Using PinnableElement in the TOC allows us to reuse styles associated with all PinnableElements, and helps ensure consistency between page tools and the TOC. This is something that would probably need to be done for T324505 anyway
Rather than updating PinnableElement to accept a class prop, I chose to rename '.sidebar-toc' to '.vector-toc', making the TOC's naming conventions consistent with the rest of Vector.
"sidebar" doesn't make sense anymore anyway, as the TOC can be in multiple locations and it's best to avoid the "sidebar" term
Bug: T324877
Change-Id: I5b9228380f5c4674ef424d33127a5cb4010822da
It's used by click tracking so marking with inline comment that
we need to be careful if we plan to remove it in our current
refactoring.
Change-Id: I6aee01363d2d150844e61fb27ab584c8ee78c4ae
Test ServiceWiring.php using tests copied from CentralAuth. Because
phpunit does not support marking a file as covered, the file is
marked as ignored for code coverage. However, these tests do not
fully cover the single service as these tests only verify that the
call to the service does not result in an error.
Change-Id: I6100d242aa61975cab7c44b35616cb2b19762cbf
- Rename data-portlets-main-menu to data-main-menu
- Rename MainMenuContents.mustache to MainMenu
- Replace usage of .vector-main-menu-contents in favor of .vector-main-menu
- Add classes to PinnableContainer
- Use PinnableContainer and PinnableElement in main menu
- Remove MainMenuDropdown.less
Bug: T317900
Change-Id: I59b3acd3d56cd5761e5978607634dfb9a88f60e3
This allows '.vector-main-menu' to correspond to MainMenu.mustache, and 'vector-page-tools' with PageTools.mustache
Bug: T317900
Change-Id: I65c0d7cffbdf1cf9e59cde0c0fc4dca788e799de
- getTocData is moved into VectorComponentTableOfContents and it's test file
The following changes were made to the main menu, toc and page tools PHP components
- Avoid passing in $skin to the constructor
- Handle isPinned logic inside the component
- Add a public ID constant to the components
- Dropdown data for each feature use the same naming convention
Bug: T317900
Change-Id: I77a617a6c1d93bccd3b6e59353299f5534624e53
The UnpinnedContainer and PinnedContainer are versions of the
same component but with different states. Recognize them by
organizing them in the same subfolder. Update all existing usages.
Bug: T317900
Change-Id: I2c25937190997764fa23e0b581be5538ba4d06de
While removing this we also noticed that we check the value of
isMainPage. This doesn't seem like a good idea as most main pages
do not have a table of contents, so it seems like adding
unnecessary complexity for a state that doesn't exist in practice.
The existing code path also doesn't work as it adds a table of contents
unstyled to the page.
Bug: T324874
Change-Id: Idaeff6ace5912ea74ed9d335526027c4690ac8fa
* Leverage the infrastructure around feature management to handle the page tools
pinning and persistence
* Make pinnableHeader.js leverage features.js if the data-feature-name attribute
is set
* Sets tests/.eslintrc.json ecmaVersion to 2018 to enable destructuring in test
files.
* Adds a isPinned helper method to pinnableElement
* Add a logged in requirement so that the pinned feature is disabled for
anon users.
Bug: T322051
Change-Id: Ib86282216882fa94e37b7088a3f4bd0c1bcf6cd4
In preparation for 856718 where pinnableElement.js makes use of features.js,
move features.js and limitedWidthToggle.js out of the skins.vector.js module and
into the skins.vector.es6 module. This will make it easier to use by
pinnableElement.js without needing the es6 module to depend on the es5 module.
This does have the negative side-effect of causing the limited width feature to
not be supported by IE11 (and other non-ES6 browsers), however this tradeoff was
discussed with our product manager to be acceptable. Additionally, this
maintains the status quo as the toggle button does not currently show in IE11
(which may be a bug).
Bug: T322051
Change-Id: If0e8cb98deabe847c2cc71fddb90ca36d15e5f8f
As far as I can tell, this has been dead code ever since it was
introduced in change Ica040cd18d (commit aa10668e6d), and was never
called; the effective search URL was initially constructed via inputs to
the search form (e.g. a hidden input for wprov), and later the separate
urlGenerator.js was added. (Also, “suggestion” is not a valid
Special:Search paramater as far as I can tell.)
Also add some tests for the separate urlGenerator.js, to make up for the
other tests being removed.
This reduces the module size of skins.vector.search enough that we can
decrease the bundle size test config to 3.2 kB [KiB].
Change-Id: I6be5ba362402c2c2ec582d9a0192c80f46e7a7ce
If the wgVectorSearchClient supports it (the default implementation
doesn’t), add a visible-item-limit to the cdx-typeahead-search, and wire
up the resulting load-more event (new in Codex v0.3) to load additional
search results on scroll.
The hard-coded visibleItemLimit (7) is chosen to match the default limit
of the wbsearchentities API, but ultimately arbitrary; we can look into
how to make this number configurable later, if necessary.
This increases the module size enough that we need to bump the bundle
size a bit more.
Bug: T322333
Change-Id: Iadade9cbf48457cfeabc78439624602ec3f98782
Co-Authored-By: Jon Robson <jdlrobson@gmail.com>
Needed-By: I67fac3b209d6a1ab2661e1e1c0681edd8472ac6c
Generalize LimitedWidthRequirement into a more reusable
UserPreferenceRequirement that can be used by both the limited width feature and
the persistent pinning feature (and possibly others in the future).
* Removes existing logic that checks whether the option is not null. Given that
skin.json sets the default [1], presumably this isn't needed.
* Adds unit test
[1] 65af26a258/skin.json (L163)
Bug: T322051
Change-Id: I7f228cf81a65b2eb22dbe94d2384b6c9f6da91f2
This patch involves cached HTML changes
- Replace '-content-container' id with '-unpinned-container'
- Replace '-content' id with '-pinnable-element'
- Rename pinnableHeader.js to pinnableElement.js
- Replace PinnableDropdownContents.mustache with PinnableElement/Open.mustache and Close
- Add PinnedContainer/Open & Close and UnpinnedContainer/Open & Close
- Rename .vector-dropdown-content to .vector-pinnable-element
- Add new PinnableElement.less stylesheet
Bug: T318013
Change-Id: I85aec387f87126a17e760fd9fd10e10572ff3152
The Dropdown component should be considered the dropdown folder going
forward. This should be distinguished as a different component that
composes the Dropdown component
Since the template itself is rather simple it seems preferable to have
duplicate templates that describe the actual component (PageTools,
VariantsDropdown and LanguageButton)
MenuDropdown is renamed Dropdown and styles that don't correspond to the
Dropdown component are moved to the relevant components.
MenuContents mustache template (future component) is added to DRY up
several of the new templates.
