Automatic update from web-platform-tests
[css-font-loading] Add a test for FontFaceSet.has() (#30322)
* [css-font-loading] Add a test for FontFaceSet.has()
* syntax
Co-authored-by: Tab Atkins Jr <jackalmage@gmail.com>
--
wpt-commits: 1b222dc282e3c4f0e4647b3b6aa2a0eab06b393c
wpt-pr: 30322
Automatic update from web-platform-tests
binding: Fix an infinite loop issue at window.opener attribute set
Fixes an infinite loop at window.opener attribute set. The loop
happened because window.opener attribute set used property
assignment instead of defining a property. This patch defines
a property appropriately.
Bug: 1237774
Change-Id: Ibcb6450f5cc7757e49222fcb03d2d8e489c85def
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/3139739
Reviewed-by: Kenichi Ishibashi <bashi@chromium.org>
Reviewed-by: Kentaro Hara <haraken@chromium.org>
Commit-Queue: Yuki Shiino <yukishiino@chromium.org>
Cr-Commit-Position: refs/heads/main@{#918039}
--
wpt-commits: 7be37a1d311c99915ddd1a5ea202f094dc2a124c
wpt-pr: 30325
Automatic update from web-platform-tests
[ResourceTiming]: Update cached-resources tests
This change updates the resource_reuse.sub.html and
single-entry-per-resource.html Resource Timing WPTs. They covered the
same case; when no network request is issued for a resource, there
should be no associated PerformanceResourceTiming entry. The tests have
been consolidated into cached-image-gets-single-entry.html and the style
updated to conform to the guidelines in
wpt/resource-timing/CodingConventions.md.
GithubIssue: https://github.com/w3c/resource-timing/issues/254
Bug: 1171767
Change-Id: Ibdc67dcedf70b96e317ace165eed4e5ec897e996
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/3138444
Reviewed-by: Yoav Weiss <yoavweiss@chromium.org>
Commit-Queue: Tom McKee <tommckee@chromium.org>
Cr-Commit-Position: refs/heads/main@{#918104}
--
wpt-commits: 24836df509f65b6c2fb896dae46d99ec1b74d4e8
wpt-pr: 30314
Automatic update from web-platform-tests
Correct postMessage() test
This test was passing but no assert was being run.
Fixes#29536.
--
wpt-commits: e1be3c74b86007b8a6118392040a9c02b51ad952
wpt-pr: 29987
Automatic update from web-platform-tests
Fix the remaining instances of mixed-case `WebIDL`
The `WebIDL` directory was renamed to lowercase `webidl` in #30327.
This change fixes two remaining instances of mixed-case `WebIDL` in
`update-built-tests.sh` and `tools/ci/jobs.py`.
--
wpt-commits: 8300a95a270ed8c9c4685b906d927c568559deef
wpt-pr: 30331
Automatic update from web-platform-tests
Align Web IDL directory name with shortname
--
wpt-commits: 9f3fa5056ae29bbd4cb668aa87b8559c9f4500f5
wpt-pr: 30327
Automatic update from web-platform-tests
EventSource: use .any.js and .window.js more
--
wpt-commits: 00e8611b23fed04f014e882c12980cafc77e2c1b
wpt-pr: 29765
Automatic update from web-platform-tests
[mathml] Embellished operators and stretch properties
This is part patch of a series of CL to implement MathML stretchy
operators as well as embellished operators [1]. It handles the cases
"If there is a block stretch size constraint or an inline stretch size
constraint then the element being laid out is an embellished operator"
from [2] and [3].
A typical use case is the following, the inner <mover> element is an
embellished operator (a over brace with the " k times" as over script).
Its core operator should stretch horizontally to cover the inline size
of the <mrow> element "x + x + … + x". Without that CL, the horizontal
over brace would just cover the size of the " k times" over script.
<math>
<mover>
<mrow>
<mi>x</mi>
<mo>+</mo>
<mi>x</mi>
<mo>+</mo>
<mo>…</mo>
<mo>+</mo>
<mi>x</mi>
</mrow>
<mover>
<mo>⏞</mo>
<mrow>
<mi>k</mi><mtext> times</mtext>
</mrow>
</mover>
</mover>
</math>
The outer <mover> passes inline stretch constraint to its over script
per [3]. Now using the rule from [3] implemented in this CL, the inner
<mover> will just use that inline stretch constraint to stretch the
over brace.
Note, that elements can receive a stretch constraint orthogonal to the
direction they normally stretch their children. In the example below,
the <munder> is an embellished operator (a left parenthesis with a "1"
under script). Although it normally stretches its children horizontally,
the parenthesis is stretched vertically using the stretch constraint
received from its parent <mrow>.
