for $context === null in getContentObj"
jenkins-bot [Sat, 28 Sep 2019 20:00:24 +0000 (20:00 +0000)]
Merge "resourceloader: Remove support for $context === null in getContentObj"

22 months agoStringUtils: Clarify that isValidRegex is for PCRE regexps
Daimona Eaytoy [Thu, 19 Sep 2019 09:55:19 +0000 (11:55 +0200)]
StringUtils: Clarify that isValidRegex is for PCRE regexps

As suggested in c14571ba26e (I257a096319f1e).

Change-Id: Ia91d037f2f4bf8a1bad4eac65fecb1d3e2679d2d

22 months agoresourceloader: Simplify makeLoaderStateScript and makeLoaderSourcesScript
Fomafix [Sat, 28 Sep 2019 11:32:29 +0000 (13:32 +0200)]
resourceloader: Simplify makeLoaderStateScript and makeLoaderSourcesScript

The signatures without array are not used anymore.

Change-Id: Ice6a09ff6cba0c605ed1c89a25fd8e02af041b05

22 months agoMerge "docs: Fix typo on Doxygen main page"
jenkins-bot [Sat, 28 Sep 2019 12:24:40 +0000 (12:24 +0000)]
Merge "docs: Fix typo on Doxygen main page"

22 months agoFix double dot in filename
Reedy [Fri, 27 Sep 2019 16:22:11 +0000 (17:22 +0100)]
Fix double dot in filename

Follows-Up: I7617616df57f7468d06e9b52426b6851bfef0e7d
Change-Id: Ia8be2fd6a13d095e27a6489b0126763573dd6194

22 months agoMerge "resourceloader: Add $modules parameter to makeVersionQuery()"
jenkins-bot [Sat, 28 Sep 2019 01:26:59 +0000 (01:26 +0000)]
Merge "resourceloader: Add $modules parameter to makeVersionQuery()"

22 months agoresourceloader: Add version to ResourceLoaderImage urls for long-cache
Timo Tijhof [Thu, 26 Sep 2019 16:26:52 +0000 (17:26 +0100)]
resourceloader: Add version to ResourceLoaderImage urls for long-cache

The code previously here did not work well as it merely forwarded the
hash from the current web request. This had numerous issues:

1. It was often null because requests for stylesheets do not cary
   a version hash.
2. When requested by JavaScript, the version hash would be a
   combination-hash of many unrelated modules, thus when requested as
   part of different batches, it would produce different urls which
   is not ideal.

The impact of this is minimal currently because we basically never use
these urls, as SVGs are almost always embedded instead of ref'ed by url.
PNG urls are only generated for non-JS modules and then only used in older
browsers not supporting SVG. And, even after all that, for the edge case
of an SVG being ref'ed by url, they would be stored in LocalStorage by
mw.loader with the name+version of the module the image belonged to, not
the version hash of the batch request it came with.

Which means that, yes, localstorage key for "somemodule+someversion" would
have different values for different users, based on which batch the value
came with, because the image urls were using the version hash of the batch
request from ResourceLoaderContext. This is weird, but didn't cause bugs
or inefficiencies because the user would never be exposed to the other
possible urls for that image because we always check LocalStorage first.
It did cause fragmentation server-side in Varnish, though.

This is all fixed now by always including a version, and setting it to
the version of the module. This means there is no more Varnish fragmentation
for these. And it means that browsers are now allowed to cache the images
served from these urls for 30+ days (immutable) instead of only 5min,
which is what happened when they didn't have a version parameter (or set to
null).

Bug: T233343
Change-Id: I4af7fda03698ed4c288d154e7787fb2f3cbbe6c5

22 months agoMerge "Mark mediawiki.page.watch.ajax as targetable to mobile"
Jdlrobson [Fri, 27 Sep 2019 23:03:21 +0000 (23:03 +0000)]
Merge "Mark mediawiki.page.watch.ajax as targetable to mobile"

22 months agoMerge "tests: Replace PHPUnit's loose assertEquals(null) with assertNull()"
jenkins-bot [Fri, 27 Sep 2019 22:43:09 +0000 (22:43 +0000)]
Merge "tests: Replace PHPUnit's loose assertEquals(null) with assertNull()"

22 months agoMerge "build: Lower default Karma logLevel during local development"
jenkins-bot [Fri, 27 Sep 2019 22:23:51 +0000 (22:23 +0000)]
Merge "build: Lower default Karma logLevel during local development"

22 months agoMark mediawiki.page.watch.ajax as targetable to mobile
jdlrobson [Thu, 26 Sep 2019 15:56:58 +0000 (08:56 -0700)]
Mark mediawiki.page.watch.ajax as targetable to mobile

This is explicitly disabled by Minerva so this should have
no impact other than allow the loading of this module on
mobile.