Bug: T320927
Change-Id: I220d0a477b3f05b278bd2cec7dd601c7b3b17c6f
[Visual Changes in this patch]
This patch introduces changes to whitespace inside the Mustache template
which corresponds to an HTML change, resulting
33 minor changes in pixel, the spacing of the ToC label shifts by
1 pixel
- Remove page tools feature flag from TOC
- Remove usage of .vector-toc-collapsed and .vector-toc-not-collapsed
- Remove now unused `moveToc()`
- Update jest tests to use PinnableHeader
Bug: T318013
Change-Id: I82f23e69b0249c844af9e45fec342217a0755893
Visual changes when the flag is on
Follow up to If3aed0ac401b0abc80c3ad52806eb85b33d43f06
More context at T322673#8383888
Bug: T322673
Change-Id: I64a47fa869247f758c6369c2e0ff8b12f87f9d9e
Changes:
* Code is moved from SkinVector to SkinVectorLegacy verbatim since it is now only
needed by SkinVectorLegacy
so we remove it:
** This step broke MainMenu rendering so this was captured in VectorComponentMainMenu
** The isLegacy code no longer makes sense and is removed in a follow up
* storybook no longer run on CI
Authors: Bernard Wang and Jon Robson
Bug: T321102
Bug: T319349
Change-Id: Ib424ca1c767161fdae8c0aeedfe662b7fa039ff6
To aid moving towards a MainMenu component we must first make subcomponents
for the language alert and opt-out link (which share code which
I've captured in VectorComponentMainMenuAction)
Code is lifted into SkinVector22 to allow us to easily make the template
change without any breaking visual changes
Bug: T319349
Bug: T322089
Change-Id: Ibb69d029a9fb6cee3482e15a60a7358361bd2405
* Introduce a generalized feature toggle system that uses user
preferences or localStorage for anons (right now the latter is out
of scope but will be explored in a follow up)
* Feature flagged to VisualEnhancementNext for now, given the dependency
on icon size
Bug: T319449
Change-Id: I7343a3f38b720411d5ef5f3414f25f475b0bb84a
Restricted to the feature flag (?vectorvisualenhancementnext=1)
for now to allow us to not worry about caching.
mw-ui-icon-flush-left and mw-ui-icon-flush-right have been in core
for some time and using them means we don't have to manage hardcoded
values for these.
We'll need to think of .mixin-vector-flush-left-left
and .mixin-vector-flush-right-margin-left separately - I wonder
if these could be revised to use these 2 standard classes.
Additional changes:
- Drop unused mixin-vector-flush-right-margin-left mixin
Bug: T321504
Bug: T317583
Change-Id: I51f8e31be6771a3fb32fb07dc22d0c2872b5706d
The legacy search wprov behavior, implemented in the WikimediaEvents
extension in searchSatisfaction.js, is to use acrw1_ as the prefix (i.e.
with an underscore before the index), and acrw1_-1 for user searches
that didn’t select a suggestion (not acrw1). Make the Vue search match
this in both cases.
Bug: T317682
Change-Id: I6c84533e1811233ff2727d501327471d73fc7b62
Making this a feature part of the feature management system is integral
to making this a toggle and will allow us to explore making this
persistent in future.
Bug: T319447
Bug: T319449
Change-Id: I80c7b892a6891094854b4154db90917b67986102
To avoid array query values to be used, see warning there:
"Array values are discarded for security reasons. Use {@see getArray} or
{@see getIntArray}."
This also fix that the falsy value "0" is checked against the regex
Bug: T321267
Change-Id: I29bc4a9a7fef5a6cadc0c6aa9fa1f4a03ccf9705
This ensures that the wprov parameter is included when users follow the
link as a link (click, middle-click, etc.), and also prepares us for a
future where pressing Enter after selecting a search result navigates to
that result’s URL instead of submitting the form. It also matches the
behavior of the legacy search form.
We put this in App.vue + instrumentation.js, not in urlGenerator.js,
because we also want wprov to be added when custom URL generators or
search clients are used. (The reason for instrumentation.js instead of
purely App.vue is just that it’s easier to test there.)
In the tests, we need to update @wikimedia/mw-node-qunit so that we have
a sufficiently functional mw.Uri() mock.
Bug: T317682
Change-Id: I765d3bbf89b2253add7b50305c362e4bbc9ecceb
* Removes server rendered vector-tab-noicon class from legacy Vector as it is
not currently used.
* Adds menuTabs.js responsible for adding the vector-tab-noicon class when menu
items are added to tabbed menus. This is only used in vector-22.
* Adds unit test
Bug: T320691
Change-Id: Iffad86125f8754305f592ddc19d894866bd6dedc
This allows the URL to the other wiki's rest.php to be configured
exactly, rather than assuming that it has the same wgScriptPath as the
current wiki. This is necessary to make this feature work on PatchDemo,
where wgScriptPath looks like '/123abc456/w'.
$wgVectorSearchHost is removed, since nothing uses it except PatchDemo
(where it's broken) and development setups.
Bug: T319494
Change-Id: Ife042f4f683d366a31a642723746d4aa80774c03
* Ensure addPortletLink hook is only run once
* Mark more menu as not supporting icons so icons are not added
Bug: T317491
Bug: T318495
Change-Id: I99450a5b0410e88cc7cdb2753b9b4256e3fe41db
Change I765d3bbf89 pushes the module over the configured maximum allowed
size (3 KiB, see bundlesize.config.json); shave off some bytes elsewhere
to bring it below the limit again. IMHO all of these changes should be
acceptable:
* arrow functions are already used elsewhere in this module;
* using the mw.config.get() fallback argument is normal (it slightly
changes behavior, but I don’t think explicitly setting the search
client or URL generator to a falsy value and expecting to get the
default behavior should be considered supported);
* not quoting [name="search"] matches [name=title] immediately above;
* using forEach() with a function reference is still readable (initApp()
is now called with extra arguments, but that doesn’t matter).
Change-Id: I45dda26cb59279d91804b0c2bbf12174fa78ee12
Change I45dda26cb5 removes some code, and in the process reduces the
branch coverage of the skins.vector.search module below the threshold of
25% that CI requires; add some tests for instrumentation.js to bring the
branch coverage up again.
Change-Id: I5d7ae019b3219b410c658583a6d72cf56d0cc1c5
- Changes the "skip to content" link to point to "#bodyContent"
instead of "#content".
- Adds a `aria-labelledby` attribute to #bodyContent.
- Changes the table of contents "back to top" link to
point to "#" instead of "#top-page".
- Removes "#top" link as it appeared unused.
- Updates Jest snapshots.
Bug: T312818
Change-Id: I86443066d45c818d6f137d70ebe36798eba66f13
The removal here is actually relatively straightforward.
References inside OverridableConfigRequirementTest are kept as
the A/B test capability remains after this removal and the tests
here do not relate to the code removed, just examples of how it
can be used.
Bug: T302027
Change-Id: I2dd92adaff1221d8213723e9ee60aa02787d4cc7
Can be tested by appending ?vectorvisualenhancementnext=1
to URL
Bug: T310838
Bug: T234990
Bug: T234550
Depends-On: I76d0d94c9006cc5f5680849ecdd1c382c16e34ba
Depends-On: Ib7c3021db014827b4b88cac855afc0b54a360f8c
Change-Id: Ie2ffa5c3ecf270c1bb1f315937023ae7ace5ed30
* Updated usages to use SkinTemplateNavigation::Universal
in preparation for the hard deprecation and removal of the hook.