<math>
<mrow>
<munder>
<mo>(</mo>
<mn>1</mn>
</munder>
<munderover>
<mn>2</mn>
<mn>3</mn>
<mn>4</mn>
</munderover>
</mrow>
</math>
New tests integrated into this CL cover all possible combinations:
- third_party/stretch-along-block-axis-001.html is extended to test
case when the <mo> is replaced with an embellished operator with
<mrow> or <munderover>. It tests when the mrow and underover
algorithms forward any inherited block stretch size constraint to
their children. It also tests when the mrow algorithm calculates
and passes a block stretch size constraint to embellished operator
children.
- presentation-markup/scripts/underover-stretchy-00[23].html are copies
of underover-stretchy-001 with the <mo> embellished with <mrow> and
<munderover> respectively. They respectively test when the mrow and
underover algorithms forward any inherited inline stretch size
constraint to their children. They also test when the munderover
algorithm calculates and passes an inline stretch size constraint to
embellished operator children.
[1] https://w3c.github.io/mathml-core/#embellished-operators
[2] https://w3c.github.io/mathml-core/#algorithm-for-stretching-operators-along-the-block-axis
[3] https://w3c.github.io/mathml-core/#algorithm-for-stretching-operators-along-the-inline-axis
Bug: 6606, 1124298, 1124301
Change-Id: I682e5343c1440e657b10933e1bfe7fdf63d9e112
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/3121110
Commit-Queue: Frédéric Wang <fwang@igalia.com>
Reviewed-by: Ian Kilpatrick <ikilpatrick@chromium.org>
Cr-Commit-Position: refs/heads/main@{#918027}
--
wpt-commits: 71597bb9442e9433049076ef2474a3b6a87f81c9
wpt-pr: 30193
Automatic update from web-platform-tests
[wave] Include all js tests in a wave run
For some unknown reason the wave loader skipped the first
test that was listed under js tests. This change makes the
wave runner run as many tests as the online web based wpt
runner.
--
wpt-commits: fbd337d3e51475927889f19cc2db72ed41200140
wpt-pr: 30251
Automatic update from web-platform-tests
[wave] Python 3 fixes for tools/wave/
A bunch of minor changes to make the wave test runner work with
Python 3:
* Sort() no longer accepts a cmp function argument, use sorted()
instead with a key function.
* headers.get('Content-Length') returns byte value now, explicitly cast
to int. Also return -1 if the header does not exist (instead of None)
so that it is accepted as int value and to get the same behavior in
InputFile.read() as with Python 2.
* Pass header values as comma separated byte strings to
HTTPConnection.request() instead of lists. HTTPConnection.putheader()
does not like getting lists.
* hashlib.update() accepts a "bytes-like" object so encode the token
string before passing it in.
* dict.iteritems() was renamed to dict.items().
* Explicitly open zipfiles as binary files. If not, file.read() will
try to read them as utf-8 which fails.
--
wpt-commits: 8bd6281eea3983b9ced7363dc3625eb04bee271f
wpt-pr: 30228
Automatic update from web-platform-tests
[wave] Support calling window.gc() between tests in the wave runner
This patch make the wave runner call window.gc() if the Chromium
commandline flag --js-flags="--expose-gc" is supplied.
Garbage collecting between test groups provides much more stable
test results when running tests in a single window. Especially on
slow machines where garbage collection might not kick in as
frequently.
--
wpt-commits: 21868a414f0555086247b730f742d933b5530c32
wpt-pr: 30253
Automatic update from web-platform-tests
Fix and make automatic sandboxed plugins test
Follows https://github.com/whatwg/html/pull/6946.
--
wpt-commits: 789f310832f7746b34bdc6afa16ff1e1a7a8f4af
wpt-pr: 29951
Automatic update from web-platform-tests
SharedWorker: interface constructors exposed (#29988)
As per https://html.spec.whatwg.org/#shared-workers-and-the-sharedworker-interface, SharedWorker should not be exposed. The other constructors, however, should not be missing.
Fixes#29928.
--
wpt-commits: 4606293de6223e69ababd4aa0a1c4a0bbe888666
wpt-pr: 29988
Automatic update from web-platform-tests
Revert "Integrated Progress based animations with CSS Animations"
This reverts commit 149290ad09de842c169957a8a773ba9b07e75b50.
Reason for revert: Speculative revert for consistent failure of
at-scroll-timeline-dynamic.tentative.html on Linux Tests (dbg)(1).