Change-Id: Ib73fb995f5ccb16d99b505abad2466187c3dc5a5

22 months agoMerge "Remove tests that were skipped for MCR-enabled schema"
jenkins-bot [Fri, 27 Sep 2019 21:34:39 +0000 (21:34 +0000)]
Merge "Remove tests that were skipped for MCR-enabled schema"

22 months agoMerge ".pipeline/config.yaml: rename dev stage to publish"
jenkins-bot [Fri, 27 Sep 2019 20:58:32 +0000 (20:58 +0000)]
Merge ".pipeline/config.yaml: rename dev stage to publish"

22 months agoMerge "resourceloader: Add $context to static functions in ResourceLoader"
jenkins-bot [Fri, 27 Sep 2019 20:48:15 +0000 (20:48 +0000)]
Merge "resourceloader: Add $context to static functions in ResourceLoader"

22 months agoMerge "resourceloader: Add type hints for type ResourceLoaderContext"
jenkins-bot [Fri, 27 Sep 2019 20:22:10 +0000 (20:22 +0000)]
Merge "resourceloader: Add type hints for type ResourceLoaderContext"

22 months agoRemove tests that were skipped for MCR-enabled schema
Petr Pchelko [Fri, 27 Sep 2019 19:05:42 +0000 (12:05 -0700)]
Remove tests that were skipped for MCR-enabled schema

Pre-MCR schema is no longer supported, thus we don't need
the tests that are skipped for MCR-enabled schema.

Bug: T214308
Change-Id: I2d78ec668fde1e49710c54be8ab72c474f30f6b8

22 months agotests: Replace PHPUnit's loose assertEquals(null) with assertNull()
Thiemo Kreuz [Tue, 17 Sep 2019 14:03:28 +0000 (16:03 +0200)]
tests: Replace PHPUnit's loose assertEquals(null) with assertNull()

assertEquals( null, … ) still succeeds when the actual value is 0, false,
an empty string, even an empty array. All these should be reported as a
failure, I would argue.

Note this patch previously also touched assertSame( null ). I reverted
these. The only benefit would have been consistency within this codebase,
but there is no strict reason to prefer one over the other. assertNull()
and assertSame( null ) are functionally identical.

Change-Id: I92102e833a8bc6af90b9516826abf111e2b79aac

22 months agoMerge "Fix permission check on protection log"
jenkins-bot [Fri, 27 Sep 2019 18:38:16 +0000 (18:38 +0000)]
Merge "Fix permission check on protection log"

22 months agoMerge "Remove Revision::getRevisionText and gated pre-MCR schema access"
jenkins-bot [Fri, 27 Sep 2019 18:38:09 +0000 (18:38 +0000)]
Merge "Remove Revision::getRevisionText and gated pre-MCR schema access"

22 months agoMerge "rdbms: Remove obsolete comment in LoadBalancerSingle"
jenkins-bot [Fri, 27 Sep 2019 18:35:44 +0000 (18:35 +0000)]
Merge "rdbms: Remove obsolete comment in LoadBalancerSingle"

22 months agoRevisionStore: Introduce getContentBlobsForBatch
daniel [Thu, 26 Sep 2019 12:23:40 +0000 (14:23 +0200)]
RevisionStore: Introduce getContentBlobsForBatch

getContentBlobsForBatch provides a more low level interface than
newRevisionsFromBatch, allowing bulk access to serialized content
without the need to instantiate Title, RevisionRecord, RevisionSlots,
SlotRecord, and Content objects.