Bug: T310017
Change-Id: I133c7479da294c0b8c908ad8d34690297f08200b
This is a common need for features, and having these use a standardized
class name will make using them a lot easier.
Change-Id: I0e16c26878e7d4399d2bf57f236523d214951a27
This class is no longer needed in Codex and will be removed.
This patch adds a similar dynamic class to the Vector search
app, to be used for showing/hiding the search button on
focus/blur or hover/leave.
Bug: T316893
Change-Id: I738c0f24dcd06ddeb9179cfedc85ed73a6504f1e
Adds tests for the addPortletLinkHandler, which should
add an icon element to items added via 'mw.util.addPortletLink'
and conditionally move elements from the toolbar into
the more menu at narrow widths.
- Adds tests for dropdownMenu.js
- Adds associated Jest snapshots
- Stubs the mediawiki.page.ready module
- Updates mockMediawiki.js to 4.6.0
- Adds global mock for mw.util.showPortlet()
- Cleans up tests using local versions of these mocks
Bug: T314798
Change-Id: I81394d840321916756a41a23c40d96c4b6341931
Throw a runtime exception if the TOC experiment is setup correctly.
Otherwise this leads to confusion.
There must be 3 buckets, with 0 unsampled and equally weighted
control and experiment.
Bug: T313435
Change-Id: I09da238ff17b77a053b4dc132624ea19a86253b9
- Make 'updateMenuItem' generic, rename to 'updateItemData'
- Make 'appendListItemClass' generic, rename to 'appendClassToItem'
- Remove 'appendClassToListItem' and replace with 'appendClassToItem'
- Create 'updateMenuItemData', 'updateLinkData', and 'updateDropdownMenuData' wrapper functions to be used in Hooks and SkinVector
- Add and update tests
Bug: T307901
Change-Id: I4b12a0c1aab1224d2658bad30ee629f7266c5b7e
Visual changes:
- TOC icon in sticky header
- Small update to the narrow screens collapsed TOC spacing when open
Bug: T311103
Depends-on: Ic94fda79bd14856ccda14985d5860aa54d3118d0
Change-Id: I977092f951ba2843816609e706bce4b99583f623
Removes the .mw-ui-button styles from the userpage link
and create account link in the user menu.
Overrides the userpage redlink style so that the link
remains blue even if the userpage doesn't exist.
Visual changes: userpage link and create account link
in header are no longer .mw-ui-buttton styles, but
look like standard blue links. The user links menu shifts
slightly because of this as well.
Bug: T312157
Change-Id: Id98e566952e5875527f38d1edbc8f6e058af4518
- Include temporary feature requirement for TOC A/B test.
- Assumes 100% of logged-in users with even/odd user ids
being assigned to treatment/control buckets respectively.
- Sampling rates passed in by config are not considered
during bucketing.
- Update hook for adding needed TOC A/B test body classes.
- Add test for temp feature.
Note: the temporary feature requirement and associated hooks
should be removed once the 2nd TOC A/B test concludes.
Bug: T313435
Change-Id: If9c75235614af289cd50182baab29bec3155eb81
For idwiki/viwiki, we wish to run the sticky header edit button AB
test so that treatment1 group sees the sticky header without edit
buttons, treatment2 groups sees the sticky header with edit buttons,
and the control/unsampled groups see no sticky header at all.
This patch overrides the configuration to make the sticky header
w/o edit buttons for treatment1, sticky header w/ edit buttons for
treatment2, and hides sticky header for everyone else. This depends
on a configuration with the treatment groups having "treatment1"
and "treatment2" as substrings in their bucket names.
The full configuration for idwiki/viwiki would be something like
the following:
```
$wgVectorStickyHeader = [
"logged_in" => true,
"logged_out" => false,
];
$wgVectorStickyHeaderEdit = [
"logged_in" => true,
"logged_out" => false,
];
$wgVectorWebABTestEnrollment = [
"name" => "vector.sticky_header_edit",
"enabled" => true,
"buckets" => [
"unsampled" => [
"samplingRate" => 0
],
"noStickyHeaderControl" => [
"samplingRate" => 0.34
],
"stickyHeaderNoEditButtonTreatment1" => [
"samplingRate" => 0.33
],
"stickyHeaderEditButtonTreatment2" => [
"samplingRate" => 0.33
]
],
];
```
Bug: T312573
Change-Id: I15c360fdf5393f5594602acc33b5b916e904016d
- Rename 'vector-user-menu-more' class to 'vector-user-menu-overflow'
- Update storybook and tests
Visual changes
- Small intentional change on the create account button
Bug: T306662
Change-Id: I371bb11903d8cdd8f0da89266fcf549050c0da8c
There are two issues with context in this test case:
1. The $vectorTemplate object was not given the $context object.
Calling setContext() is the same as MediaWiki would normally do
after calling SkinFactory, e.g. as in RequestContext::getSkin.
2. The "main" glboal state was leaking into the test by mutating
the result of getMain() instead of creating your own for use
in the test.
Change-Id: I09eb8c3a38c0d4e2c85461b61e6f14ab00995016
- Update related selectors, styles.
- Remove unneeded styles.
- Remove link hijack js.
- Simplify hook to only add experiment name to body.
Bug: T310527
Change-Id: I25527261d529a16e28f1b90f2f5af234d26fd40f
[Visual changes]
This should result in 9 visual regression failures relating to
increased height of search results and loading bar
[More details about change]
- Migrate search app from Vue 2 to Vue 3; update tests
accordingly
- Remove dependence on WVUI and use Codex instead, via the special
`@wikimedia/codex-search` package
- Update search app to use CdxTypeaheadSearch, which no longer
takes in props related to the search client or fetch start/end
instrumentation. Instead, directly use the restSearchClient
and call fetch start/end events in the search app.
- Handle hideDirection in the search app/API response formatting
code, not within the TypeaheadSearch component
- Handle showing/hiding the search button in the app
- Move the WVUI URL generator into Vector
- Update server-rendered search box styles to match design updates
included with CdxTypeaheadSearch
- Replace references to WVUI with references to Codex
- Update values of various LESS variables to match Codex, and update
searchBox styling to prevent jankiness when the searchBox is replaced
with the Codex TypeaheadSearch component
The VectorWvuiSearchOptions config variable has been maintained and
will be updated to a code-agnostic name in a future patch.
Bug: T300573
Bug: T302137
Bug: T303558
Bug: T309722
Bug: T310525
Co-Authored-By: Anne Tomasevich <atomasevich@wikimedia.org>
Change-Id: I59fa3a006d988b14ebd8020cbd58e8d7bedbfe01
Use overrideConfigValues instead of installMockMwServices and setMwGlobals.
This ensures that configuration overrides are handled consistently.
This also fixes a data provider that relied on a service, which causes
the test to fail when not run via the pgpunit wrapper.