Blamelist of the first failure: https://ci.chromium.org/ui/p/chromium/builders/ci/Linux%20Tests%20(dbg)(1)/98587/blamelist
Original change's description:
> Integrated Progress based animations with CSS Animations
>
> Timeline duration can now be stored on an AnimationEffect.
> This was needed since for CSS animations, InertEffects with
> no owning animation are used until later in the pipeline.
> Normally the owning animation of an effect is used to gain
> access to the corresponding timeline which provides the timeline
> duration needed for timing normalization math, but in some
> cases the duration was needed before an effect had an owner
> assigned. As such it is now possible to create an InertEffect
> and pass in a timeline duration if needed.
>
> Added logic for when CSS animations swap from progress based
> timeline to a time based timeline. This new logic preserves
> the current progress of the animation even when current times
> do not match up.
>
> I also removed time range from some CSS tests since they are
> now supported by progress based animations.
>
> Bug: 1216655, 1140602
> Change-Id: I7dde1f117968c6465eb7c4c0f4b8b66d9371776f
> Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/3130703
> Reviewed-by: Kevin Ellis <kevers@chromium.org>
> Reviewed-by: Anders Hartvoll Ruud <andruud@chromium.org>
> Commit-Queue: Jordan Taylor <jortaylo@microsoft.com>
> Cr-Commit-Position: refs/heads/main@{#917801}
Bug: 1216655, 1140602
Change-Id: I5d993ec0ee6de610347916f8b1fd09d0ea10e34f
No-Presubmit: true
No-Tree-Checks: true
No-Try: true
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/3140514
Bot-Commit: Rubber Stamper <rubber-stamper@appspot.gserviceaccount.com>
Owners-Override: Yuki Yamada <yukiy@chromium.org>
Auto-Submit: Yuki Yamada <yukiy@chromium.org>
Commit-Queue: Yuki Yamada <yukiy@chromium.org>
Cr-Commit-Position: refs/heads/main@{#917986}
--
wpt-commits: 5716a8027bea38107c117d2f6a5ecf2b09ea88de
wpt-pr: 30323
Automatic update from web-platform-tests
Clamp the infinity parsed value from html_dimension
The DCHECK(CSSCalcInfinityAndNaNEnabled() || std::isfinite(num)) in CSSNumericLiteralValue::Create could be failure when html_dimension parse the large number string.
Therefore, add a clamping for html_dimension to prevent infinity CSSNumericLiteralValue creation.
Bug: 1243557
Change-Id: I24a0736d1641a088a5e9f21f3ce4a043d805fa94
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/3121744
Reviewed-by: Mason Freed <masonf@chromium.org>
Reviewed-by: Xiaocheng Hu <xiaochengh@chromium.org>
Commit-Queue: Seokho Song <0xdevssh@gmail.com>
Cr-Commit-Position: refs/heads/main@{#917971}
--
wpt-commits: 8c780b2dbff62b7be32828f8792aadd389ab49a0
wpt-pr: 30220
Automatic update from web-platform-tests
[WebCodecs] Prevent deeply nested frames.
It's not correct to directly access the wrapped frame since it may
itself be wrapping another frame. Prevent nesting, but also switch
call sites to not operate directly on |wrapped_frame_| for ease of
reasoning.
R=sandersd
Fixed: 1246173
Change-Id: I1475aa695a1801620eebdcb88a0d421652696261
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/3140495
Commit-Queue: Dale Curtis <dalecurtis@chromium.org>
Commit-Queue: Dan Sanders <sandersd@chromium.org>
Auto-Submit: Dale Curtis <dalecurtis@chromium.org>
Reviewed-by: Dan Sanders <sandersd@chromium.org>
Cr-Commit-Position: refs/heads/main@{#917938}
--
wpt-commits: a607f8c0330f6d6ce92fac257bc47a1fd7fd8706
wpt-pr: 30319
Automatic update from web-platform-tests
Skip relayout on scrollbar changes if side-effects aren't allowed.
The root cause here is that legacy multicol containers don't handle
overflow correctly, but that's not easily fixable.
Instead, just disallow additional layout passes if scrollbars (somewhere
in the subtree) change when side-effects are disabled
(NGDisableSideEffectsScope::IsDisabled()). Calling ClearLayoutResults()
(which is what the scrollbar stabilization loop does) while in this
state is forbidden.
Bug: 1245440
Change-Id: I5b959cf1cfca9bb65b7768e6abfae47155d2038d
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/3135415
Commit-Queue: Morten Stenshorne <mstensho@chromium.org>
Reviewed-by: Christian Biesinger <cbiesinger@chromium.org>
Reviewed-by: Koji Ishii <kojii@chromium.org>
Reviewed-by: Ian Kilpatrick <ikilpatrick@chromium.org>
Cr-Commit-Position: refs/heads/main@{#917877}
--
wpt-commits: 51d8c24e16deaf156a7eab496cf06721e55c4ebf
wpt-pr: 30285
Automatic update from web-platform-tests
FS: Add check for removal of sandbox root dir
The FileSystemHandle::Remove() method was recently added to allow for
the removal of files and directories from the underlying file system.