Bug: T228675
Bug: T234034
Change-Id: I8481ad211e2d9f11bc10ea10c16e78b74538d95b

22 months agoMerge "Force usage of MCR aware database schema"
jenkins-bot [Fri, 27 Sep 2019 17:52:33 +0000 (17:52 +0000)]
Merge "Force usage of MCR aware database schema"

22 months ago.pipeline/config.yaml: rename dev stage to publish
Brennen Bearnes [Fri, 27 Sep 2019 17:43:52 +0000 (11:43 -0600)]
.pipeline/config.yaml: rename dev stage to publish

This should align with the "publish" pipeline defined in
integration/config.

Bug: T218360
Change-Id: Iba27755416843dd547ac1889d932a3b5ea9d1fd2

22 months agoFix permission check on protection log
Ammar [Fri, 27 Sep 2019 13:39:15 +0000 (14:39 +0100)]
Fix permission check on protection log

*I2341e6f inverted the permission check, such that now the link is only shown to unprivileged users

Bug: T234017
Change-Id: I0977f1ab1a72840303aeca2367a30546d83117d4

22 months agordbms: Remove obsolete comment in LoadBalancerSingle
Lucas Werkmeister [Thu, 26 Sep 2019 15:49:48 +0000 (17:49 +0200)]
rdbms: Remove obsolete comment in LoadBalancerSingle

This comment was added in commit 0b583944e4, but
later rendered obsolete by commit 143333b172.

Change-Id: I1a64ab75a9883a88d5767722e27724520f8a0dcc

22 months agoRemove Revision::getRevisionText and gated pre-MCR schema access
Petr Pchelko [Fri, 20 Sep 2019 18:11:35 +0000 (11:11 -0700)]
Remove Revision::getRevisionText and gated pre-MCR schema access

- fixT22757.php appears to have been written for T22757 which has been
resolved, I assume it can be removed.

- compressOld.php - rev_text_id access is gated on MCRSchemaMigrationStage
READ_OLD. IT's safe to use the old_text and old_flags directly as we
select for them. Also the query ensures external blobs are not selected,
so we can use decompressData instead of Revision::getRevisionText.
Tested locally.

- recompressTracked.php uses a temporary table for tracking the blobs.
Table is populated by trackBlobs which explicitly queries for old_id,
so using it directly in recompressTracked is ok.
Tested locally.

- checkStorage.php the affected code attempts to fix broken revisions
from a dump, however, it looks like it doesn't work - the code expects
revision IDs as input, however it receives an array of text table old_id
values. Untested.

Bug: T198343
Change-Id: I753019565c15d270c831c995c07c7f1aad887cb6

22 months agoresourceloader: Add $context to static functions in ResourceLoader
Fomafix [Mon, 9 Sep 2019 15:50:13 +0000 (17:50 +0200)]
resourceloader: Add $context to static functions in ResourceLoader

This change allows to use the context in the functions.

The following internal static functions from ResourceLoader get now a
reference to the ResourceLoaderContext object:
* makeLoaderImplementScript
* makeLoaderStateScript
* makeLoaderRegisterScript
* makeLoaderSourcesScript

ResouceLoader::encodeJsonForScript is duplicated to
ResourceLoaderContext::encodeJson loading the debug mode from context.

ResourceLoader::encodeJsonForScript is kept for other usages without
context.

The debug mode is loaded from $context->getDebug() instead of from
ResourceLoader::inDebugMode(). This does not support to enable the debug
mode by setting the cookie 'resourceLoaderDebug' or the configuration
variable wgResourceLoaderDebug. Only the URL parameter debug=true
enables the debug mode. This should be sufficient for the subsequent
ResourceLoader requests. The tests don't need the global variable
wgResourceLoaderDebug anymore. The initial ResourceLoader context in
OutputPage still uses ResourceLoader::inDebugMode() with cookie and
global configuration variable.

This change adds the parameter $context with a ResourceLoaderContext
object to ResourceLoaderModule::getDeprecationInformation and deprecates
omitting the parameter. Ifa1a3bb56b731b83864022a358916c6aca5d7c10
updates this in extension ExtJSBase.