Depends-On: I898927717ce961d98617a7fcd9c7fa8e19bec412
Change-Id: I6354fa39e1e9adf4be6eb6b26db82b8f106c593e
SkinVector and Hooks both had code to add classes and handle Vector specific template data. This patch simplifies the way we handle menu data to always use Hooks:updateMenuItem. This has an additional side effect of removing instances of mw-ui-icon-before.
Bug: T306628
Change-Id: I73514a0eada4d92705b70e7c2ebd91092fc12544
Follow up to c5cfd4d
1) Partial paths are incorrect
These are not a problem with our current Mustache template
parser but could break with any changes in our PHP implementation
2) Add dedicated class to legacy menus
On the longer term this will allow us to further separate the
old and new CSS.
Change-Id: I056b033855c28f919a4af99784620503f10b9dcb
When the vector-sticky-header-enabled class is removed from the body
it means the feature is not enabled, it doesn't mean the sticky header
should be invisible.
Call the hide method instead, and move it out of the function given
its a side effect
Bug: T308343
Change-Id: I4ecd6524146f203af926847812e20275c9573cab
Add action=edit to the exclusions in $wgVectorMaxWidthOptions
and update the parsing of the query string values for that config
variable to accept a regex.
Bug: T307725
Change-Id: Iba526033d45e18cb340a2648378d3d90ef3ae3c6
- `MediaWiki\Skins\Vector\Tests` is now the prefix for all tests in the skin
- we followed PSR conventions of following folder structure after the prefix
- Optimize imports/use order
- update namespace in skin.json
Bug: T303102
Change-Id: Ib76374d81d973c83adfd6c8e7863ff6d797e655d
If LOG_DIR environment variable is set, create report there.
If it is not set, create the report in 'a11y/'.
Bug: T301184
Change-Id: I74fa6e8ee59d43e5a6e57083999107ac27446a62
The report was created in log/a11y folder until now. Looks like Jenkins
expects it to be in log folder.
Bug: T301184
Change-Id: I9926e7446e4fc86526dab963066511bf8da2271e
When browser preference for reduced motion is enabled:
* Disables bolding of table of contents
* Disables sticky header transition
Bug: T254399
Change-Id: I8ef9e59b258fed977ce370da352b1924832d842b
Ideally the TOC has an h2 after the article title h1, but because of the TOC sticky positioning and its placement in the grid, we cant use an h2 without messing up the heading order.
Bug: T301051
Change-Id: I462ddfb618ddd422c9f71293280d1790c4846f50
Replaced “vector-menu-heading” with ‘’(empty string) in “$portletData['heading-class'] = '';”
in includes/SkinVector.php and updated accordingly the includes/templates/Menu.mustache file.
Updated “vector-menu-heading” with ‘’(empty string) in stories/LanguageButton.stories.data.js,
stories/UserLinks.stories.data.js, and tests/phpunit/integration/SkinVectorTest.php respectively
SkinVector and Storybook are cleaned up.
Bug: T290281
Change-Id: I4ca16953799b3dc52e45674bb398c78f14cfc842
This makes built assets available in jenkins,
Also addresses miscellaneous feedback relating to configuration
variable name and track topic name.
Change-Id: Ic70d517eba96aab6512a96ea19d6be56b3664871
It didn't vary by the skin version as documented in the Constants class,
and merely another instance of GlobalVarConfig.
Change-Id: I0acd0366a241e04bb79f6aae5dc52284dfa578df
* Bucket and sample on server by using the
`WikimediaEvents.WebABTestArticleIdFactory` service from
WikimediaEvents (soft dependency)
* Add linkHijack.js so that users bucketed in one group have the
possibility of remaining in that group if they click a link to another
page.
Bug: T302046
Depends-On: Ie6627de98effb3d37a3bedda5023d08af319837f
Change-Id: Iff231a976c473217b0fa4da1aa9a8d1c2a1a19f2
The LatestSkinVersionRequirement is problematic for various reasons:
1) It uses the "useskin" query string
parameter which may or may not refer to the correct
skin (in the case of a non-existent skin it will always come to the conclusion that
it is not modern Vector and thus must be legacy).
2) It uses the User object which may or may not be safeToLoad
depending on when called.
The feature seems redundant at this point, as we are separating code
into separate classes Vector and Vector22 and all the features only apply
to modern Vector. I suggest we remove it and use the features explicitly in the skin
intended.
Bug: T305232
Bug: T305262
Bug: T302627
Change-Id: I92fa33547bd601e05ddc8c1468e681892e47c16b
Can be merged when database rows have been updated to
no longer use skin version.
* Drop migration code
* Drop skin version preference
* Drop default skin version and existing accounts skin version
* Move skin definitions into skin.json
* Repurpose SkinVector as an abstract class
* Update READMEs
Bug: T301930
Bug: T294995
Bug: T302627
Change-Id: I7454d8f1cfdef81e7f3df476d8ce86736b46fff2
* Eliminates AB.js dependency on sticky header
* Code coverage has been raised to 100%
* Instead of importing ABTestConfig, these props are now passed into the
function along with a token.
* WikimediaEvents hook is now fired when experiment is initialized. The
experiment should not be initialized if it is not enabled.
* Removes several methods (e.g. initAB, getEnabledExperiment) due to the
preceeding changes.
* Adds `isInSample` and `isInTreatmentBucket` methods so that the client
has less work.
Treatment buckets now follow a naming convention so that the client can
do less work querying if the subject is part of the treatment:
* Treatment buckets should have the case-insensitive `treatment`
substring somewhere in their name (e.g. 'treatment',
'stickyHeaderTreatment', 'sticky-header-treatment' )
Bug: T302046
Change-Id: I4febec42b4c471b2f2ef02be2e334bd6d2c31eec
A user with an ID is a registered user and this test requires
an anonymous user.
Follow up to a1afa7ccb3
Not sure why this was working before, but I'm guessing
setId doesn't modify the User associated with the skin
Change-Id: I32bd74bd5aec1b14fb8b725fca2f8cef5f9d2ba1
Removed class "vector-menu-checkbox-expanded" and "vector-menu-checkbox-collapsed" in
includes/templates/Menu.mustache and deleted the necessary lines in
resources/common/components/MenuDropdown.less, tests/jest/stickyHeader.test.js,
includes/templates/skin.mustache, skin.json, i18n/en.json and i18n/qqq.json.
Bug: T299173
Change-Id: Ibf8a08e6e5d1a6c607abf170c030a0285e84ad74
Namespaceless class aliases are left behind for migration purposes.
They can be removed at a later date when dependant extensions
and skins are fully updated.
Bug: T301204
Change-Id: I2b37c1889ff862ec8bb41325fc9f654c673cd115
- Add test for SkinVector::shouldLanguageAlertBeInSidebar() method.
- Change some methods to protected status -- acknowledging this is not
an optimal change. Until we can pull out some of the language checks
into its own class with injected dependencies (title, feature manager,
languages), this is an interim solution to try to catch regressions
since the conditional states for showing the alert are complex and
ultimately temporary. Extracting the language checks into their own
class can be done in a follow up ticket to help optimize testing
language checks in isolation.