This method now fails when attempting to delete the root of a sandbox
file system.
Bug: 1242231
Change-Id: Ibe3ed59b79678ce168d0803c6607abd8ac06594c
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/3123735
Commit-Queue: Austin Sullivan <asully@chromium.org>
Reviewed-by: Marijn Kruisselbrink <mek@chromium.org>
Cr-Commit-Position: refs/heads/main@{#917845}
--
wpt-commits: ea05faa7f5b21494fbf0b840267d8ccc3e3229d9
wpt-pr: 30261
Automatic update from web-platform-tests
[ResourceTiming]: Move more tests to cross-origin-redirects.html
As part of the work tracked in crbug.com/1171767, we modernized some
Resource Timing WPTs concerned with cross-origin behaviour. See
crrev.com/865600 for context.
This changes modernizes and centralizes some more of the test cases that
exercise cross-origin redirects.
Bug: 1171767
Change-Id: If56793f09b8953c5bccd0e0357c24bc6b6392711
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/3129164
Reviewed-by: Yoav Weiss <yoavweiss@chromium.org>
Commit-Queue: Tom McKee <tommckee@chromium.org>
Cr-Commit-Position: refs/heads/main@{#917827}
--
wpt-commits: 6699804bd6f7d5ed3f437c0d616c79f0dddf6f74
wpt-pr: 30234
Automatic update from web-platform-tests
Integrated Progress based animations with CSS Animations
Timeline duration can now be stored on an AnimationEffect.
This was needed since for CSS animations, InertEffects with
no owning animation are used until later in the pipeline.
Normally the owning animation of an effect is used to gain
access to the corresponding timeline which provides the timeline
duration needed for timing normalization math, but in some
cases the duration was needed before an effect had an owner
assigned. As such it is now possible to create an InertEffect
and pass in a timeline duration if needed.
Added logic for when CSS animations swap from progress based
timeline to a time based timeline. This new logic preserves
the current progress of the animation even when current times
do not match up.
I also removed time range from some CSS tests since they are
now supported by progress based animations.
Bug: 1216655, 1140602
Change-Id: I7dde1f117968c6465eb7c4c0f4b8b66d9371776f
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/3130703
Reviewed-by: Kevin Ellis <kevers@chromium.org>
Reviewed-by: Anders Hartvoll Ruud <andruud@chromium.org>
Commit-Queue: Jordan Taylor <jortaylo@microsoft.com>
Cr-Commit-Position: refs/heads/main@{#917801}
--
wpt-commits: 9e3b33267d5f983f77bf9600a77dac54a5a37787
wpt-pr: 30247
Automatic update from web-platform-tests
[layout] Remove limit from LayoutInline::SplitInlines.
After 200 elements the code "gave up" causing the layout tree to be
"strange".
This caused a To<LayoutInline> to fail in the OOF code. Relaxing this
To<> to a DynamicTo<> caused additional CHECKs / DCHECKs all over the
place (not just in NG but in Legacy as well).
This patch removes the limit at which we "give up". This may cause
additional render hangs.
However we currently have a project "block-in-inline" which will (for
most cases) stop inline-splitting for occuring (except in legacy
fallback).
Bug: 1245786
Change-Id: I5f1c4d6a4b81a8345974de40c0c50a27a839b7b4
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/3140144
Reviewed-by: Koji Ishii <kojii@chromium.org>
Commit-Queue: Ian Kilpatrick <ikilpatrick@chromium.org>
Cr-Commit-Position: refs/heads/main@{#917771}
--
wpt-commits: 37e4ca8ce64512cb1da8159c3a6a757bad604c43
wpt-pr: 30306
Automatic update from web-platform-tests
Fix delete_all_cookies documentation
This function (like the others) returns a promise.
We should make that explicit in the docs.
Bug: 1233305
Change-Id: I90deeb13988267dc2b6069a22e44de82e8b09015
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/3140046
Auto-Submit: Ari Chivukula <arichiv@chromium.org>
Reviewed-by: Steven Bingler <bingler@chromium.org>
Commit-Queue: Ari Chivukula <arichiv@chromium.org>
Cr-Commit-Position: refs/heads/main@{#917770}
--
wpt-commits: a4fc13cb6502d00e60a0c1911d192b87538d1c20
wpt-pr: 30311
Automatic update from web-platform-tests
Revert "Fix test picture generation and add test."