Bug: T229311
Change-Id: I5341f18625209446a6d006f60244990f65530319

22 months agoMerge "Split down patch-comment-table.sql"
jenkins-bot [Fri, 27 Sep 2019 15:18:11 +0000 (15:18 +0000)]
Merge "Split down patch-comment-table.sql"

22 months agoMerge "MWTestCase: include table prefix in domain for temp table only mode."
jenkins-bot [Fri, 27 Sep 2019 14:47:09 +0000 (14:47 +0000)]
Merge "MWTestCase: include table prefix in domain for temp table only mode."

22 months agodocs: Fix typo on Doxygen main page
apaskulin [Fri, 27 Sep 2019 14:45:59 +0000 (07:45 -0700)]
docs: Fix typo on Doxygen main page

This PR fixes a typo and removes a few extra words
from the Doxygen landing page.

Change-Id: Ieb09c0e72129f4c6b0ccd574daf2f9316f580259

22 months agoMerge "Speed up ApiQuerySiteinfoTest:testContinuation"
jenkins-bot [Fri, 27 Sep 2019 13:42:20 +0000 (13:42 +0000)]
Merge "Speed up ApiQuerySiteinfoTest:testContinuation"

22 months agoMerge "API: Use ConvertibleTimestamp::setFakeTime for testing curtimestamp"
jenkins-bot [Fri, 27 Sep 2019 12:01:55 +0000 (12:01 +0000)]
Merge "API: Use ConvertibleTimestamp::setFakeTime for testing curtimestamp"

22 months agoForce usage of MCR aware database schema
Clara Andrew-Wani [Tue, 24 Sep 2019 22:07:36 +0000 (18:07 -0400)]
Force usage of MCR aware database schema

Bug: T214308
Change-Id: I2ae4e82a12f730617563eb6f911fc406b9195391

22 months agoMWTestCase: include table prefix in domain for temp table only mode.
daniel [Fri, 27 Sep 2019 11:10:22 +0000 (13:10 +0200)]
MWTestCase: include table prefix in domain for temp table only mode.

This patch allows PageUpdaterTest and DerivedPageDataUpdaterTest to
function properly. Without this patch, the fail when run by themselves.

See phab ticket for analysis.

Bug: T234031
Change-Id: I0804b503bc8b860343a5141c0bfb25fe6855c87d

22 months agoRemove MySQL index aliasing for user_newtalk indexes
Brad Jorsch [Wed, 18 Sep 2019 16:44:15 +0000 (12:44 -0400)]
Remove MySQL index aliasing for user_newtalk indexes

While these have not yet caused problems, we should probably deprecate
the whole confusing "index aliasing" feature. Cleaning this up needs to
be done first.

Method MWLBFactory::setSchemaAliases() is being considered private due
to the @internal notation, and so is being removed without deprecation.

Bug: T233240
Change-Id: Ib04b7bc4af53382fde01bd8550cc7e361124255f

22 months agoSpeed up ApiQuerySiteinfoTest:testContinuation
Antoine Musso [Fri, 27 Sep 2019 08:42:56 +0000 (10:42 +0200)]
Speed up ApiQuerySiteinfoTest:testContinuation

The test fills $wgUrlProtocols with 86000 entries, forging oversized
strings which ends up eventually truncated to 100 bytes.  The aim is to
overflow $wgAPIMaxResultSize to trigger an apiwarn-truncatedresult
warning.

The test takes 230 ms on my machine and when generating code coverage
almost a minute.

A faster approach is just to lower the $wgAPIMaxResultSize limit to
trigger the warning.

Test plan:

  $ php tests/phpunit/phpunit.php \
      tests/phpunit/includes/api/ApiQuerySiteinfoTest.php \
      --filter testContinuation

Bug: T234016
Change-Id: I3dc96d394d8b98dda8f151d14544b829c92f9b10

22 months agoLocalisation updates from https://translatewiki.net.
Translation updater bot [Fri, 27 Sep 2019 06:24:47 +0000 (08:24 +0200)]
Localisation updates from https://translatewiki.net.