Bug: T302018
Change-Id: I99b7df3029e0af86e4d67b3f446d4ce99260d33e
Server render the table of contents in a collapsed state when the total
number of headings is equal or greater than the value of
`$wgVectorTableOfContentsCollapseAtCount`. Otherwise, the table of
contents will be server rendered in its "expanded" state.
In addition:
* Revise table of contents tests to call one `assertion` per element so
that it is easier to see the exact element that may fail an assertion.
* Revise table of contents tests to call a mount function that can merge
props to allow for a more flexible set of tests.
* Revise table of contents tests by wrapping a `describe` around tests
that expect the same prop state.
* Adds typedef for table of sections props
Bug: T300973
Depends-On: Ifaee451e1903f2accd0ada2f2ed6dfa3f83037b6
Change-Id: I382200bc603b6abf757a91f14a8a55a6581969bd
Collapses sub-sections in the new table of contents by default
(except for non-js and reduced-motion users) and expands the
sections when the top-level section link has been clicked.
Refactors the `activateSection` TableOfContents methods into separate
`activateSection` and `deactivateSection` functions.
Adds `expandSection` and `collapseSection` methods.
Adds triangle icon as a visual expand/collapsed indicator
next to all ToC section headings and are hidden via CSS based on
whether or not the section contains subsections.
Adds test for tableOfContents.
Bug: T299361
Change-Id: I36b3ae7f9f633877683bc17a9444c970d7fa7293
We want the link that the user has clicked inside the TOC to be "active"
(e.g. bolded) regardless of whether the browser's scroll position
corresponds to that section. Therefore, we need to temporarily ignore
section observer until the browser has finished scrolling to the section
(if needed).
However, because the scroll event happens asyncronously after the user
clicks on a link and may not even happen at all (e.g. the user has
scrolled all the way to the bottom and clicks a section that is already
in the viewport), determining when we should resume section observer is
a bit tricky.
Because a scroll event may not even be triggered after clicking the
link, we instead allow the browser to perform a maximum number of
repaints before resuming sectionObserver. Per T297614#7687656, Firefox
wasn't consistently activating the table of contents section that the
user clicked even after waiting 2 frames. After further investigation,
it sometimes waits up to 3 frames before painting the new scroll
position so we have that as the limit.
Bug: T297614
Change-Id: If3632529f58c15348a7200258f4f5999ea0dadc4
The behavior of data loading can differ between submitting and display,
so use FauxRequest to customize the method.
Also fix the order of parameters passing to assertSame(), the first one
should be the expected one.
Change-Id: Icfa062eada75c50cd2c8bc5db2930602d80e9ae7
Given Wikidata is the only project using modern Vector,
and the only project where the search API is not applicable,
this will result in a loss of autocomplete on Wikidata.org
which will fall back to the non-JS mode.
Bug: T290688
Change-Id: Iece5a4efd43e09cd90c842c9c134ca115b35f2b2
In I0cd49e6d621cd437e440ac7f7627eaa064ab870c a new field will be
added. Our tests in Vector shouldn't fail every time this happens
Change-Id: Ieb4923e9f58f950ee02ce3eb1446b982d1f5724a
Field name without 'wp' prefix (which is used as the key in html descriptor) is required.
After I58f9df384df8ecc5ebae8cac68ec2251351bc984, values of fields that are supposed to be disabled would be loaded from default, use a miss-matched field name would be treated as disabled.
It works in UI now, but it's not a good idea to strip the 'wp' prefix on the server-side.
Bug: T298819
Change-Id: If98368ad400986afaef3187867f201044ebf0efb
A new vector-2022 skin is added. This will be the eventual home
of the new Vector skin when we are ready to migrate.
Please see SkinVector class for the migration plan to simulate this
as part of testing.
Bug: T291098
Change-Id: Ibaddf94a5bfb5e21bbbaf1e0aa1b343a3f566d2d
The generation of JavaScript will throw a RuntimeException
making it obvious when an invalid A/B test has been setup.
Bug: T297662
Change-Id: I75b0e923463bf52f8fc5b5c6b7f9baf586053154
The main motivation is to reduce runtime by not invoking an unneeded
Selenium test phase in Quibble, which currently needs about 1 minute to
install npm dependencies.
Depends-On: Ieb7591f3c0ac843677a2a61b4ad47f920fc2ca2d
Bug: T255149
Change-Id: I0563421b3a3d3e6e81abec987c266ade275b47a1
Remove using of User:getOption since this method
will be hard-deprecated. Now it is soft-deprecated.
Bug: T296083
Change-Id: I3194a9c1c5c70592f88bc4dbedc78846d1141768
After I628435a4a, we were asserting a boolean was given because we're
extending HTMLFormField which requires a boolean value. This was safe
because GlobalPrefs would provide a boolean, but that changed with
I594f6297.
We could rework GlobalPrefs once again to ensure only a boolean is
passed in, but since HTMLLegacySkinVersionField already has special
handling around the data types, it seems to make sense to contain the
type transformation in this class.
Simply removing the Assertion is enough to prevent T296068, however
depending on when the global preference was saved (such as since MW
1.38.0-wmf.9 but before wmf.10), it's possible either a bool or a string
was saved, hence we check for both to ensure correct display.
Bug: T296068
Change-Id: If10b948617d2bb8346475f207fe425fb768cb987
- Separate icon classes from button classes in user links/language
- Upgrades the personal tools language button preference to
a mw-ui-button with icon
- Adds a generic selector for dropdown menus without an icon
- Cleans up user links CSS now mw-list-item class is available
- Removes icon hack CSS
Bug: T289630
Bug: T283757
Change-Id: Ib518858e06549f252d73d57fd4768f446cc561b9
Update/remove config, constants, hooks, templates, styles, logic, tests, stories to check legacy vs modern Vector where applicable instead of the decommissioned user links feature flag.
Bug: T288852
Change-Id: I5c5831091a10711838a8a2877c782df4996d4596
User::setOption() is deprecated and should be replaced with UserOptionsManager::setOption()
Bug: T277818
Change-Id: If867b4f97918db581d337a32b33cbca2315a71f6
- Add new OverridableConfigRequirement class.
- Add query parameter constant for user links.
- Update Feature Manager with new requirements.
- Use new class for LanguageInHeader requirement.
- Remove LanguageInHeaderTreatmentRequirement class and test.
- Add unit test to cover user links and language in header.
Bug: T285855
Change-Id: I56b729a9e245ed2ddc85625c0be39f5c26320ac4
Append mw-ui-icon classes to list item not list link
This allows us to apply a custom padding separate from the icon.
Note due to a bug in how core handles personal user items,
this will result in the icons temporarily disappearing for several
items until If399dfff9bbdd3b03b2ca702face3ec5164bef11 is resolved.
This is okay given the user menu is currently feature flagged.
Bug: T191021
Change-Id: I766aeb4d1bb36cebd0d80ad43ced940dbea96477
Since we have feature flagged the new user menu feature, it is
imperative we load both sets of styles until the feature has
shipped. This allows us to switch seamlessly between the two
without worrying about cached HTML being served with updated CSS.