This reverts commit 84222e05b7a97c034e40d2d118de4aac05a94c50.
Reason for revert: caused test to fail external/wpt/webrtc/RTCPeerConnection-helper-test.html, ex build: https://ci.chromium.org/p/chromium/builders/ci/Mac10.15%20Tests
Bug: 1245692
Original change's description:
> Fix test picture generation and add test.
>
> This CL changes the test picture from a growing box to a small moving
> box. It adds a test that the expected bandwidth target is achieved at
> the expected resolution.
>
> In a browser, 640x480 is not reached within 5 seconds. This is odd.
>
> Bug: None.
> Change-Id: I2d94f1c190c26cec6759ecc0f788feea8e52ed7c
> Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/3135833
> Commit-Queue: Harald Alvestrand <hta@chromium.org>
> Reviewed-by: Florent Castelli <orphis@chromium.org>
> Cr-Commit-Position: refs/heads/main@{#917211}
Bug: None.
Change-Id: I205c46cca2245f3f906d52cca87ad878b486ee82
No-Presubmit: true
No-Tree-Checks: true
No-Try: true
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/3137553
Bot-Commit: Rubber Stamper <rubber-stamper@appspot.gserviceaccount.com>
Reviewed-by: Kevin Marshall <kmarshall@chromium.org>
Reviewed-by: Harald Alvestrand <hta@chromium.org>
Commit-Queue: Harald Alvestrand <hta@chromium.org>
Cr-Commit-Position: refs/heads/main@{#917317}
--
wpt-commits: 1960b5daf5d84a0d9fd8b7064795267d79d70039
wpt-pr: 30308
onInspectorFrontDestroyed was throwing because when it
is called from the destroy callback of watchFronts, it
doesn't get a second argument, and the destructuring fails.
This patch defaults the second parameter to an empty object
in order to avoid this.
Differential Revision: https://phabricator.services.mozilla.com/D126235
With dedicated targets for each window, we can now only check that the message
originates from the target window, and not from its window tree
Some tests that were assuming the iframe had the same target as the top level document
are updated.
Differential Revision: https://phabricator.services.mozilla.com/D125935
A few tests were making assertions on targets
based on Fission being enabled or not. Since we're
creating targets for every iframe (with EFT) independently
of Fission, those tests were breaking.
Differential Revision: https://phabricator.services.mozilla.com/D125833
When we get data from D&D operation by nsWindow::OnDragDataReceivedEvent, speed up data transfer to child process by
WindowDragMotionHandler() call.
Differential Revision: https://phabricator.services.mozilla.com/D126211
Right now we query all MIME data types during D&D operation and rely on Gtk to filter out
missing ones. It leads to possible D&D failures as we can't process D&D events
from Gtk during it.
This patch does:
- Query MIME types available for D&D transfer by nsDragService::GetDragFlavors().
- Quit early at nsDragService::GetTargetDragData() when requested MIME can't be transferred by D&D.
- Use gdk_drag_context_get_suggested_action() instead of gdk_drag_context_get_actions().
- Check mTargetWindow before we use it.
Differential Revision: https://phabricator.services.mozilla.com/D126200
This is pretty much impossible to review as-is, so it may be more useful to review the script that made them. The repo is at https://github.com/htwyford/create-theme-script. parse-themes-script.js is the main script. It reads input.json. input.json is pulled from https://github.com/FirefoxUX/themes/blob/main/tokens/color/base.json, with some light edits to fix inconsistencies. The file in the FirefoxUX repo is an export of the colors in the Figma file: https://www.figma.com/file/xaRff6432QsirRftX8NZgb/MR2-Themes?node-id=86%3A17747.
The themes aren't perfect yet. For example, the text color in the Urlbar chiclet is sometimes wrong. They also don't consider the UX spec on badge colors, since UX is still actively updating that part of the spec. Since these themes are behind a pref, I think we should land it and fix the issues in followups. It will make it more clear what is changing when they're not being added en masse like this. Getting them in the tree ASAP also lets UX and QA get a head start on testing them.
Differential Revision: https://phabricator.services.mozilla.com/D125755
We sometimes miss reply to D&D motion event due to delayed DnsDragService::RunScheduledTask() call.
We can't call D&D handler directly from drag-motion event as it causes nested recursions and re-entrance
in glib loop so use higher priority for the delayed handler call at least.
Disable re-entrance of nsDragService::RunScheduledTask() which may happen when
we get D&D data from drag_drop event.
Provide more loggig to D&D code.
Differential Revision: https://phabricator.services.mozilla.com/D126182