Change-Id: Ic52edc08609d9b9d03b7173ae43810f0dad140c1

22 months agoresourceloader: Add type hints for type ResourceLoaderContext
Fomafix [Fri, 27 Sep 2019 04:43:44 +0000 (06:43 +0200)]
resourceloader: Add type hints for type ResourceLoaderContext

Change-Id: I10fc4b9277f94fc495149fe0d0077b054090387b

22 months agoIn recompressTracked.php, fix the names of things related to child processes
Tim Starling [Fri, 27 Sep 2019 00:58:21 +0000 (10:58 +1000)]
In recompressTracked.php, fix the names of things related to child processes

The forbidden word was used in two senses, confusing the search and
replace done in 3d5016569da984050e.

Change-Id: I220a58fd536c828ab5e03a779e45f222d0298f9a

22 months agoMerge "objectcache: fully respect "pcTTL" in WANObjectCache instead of using INF...
jenkins-bot [Fri, 27 Sep 2019 00:13:37 +0000 (00:13 +0000)]
Merge "objectcache: fully respect "pcTTL" in WANObjectCache instead of using INF when >= 0"

22 months agoMerge "filebackend: avoid use of wfWikiId() in FileBackendGroup"
jenkins-bot [Fri, 27 Sep 2019 00:13:30 +0000 (00:13 +0000)]
Merge "filebackend: avoid use of wfWikiId() in FileBackendGroup"

22 months agoMerge "Add .pipeline/ with dev image variant"
jenkins-bot [Thu, 26 Sep 2019 23:20:56 +0000 (23:20 +0000)]
Merge "Add .pipeline/ with dev image variant"

22 months agoMerge "Add a run mode to $wgDisableQueryPageUpdate"
jenkins-bot [Thu, 26 Sep 2019 23:17:16 +0000 (23:17 +0000)]
Merge "Add a run mode to $wgDisableQueryPageUpdate"

22 months agofilerepo: make FileRepo::store/storeBatch() accept FSFile as similar methods already do
Aaron Schulz [Sun, 15 Sep 2019 19:32:08 +0000 (12:32 -0700)]
filerepo: make FileRepo::store/storeBatch() accept FSFile as similar methods already do

Change-Id: I92ce979a71fabd9e2095e7db43fa6fc8c2e0ef75

22 months agoAdd a run mode to $wgDisableQueryPageUpdate
Umherirrender [Mon, 19 Aug 2019 14:49:43 +0000 (16:49 +0200)]
Add a run mode to $wgDisableQueryPageUpdate

The run mode is used to show a different message on the special page,
instead of the current one "Updates for this page are currently
disabled. Data here will not presently be refreshed." even the data gets
updated with a different cron job.

Bug: T78711
Change-Id: Ib63d16bfea477dec43323b39671cc068530e2f0b

22 months agoresourceloader: Add $modules parameter to makeVersionQuery()
Timo Tijhof [Thu, 26 Sep 2019 17:34:17 +0000 (18:34 +0100)]
resourceloader: Add $modules parameter to makeVersionQuery()

The 'version' and 'modules' parameters are a somewhat problematic
part of the ResourceLoaderContext object as we often pass around
the context but may be dealing with only a subset of the modules
in the outer request, or even entirely different ones (e.g.
for OutputPage's fake context 'modules' and 'version' are both
null).

This is already visible in ClientHtml where we create a derivative
context just to call setModules() and have makeVersionQuery()
work. This is now fixed.

This change is in preparation for use in ResourceLoaderImage
(to fix T233343) where we'll need to compute a version for only
1 module of a larger set, and ideally without needing to create
another context.

Bug: T233343
Change-Id: Icc1a4fd1f58c4e49e58eee43ca4ba2de6cfffc76

22 months agoMerge "installer: Deprecate WebInstaller::getInfoBox, getWarningBox and getErrorBox"
jenkins-bot [Thu, 26 Sep 2019 20:16:01 +0000 (20:16 +0000)]
Merge "installer: Deprecate WebInstaller::getInfoBox, getWarningBox and getErrorBox"

22 months agoMerge "Clean up ar_usertext_timestamp index mess"
jenkins-bot [Thu, 26 Sep 2019 19:41:56 +0000 (19:41 +0000)]
Merge "Clean up ar_usertext_timestamp index mess"

22 months agoMerge "You are reviewing an old revision should be a warningbox"
jenkins-bot [Thu, 26 Sep 2019 19:41:48 +0000 (19:41 +0000)]
Merge "You are reviewing an old revision should be a warningbox"

22 months agoMerge "newRevisionsFromBatch: don't throw on duplicate row."
jenkins-bot [Thu, 26 Sep 2019 19:39:58 +0000 (19:39 +0000)]
Merge "newRevisionsFromBatch: don't throw on duplicate row."