To do this, we add a new class to both user menu's distinguishing
the legacy version from the modern version. The styles are then
scoped to these new selectors.
This also fixes some regressions with the legacy user menu in
modern Vector when wgVectorConsolidateUserLinks is disabled.
Notes:
* No caching selector is needed for #pt-userpage given it can only
ever be output for logged in users.
* ID selectors in general are bad, so scoping to mw-portlet-personal-user-menu-legacy
isolates the legacy component allowing it to be rendered alongside the modern UserMenu
Bug: T276561
Change-Id: I068c5233bb25a7b141e66a6726b5761841f83eb2
Before this commit the `languageinheader` query param would only take
effect if the A/B test was enabled AND the query param was set. Per
T282543, we want the query param to take effect regardless of the state
of the language/AB test config.
To see new treatment, set `languageinheader=1`.
To see old treatment, set `languageinheader=0`.
Bug: T282543
Change-Id: I6a06e90b6e46a6fd7506a5ddeaf071b893ebfe8e
* Adds ab test config to enable/disable the ab test. Defaults to `false`
(ab test disabled).
* Adds a `languageinheader` query param which only takes effect when the
ab test is enabled. The query param is cast to a bool and determines
which treatment is shown. For example, set query param to
`languageinheader=1` to see the new treatment. Set query param to
`languageinheader=0` to see the old treatment. To bucket based on the
user's id or global user's id, don't set the query param.
* Moves the language in header config work that was previously in
ServiceWiring into a `LanguageInHeaderTreatmentRequirement` class so
that unit tests can be done on most of the logic that determines whether
the language in header will show.
* Adds logic to bucket user based on [global] user id.
Bug: T280825
Change-Id: Id538fe6e09002fae6c371109769f3b7d61e7ac6d
The following rules are failing and were disabled:
* compat/compat
Additional changes:
* eslint: Renamed `wikimedia/client` profile to `client-es5` (T277085).
Change-Id: I12c1a88cef8e2c95bed496628d2fe74d031f8278
* We add the `.mw-interlanguage-selector` class to the
.vector-menu-heading in the server rendered HTML. `ext.uls.interface.js`
later attaches a click handler to this selector that loads the rest of
ULS.
* We hide the dropdown arrow for js users and only show it again if
ext.uls.interface module isn't installed or is not being loaded.
* When the `ext.uls.interface` module has been loaded, we hide the checkbox
and checkbox hack menu in favor of showing the ULS popover.
Additionally:
* Adds '.vector-menu-heading' class to menu headings.
* Change h3 selector to `.vector-menu-heading`.
Bug: T273232
Change-Id: I6f4572c16ca4096dcda3aac4d585003b93dcccfa
FeatureManager allows the logic to be centralized and allows clients to
ask about its state. For instance, SkinVector will make use of it in
I70277c1082a504fbd5f6023e9873e8071de7e35d.
Also:
* Adds WvuiSearchTreatmentRequirementTest to test A/B logic
WvuiSearchTreatmentRequirement/Test logic are adapted from
I878239a85ffbecb5e78d73aed5568c56dbd7d659.
Bug: T270202
Change-Id: Ia02349a7b41c7caf26fbd728e0be7d47488b97e5
SkinMustache in core provides most of what is required for Vector to
generate its menus. In the interest of having a canonical source of
truth for menus across all skins, Vector should use this data.
To ensure the HTML generated is (mostly) the same after this patch to
prior, a few modifications are necessary:
* The data from core is decorated so that Vector can continue having its
own custom class names on menus. This is done using the
decoratePortletClass method.
* There is no support for a menu having a header representing the
selected menu item, as is currently the case with variants. This is
achieved via an extension to getPortletData. It's assumed that later
when variants are merged with languages, this can be removed.
* Menus are agnostic to how they are displayed, so we must continue to
add the is-dropdown template variable to drop down menus. In future we
may want to rethink our Menu partial to make this unnecessary in PHP.
* The portal-first class is redundant in the modern Vector as we can
use the first-child selector. Previously we introduced a class to
service the legacy skin where this rule doesn't apply as #p-logo is
the first child. However, the legacy skin can do this using a special
next sibling selector instead.
Bug: T268157
Change-Id: I5f7adc1840441b508ffee40139b85b64021789e6
Drop support for vectorMenu, vectorTabs and
vectorMenuCheckbox, body, menu selectors in preference
for standard selectors.
This change will impact a large amount of user scripts/styles but should
not impact any gadgets.
These classes were kept around for user scripts and styles however are not
needed internally. As we transition to a more maintainable skin menu
system, it is time to lose these selectors even though this will cause
disruption.
Vector now will use the mw-portlet class rather than the vector-menu
class in its own CSS styling, however it keeps the other classes to
allow differentiation of the different types of menu.
Changes to test: Previously the tests assumed all portlets were empty
when checking the classes. This is very rare, so its better to check
the classes of non-empty portlets, so several tests are updated
accordingly to drop the emptyPortlet class.
Bug: T262092
Change-Id: I1824335eb47d613c2a4804ec1f1106c0f4c16101
Kept as simple as possible for now. The new class is added but no classes
are removed. This will be done in a follow up.
Bug: T256897
Bug: T253938
Change-Id: Ib31a9d8f2ac14e63b63e82abd4a9aa1fcb956f45
Since SkinVector provides name via skin.json the name must be
passed in the test constructor.
This will be required as part of
I5772eb760e4fc56d2062a333ba4d7ca6995f3db2
Change-Id: I4087deb8b0726c9959ac15d77a0ed2442e4890f6
A new config flag is added that buckets 50% of users into the old
header and the 50% into the new header.
Bug: T249363
Change-Id: I8b4fa475f9cd7e61ad2989e2a1485e7e64c8ab3f
I177dad88 introduced the skin version user preference field and
associated configuration values. Per T242381, the field is to presented
as a checkbox with the implied storage type of a boolean where a string
is needed. A PreferencesFormPreSave hook handler was added to adapt
values of either data type to the other. While this was a neat solution
to a minor nit, the adapter's implementation is incompatible with the
GlobalPreferences extension as the PreferencesFormPreSave hook isn't run
whilst saving global preferences.
Rather than adding an equivalent hook to the GlobalPreferences
extension, create a custom field based on a checkbox with the adapter
included. This allows us to:
- Separate the business logic concerned with preserving the user's
VectorSkinVersion preference if they've simply disabled Vector from
the adapter
- Simplify the adapter's implementation
- Forego adding hooks to the GlobalPreferences codebase
Additional changes:
- Replace repeated string literals with equivalent constants in
tests/phpunit/integration/VectorHooksTest.php
Bug: T258493
Change-Id: I628435a4ad676f55534191b8c10147be28be5d73
This hook is run on every page. The SkinTemplateNavigation hook
counter intutively is run only on pages which can exist. I think
it's clearer if we only use SkinTemplateNavigation::Universal hook
and keep the logic for when it runs inside our own code.