22 months agoinstaller: Deprecate WebInstaller::getInfoBox, getWarningBox and getErrorBox
Volker E [Tue, 17 Sep 2019 01:53:29 +0000 (18:53 -0700)]
installer: Deprecate WebInstaller::getInfoBox, getWarningBox and getErrorBox

Since MW 1.31 there are messageBox, warningBox, errorBox and successBox
available which should be used instead.

These are simple wrappers around Html::messageBox and related methods
which should be used directly instead. Internal usage has been migrated
as of a9252abb1cf0, but these public methods remain for now. Once
removed we can also removed the internal Html::infoBox() method.

Bug: T232903
Change-Id: I2756c88987754c4e3f84d98c531d8396098cd7fe

22 months agoMerge "Fill GenderCache for used pages in action=query&prop=fileusage"
jenkins-bot [Thu, 26 Sep 2019 17:03:04 +0000 (17:03 +0000)]
Merge "Fill GenderCache for used pages in action=query&prop=fileusage"

22 months agoAPI: Use ConvertibleTimestamp::setFakeTime for testing curtimestamp
Brad Jorsch [Thu, 26 Sep 2019 16:29:29 +0000 (12:29 -0400)]
API: Use ConvertibleTimestamp::setFakeTime for testing curtimestamp

Mainly to avoid spurious test failures when CI is being extremely slow.

Bug: T233752
Change-Id: Ie2cdd84dc076a852fbdce52f661ef893f9a2d45b

22 months agoParser: Add Title type hints
Fomafix [Wed, 28 Aug 2019 12:39:25 +0000 (14:39 +0200)]
Parser: Add Title type hints

getTitle() and Title() will never return NULL anymore. When mTitle is
not set then a TypeError be thrown at runtime. The explicit local
checks for is_null( $this->mTitle ) are not needed anymore when
$this-getTitle() instead of $this->mTitle is used.

Change-Id: I09a178e5c6938954edb2949f13660227d6a01fbc

22 months agoMerge "Parser: Use 'Special:Badtitle/Parser' instead of 'NO TITLE'"
jenkins-bot [Thu, 26 Sep 2019 14:15:53 +0000 (14:15 +0000)]
Merge "Parser: Use 'Special:Badtitle/Parser' instead of 'NO TITLE'"

22 months agoCommentStore: Accept SCHEMA_COMPAT_* constants
Brad Jorsch [Fri, 20 Sep 2019 18:35:39 +0000 (14:35 -0400)]
CommentStore: Accept SCHEMA_COMPAT_* constants

Extensions doing an actor and comment migration at the same time will
likely want to use the same constants for each, and that will most
likely be using write-both/read-old and write-both/read-new for the
middle stages rather than write-both/read-both and write-new/read-both
as implemented by the MIGRATION_* constants. See Ie29fd05 for example.

This patch changes CommentStore's internal logic to work correctly when
passed write-both/read-old or write-both/read-new.

Bug: T233449
Change-Id: Iebec80f969ad1c4f9f51a4f25656319cca32b0dd

22 months agoexception: Let MediaWiki.php control final output for ErrorPageError
Timo Tijhof [Wed, 18 Sep 2019 18:05:42 +0000 (19:05 +0100)]
exception: Let MediaWiki.php control final output for ErrorPageError

The same way it does already for non-error output. This makes
it so that doPreOutputCommit() consistently happens between
the staging of output and the actual sending of output.

It is still allowed for code to bypass this, such as for fatal
errors and for handlers that disable OutputPage (like Special:Export).

But for cases where we do want to perform doPreOutputCommit(), it
should be run consistently between staging and sending so that it
can make appropiate decisions based on the current state of
OutputPage.

Previously, the state of OutputPage seen by doPreOutputCommit()
would be the broken/incomplete output of a seemingly succesful
(possibly cacheable) user action, which would then, after
doPreOutputCommit() runs, be completely replaced by $e->report()/
$out->showErrorPage().