Bug: T255319
Change-Id: I0835074a6cadf6e9bdcc45299de37dd9328bf9b2
The feature manager abstracts away how a feature is enabled from the
consumer of that feature. Accordingly, replace direct instantiation of
SkinVersionLookup with usage of the Vector.FeatureManager service.
Supporting changes:
- Add Vector\VectorServices, a simple wrapper around
MediaWiki\MediaWikiServices that allows us to both document and
type-hint services specific to Vector
- Add Vector\Hooks::isSkinVersionLegacy to minimise repetition
Additional changes:
- Make the MakeGlobalVariablesScript hook handler return early if the
user isn't using the Vector skin like the other hook handlers
Bug: T256100
Change-Id: I93b5ef39802323c7ac658af8fa7cc312fff68aa7
In PHP we add collapsible classes to all elements except watchstar
so that certain tabs can be collapsed under the more menu in JS.
This adds unnecessary complexity to our codebase and is not used
if JS is disabled.
To simplify this and bring Vector's PHP consistency with core this
logic is moved to JavaScript.
Bug: T259372
Change-Id: I2acbf7089198118626368ee8a37615d2de062f83
Please note I7e06a4cc226f3434c0f655212a464b8b98bcc7f4 should be
merged at the same time as this patch.
== The background ==
All extensions have been weaned of BaseTemplate hooks in
Wikimedia projects.
This change now means that Vector will no longer run
any BaseTemplate hooks. See the epic T253809 for the
implementation details.
== The change ==
BaseTemplate will now have nothing to do with the rendering of
Vector. The skin version is added to express the significance of
breaking compatibility with 3rd party extensions.
We TEMPORARILY remove SkinVector to retain git blame. SkinTemplateVector will
be renamed SkinVector in the follow up (see 2/2)
Update skin.json to use SkinTemplateVector for the skin (this will be fixed
in a follow up).
The isLegacy method is moved to SkinTemplateVector.
Changes of note:
* html-debuglog is no longer needed. SkinMustache includes this information on
the skins behalf
* html-printtail and html-headelement are now not needed in the master template
and added by SkinMustache
* Skin::getAfterPortlet does not provide the `after-portlet` wrapping element provided
by BaseTemplate::getAfterPortlet so this is added
* SkinTemplate::getFooterIcons does not support the options that BaseTemplate::getFooterIcons
does so any icons which do not have an image must be manually checked for and unset
Known changes to HTML output as a result of intentionally
delegating their output to the core SkinMustache class:
* A new line is removed between the body element and #mw-page-base
* #mw-html-debug-log now appears at the end of the body element
* #printfooter is now a child of #mw-content-text rather than sibling.
Bug: T251212
Change-Id: I4e89beb96f6401ed7e51bafdf0aac408f5a2c42f
- Creates a new user-preference called 'VectorSidebarVisible'
which stores the sidebar hidden/collapsed state for logged-in
users.
- Updates that user-preference on the client whenever the sidebar
is expanded or collapsed.
- Refactors the sidebar related javascript into a separate file.
Bug: T255727
Change-Id: Ib1ce934f3646cd8feebf0d3b15c38b5b969ec957
Adds a bundlesize test that measures the bytesize of
individual ResourceLoader modules.
This test depends on a running mediawiki instance as well as
the $MW_SERVER and $MW_SCRIPT_PATH environment variables.
ResourceLoader modules are fetched from a URL and piped
into a bundlesize command based on a custom configuration file.
The test can be run with `npm run test:size`.
Bug: T244276
Change-Id: I4f4534921ccbc6bd4f99229c8dd36b1279dde640
Begin our journey away from BaseTemplate by
moving VectorTemplate code to SkinVector. In future all
methods will live here but to lower risk, I've only targetted
the get method.
Bug: T251212
Change-Id: I58c2ff5edaacc2d5e45492c121cf0f87d08b623f
The .menu class historically only needs to apply to dropdowns.
the .vectorMenuCheckbox is inconsistent with the other classes on the
menu so we should begin its deprecation.
Bug: T253329
Change-Id: I00b4d2fd795195cd9c8add650a3b3cafdced5465
Use SkinTemplateNavigation hook instead and copy the collapsible
behavior to the menu function
The code inside getSkinData that checks VectorUseIconWatch is
redundant as it duplicates checks already inside
SkinTemplate::buildContentNavigationUrls
It is enough to simplify check whether watch or unwatch is
present in the array.
Bug: T251212
Change-Id: If6b10b0ddcbd4b21dd13a2813e60b604c3a23415
* Standardise the menu markup. This means all menus in Vector will now
be wrapped in a div and will have a heading.
* All menus now have the vector-menu class. Styles specific to personal tools
are moved to layout since these are concerned with placement.
* The ul class will always have menu class.
* emptyPortal class is generalised into vector-menu-empty for consistency
with other classes and moved from common.less into Menu.less
* Standardise hooks - BaseTemplateAfterPortlet can now be run on any
menu.
Changes to HTML:
* lang and dir attributes are moved from the h3 up to the div element
.vectorTabs, .portal(s) and #p-personal now has hidden span element inside h3
* for non portals ul.menu" is now wrapped in a div.vector-menu-content
This change does impact the following CSS selectors which will need to be updated:
I see no matches for these selectors in code search.
```
#p-variants > ul
#p-namespaces > ul
#p-personal > ul
#p-views > ul
#p-cactions > ul
```
Using global-search.toolforge.org I see one match
for p-variants, 26 for p-namespaces, 30 for p-personal,
36 for p-views and 7 for p-cactions. I see this as acceptable
breakage provided a user notice is sent out which it has been
(T252447)
Bug: T249372
Change-Id: Id59234aa6b822a24848386bdc04d8d7ed37ca145
To complete the refactor, the Portal is also refactored
as a Menu using the getMenu function.
An old code path supporting portals outputted by hooks with
strings is marked as deprecated to simplify this code in future.
array-portals-first -> data-portals-first (the value is not
an array)
Changes:
* $this->getLanguages and $this->getToolbox() always returns an array (see BaseTemplate)
but we previously supported portals made using raw HTML. Let's move away from that
behaviour and deprecate it.
* Hooks are moved into buildSidebarProps and marked as deprecated where possible
(SkinTemplateToolboxEnd). SidebarBeforeOutput can be used instead.
Bug: T249372
Change-Id: I2549af3e24e5d51c09e9a88ca50a0d9b2e154c3f
The classes were recently changed so provided this is merged before
next branch cut no need to worry about cached HTML.
Bug: T249073
Change-Id: Ib20c7a359bda858df89ebb245e682d321dd5acd0
the PersonalMenu should be generalised. In future we will use it as
the template for all menus
Bug: T249372
Change-Id: Id1c43d2e9eefef1d7aec45f0137e27f10ad935df
VectorTemplate relies on many of SkinVector's methods. To future
proof these tests we need to set the skin in the mocked object.
Bug: T251212
Change-Id: Ifd9bbc9c909626ecfe8ccd085673bc777423d560
de76ab5 added the config,
`$wgVectorDefaultSkinVersionForExistingAccounts`. Its usage in
`Hooks::onUserGetDefaultOptions()` was invoked not only for existing
accounts but anonymous users _as well._ This is a bug, due to my own
misconceptions about the hook, that went against both the config's name
and its documentation.
Unfortunately, user sessions are unavailable in
`Hooks::onUserGetDefaultOptions()` so it does not seem to be possible to
determine whether the active user is an anonymous or existing account.
This patch drops the hook and centralizes all version determination
logic in SkinVersionLookup::getVersion(). SkinVersionLookup requires a
the active User object and can make the anonymous / existing account
determination by checking login state.
The issued was identified while responding to review feedback given by
@polishdeveloper / @pmiazga in
I52d80942b4270c008d4e45050589ed9220255a50.
Bug: T251415
Change-Id: I7982b4c34283ba81d0232ee6f501c44cf0a74b98
FeatureManager::registerRequirement registering an instance of
SimpleRequirement with ::registerComplexRequirement was awkward.
Changes:
* Rename FeatureManager::registerRequirement to
::registerSimpleRequirement, which is exactly what it does!
* Rename FeatureManager::registerComplexRequirement to
::registerRequirement
Bug: T244481
Change-Id: I612af959cee4cdcd0bdcda51a81b86ed61ee2e16
Changes:
- Add the LatestSkinVersionRequirement requirement class, which lazily
evaluates the application state to get the version of the skin for the
request.
This majority of this class is taken from Stephen Niedzielski's
Vector\SkinVersionLookup class, which was introduced in d1072d0fdf.
- Register an instance of LatestSkinVersionRequirement and register the
'LatestSkin' feature that requires that requirement
- Re-introduce SkinVector::isLegacy and make it defer to the Feature
Manager
Bug: T244481
Change-Id: If6b82a514aa5afce73e571abdd8de60b16a62fa8
As described in the readme but not implemented until now, this patch
enables the skin version to be specified as a URL query parameter. This
is useful for testing both skin versions during development and on wiki,
as well as enabling sharing URLs with a specific skin (Vector) and skin
version (1 or 2).
Obtaining the actual skin version requires tying together three input
sources, WebRequest, User, and Config. It seems simple but it'd be easy
to botch. For this reason, a helper class to correctly interrogate them
and tests are provided.
Bug: T244481
Change-Id: I52d80942b4270c008d4e45050589ed9220255a50
This commit is singularly focused on adding a link to the sidebar for
Vector, logged-in users. It does the bare minimum to fulfill the
requirements of T243281.
Additionally, it will help to answer the question "Do we need to use
abstractions (other than maybe different templates) to separate Legacy
Vector from Vector" by intentionally leaving out any abstractions in
order to make it easier to compare with a follow-up patch
(Ib2ef15180df73360cc1de25b893e49d415d23e1a) which does use abstractions.
It is a good thing to question whether or not we need addtional
abstractions in VectorTemplate and if they will help us as unnecessary
abstractions can have the opposite effect and just lead to further
frustrations down the road.
Therefore, I urge you, the reviewer, to let me know your thoughts! If
abstractions are viewed as not making our lives any easier, the
follow-up patches may be completely discarded and that's totally okay
with me. :) I think it's a good think to talk about now though.
Important changes:
* The VectorTemplate constructor was changed to allow injecting the
config, templateParser, and isLegacy boolean (only the config was
allowed before this commit). According to MediaWiki's Stable Interface
Policy, "Constructor signatures are generally considered unstable unless
explicitly declared stable for calling" [3]. Given that VecorTemplate's
constructor is not marked as stable, it is justified to do this without
warning according to the policy.
* Due to the above, the 'setTemplate' method is no longer needed and was
marked as deprecated.
* VectorTemplateTest was made to adapt to the new VectorTemplate
constructor. Additionally, it now extends from
MediaWikiIntegrationTestCase which my intelliphense server can pick up.
I *think* MediaWikiTestCase is just an alias to
MediaWikiIntegrationTestCase [1] and MediaWikiTestCase file was renamed
to MediaWikiIntegrationTestCase in [2], but I'm willing to change it
back if there is pushback to this.
Open questions:
* What are VectorTemplate's responsibilities? To me, it acts right now
as a controller (because it echos the full HTML string from the
template), a model (because SkinTemplate::prepareQuickTemplate sets data
on it which it later retrieves through `$this->get()`), a presenter
(because it adds data tailored for a web-centric view), and a view
(because it renders HTML strings instead of letting the view/template be
solely responsible for that). Arguably, some business logic might be
mixed in there as well (because it checks to see if a User is logged
in/has necessary permissions to show x which my changes here add to).
This might not be a problem if we keep VectorTemplate relatively small,
but will it remain this way as we progress further in Desktop
Improvements?
* How do we write tests for VectorTemplate without exposing unnecessary
public methods? For example, if I want to test the `getSkinData()`
method to see what state will be sent to the template, how should I do
this? One option might be to use `TestingAccessWrapper` to expose these
private methods which is what
`VectorTemplateTest::testbuildViewsProps()` does. Another option is to
accept this method as public. Is there a better way? Keep in mind that
even with access to this method, there might be many things to mock.
[1] 0030cb525b/tests/common/TestsAutoLoader.php (L64)
[2] Ie717b0ecf4fcfd089d46248f14853c80b7ef4a76
[3] https://www.mediawiki.org/wiki/Stable_interface_policy
Bug: T243281
Change-Id: I0571b041bcd7f19bec9f103fa7bccdd093f6394d
Address some feedback from I7a2cdc2dfdf20d78e4548f07cf53994563b234b3:
- Miscellaneous documentation improvements.
- Add a false case test to `DynamicConfigRequirement->isMet()`.
Bug: T244481
Change-Id: Ic5637f42da755f871c5a6d545e14effd3ac8c670
We expect the vast majority of requirements and features to be defined
in services as possible. However, there are some "complex" requirements
that require additional application/HTTP request state. Unfortunately,
service wiring is done before some of that state is available.
I65702426 attempted to work around this by requiring clients of the
Feature Manager to pass that additional state on every interaction with
the system. Those complex requirements would then select the parts of
the state that they required when it was required. However
implementations of \IContextSource are God objects and their use should
be limited.
Whilst reviewing I65702426, Stephen Niedzielski mentioned that the
application state being available is a requirement. This remarkably
simple solution:
- Keeps the Requirement interface and FeatureManager API free of God
objects;
- Is true to the nature of the Feature Manager - it makes clear and
centralizes the various checks for application state being available
across the codebase; and
- Inject a Requirement implementations' dependencies at construction
time
It just so happens that the $wgFullyInitialised variable flags whether
the application state is available...
Changes:
- Add the the FeatureManager\Requirements\DynamicConfigRequirement class
and tests. The DynamicConfigRequirement lazily evaluates a single
configuration value whenever ::isMet is invoked
- Register an DynamicConfigRequirement instance, configured to evaluate
$wgFullyInitialised while constructing the Vector.FeatureManager
service
Bug: T244481
Change-Id: I7a2cdc2dfdf20d78e4548f07cf53994563b234b3