This is a prerequisite for being able to reliably send cookie-block
cookies on error pages (next patch).

Bug: T233594
Change-Id: Iaeaf5e55a5868e6be534ddda73f3b56b9d6ef8f0

22 months agoLocalisation updates from https://translatewiki.net.
Translation updater bot [Thu, 26 Sep 2019 06:23:52 +0000 (08:23 +0200)]
Localisation updates from https://translatewiki.net.

Change-Id: I49be2a7100bff079264a48b07e14b0edf30472be

22 months agofilebackend: Convert trigger_error to PSR log warning
Timo Tijhof [Thu, 26 Sep 2019 00:57:01 +0000 (01:57 +0100)]
filebackend: Convert trigger_error to PSR log warning

* Add 'filebackend' name to the context.
* Convert message to a normalized form for better de-duplication
  in Logstash so that it is easier to see whether something is rare
  or common. Without this, each instance is unique and it might seem
  rare in a small sample of log messages.

This does not fully solve T231107, because we'd really want to
know at least the file name the issue is about, but it seems
the FileBackendStore does not have access to that right now,
that's for a future commit. (It also can't be logged from the
caller right now because this method does not expose the result
of the operation in question).

Bug: T231107
Change-Id: Iaf18d7d5c11334c84077667aa147d99fadb76a30

22 months agoYou are reviewing an old revision should be a warningbox
jdlrobson [Mon, 9 Sep 2019 22:07:12 +0000 (15:07 -0700)]
You are reviewing an old revision should be a warningbox

It's warning you about the outdated nature of the content.

Given this message can be customised and thus styled, this will
have presentation issues on certain wikis (see screenshots).

If we go ahead with this change we'll probably want to give editors
a 2 weeks heads up via tech news.

https://phabricator.wikimedia.org/F30285062
https://phabricator.wikimedia.org/F30285074
https://phabricator.wikimedia.org/F30285084

Bug: T232415
Change-Id: I7dab00d5500937c03166398d92ea65a599bf3e22

22 months agoFill GenderCache for used pages in action=query&prop=fileusage
Umherirrender [Tue, 10 Sep 2019 17:42:58 +0000 (19:42 +0200)]
Fill GenderCache for used pages in action=query&prop=fileusage

Add an utility function
ApiQueryBase::executeGenderCacheFromResultWrapper

GenderCache stops working when there are more than 1000 cache missed
and returning the default value (T200238)
Fill the cache with all needed users avoids this behaviour
and it saves one query per user page.

Change-Id: I911dcb160a7b169091b9e8f66fb3908d0f2a1ba4

22 months agoMerge "Pass "services" through from coreRoutes.json to ObjectFactory"
jenkins-bot [Wed, 25 Sep 2019 15:48:02 +0000 (15:48 +0000)]
Merge "Pass "services" through from coreRoutes.json to ObjectFactory"

22 months agordbms: fix active GTID filtering in DatabaseMysqlBase
Aaron Schulz [Wed, 19 Jun 2019 08:10:15 +0000 (09:10 +0100)]
rdbms: fix active GTID filtering in DatabaseMysqlBase

In masterPosWait(), only $pos will have the known active domain/server set
since it usually comes from getMasterPos(). However, the reference position,
from getReplicaDB(), does not have the active domains set since querying
gtid_domain_id on the replica would be incorrect and getting a connection
to the master could be expensive.

Remove obsolete hacks for jobs that used to store master positions.

Also, use the regular Database::query() method for stylistic consistency.

Bug: T224422
Change-Id: I41bbb9f337e46451aa17788dbd446db4a213a5a7

22 months agoMerge "Updated wikimedia/avro from 1.8.0 to 1.9.0"
jenkins-bot [Wed, 25 Sep 2019 07:53:14 +0000 (07:53 +0000)]
Merge "Updated wikimedia/avro from 1.8.0 to 1.9.0"

22 months agoLocalisation updates from https://translatewiki.net.
Translation updater bot [Wed, 25 Sep 2019 06:34:37 +0000 (08:34 +0200)]
Localisation updates from https://translatewiki.net.

Change-Id: I27b08eefba6d5f7e8358fb0b2ea7f84552cb38c0