Automatic update from web-platform-tests
[UserTimingL3] {}, null, undefined passed to start treated as 0
According to the spec (https://w3c.github.io/user-timing/#measure-method),
UserTimingL3 API has the second argument being:
optional (DOMString or PerformanceMeasureOptions) startOrMeasureOptions
This indicates that the arg cannot distinguish between:
1. null
2. {}
3. undefined
4. { invalidDictProperty }
According to the spec:
3.1.3 measure() method...
3. Compute start time as follows..
4. Otherwise, let start time be 0.
When these indistinguishable values are passed, they should be
interpreted as start time being 0.
Bug: 953840
Change-Id: I052a7823d9ae8b27056f53f04a26fcc93421db75
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1691862
Commit-Queue: Liquan (Max) Gu <maxlg@chromium.org>
Reviewed-by: Nicolás Peña Moreno <npm@chromium.org>
Cr-Commit-Position: refs/heads/master@{#675873}
--
wpt-commits: 1f346b050d87656a040fa4fdf9f99567164e2a2b
wpt-pr: 17735
Automatic update from web-platform-tests
Handle non-negativeness of blur radius properly
Spec on blur radius says [1]:
"... Specifies the blur radius. Negative values are not allowed."
Current implementation performs parse time range check, which is
incorrect as calc() doesn't always have a double value. And spec
requires clamping instead of parse time check for calc() [2]:
"Parse-time range-checking of values is not performed within math
functions, and therefore out-of-range values do not cause the
declaration to become invalid. However, the value resulting from
an expression must be clamped to the range allowed in the target
context."
This patch changes it to pass |kValueRangeNonNegative| to
|ConsumeLength| as it handles non-negative calc() more properly.
This is also preparation for adding DCHECK into GetDoubleValue() to
avoid calling it on calc() without a double value.
[1] https://drafts.csswg.org/css-backgrounds-3/#shadow-blur-radius
[2] https://drafts.csswg.org/css-values-4/#calc-range
Bug: 979895
Change-Id: Ia8c36641f4ec4111ecf5a94a6ffc6eaacf22049c
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1691747
Reviewed-by: Emil A Eklund <eae@chromium.org>
Commit-Queue: Xiaocheng Hu <xiaochengh@chromium.org>
Cr-Commit-Position: refs/heads/master@{#675815}
--
wpt-commits: 65d3f17c56ea812d121669e6268a0cec4a4ea4a6
wpt-pr: 17741
Automatic update from web-platform-tests
Serialization of consecutive tokens. (#17746)
* Serialization of consecutive tokens.
Per <https://drafts.csswg.org/css-syntax/#serialization>, many token pairs, if they find themselves consecutive in a value, must be serialized with a comment between them, because otherwise upon serialization+reparse they'll produce a different token stream.
--
wpt-commits: dacb51937397fd06727727614e255f223e7cb9f3
wpt-pr: 17746
Automatic update from web-platform-tests
Add new tests for Core-AAM: (#17742)
* aria-expanded on: application, checkbox, switch
* aria-required on: checkbox
* aria-posinset and aria-setsize on: row
* inclusion of children of math role
* group as valid child of listbox
* menucheckbox as valid child of group in menu
* subscript and superscript roles
* meter role
* time role
* insertion and deletion roles
--
wpt-commits: f56d498d4dde12aacf8d6bbe591d35488af06a2b
wpt-pr: 17742
Automatic update from web-platform-tests
Fix feature policy origin comparison for sandbox.
When 'self' was specified in a feature policy header for a sandboxed
page, the resulting policy would not actually allow the feature in that
page, even when it would otherwise have been allowed. This corrects that
by assigning the correct origin to the policy and accepting that origin
in allowlists.
Bug: 973880, 690520
Change-Id: I93325bf24119068f8138f6e38507598cc30cbb06
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1689958
Commit-Queue: Ian Clelland <iclelland@chromium.org>
Reviewed-by: Jeremy Roman <jbroman@chromium.org>
Cr-Commit-Position: refs/heads/master@{#675737}
--
wpt-commits: 6493155a3b1c0a0236aa4059a1ef768961578f98
wpt-pr: 17729
Automatic update from web-platform-tests
[LayoutNG] Ensure self-collapsing block relayout when ancestor resolves.
This bug was a combination of older code before we had "forced" BFC
block-offsets.
We didn't correctly detect when we had a self-collapsing block (which
had floats within it) that needed relayout if the initial estimate
was wrong.
This makes sure we give ourselves enough information to detect these
cases.
Bug: 980803
Change-Id: Id1ea5e10d819cb4509fd7664564b75b876f0f7cb
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1690720
Commit-Queue: Ian Kilpatrick <ikilpatrick@chromium.org>
Commit-Queue: Emil A Eklund <eae@chromium.org>
Reviewed-by: Emil A Eklund <eae@chromium.org>
Cr-Commit-Position: refs/heads/master@{#675729}
--
wpt-commits: 137960cfe6ce5b4dfb7a6d0bad126d517e03e009
wpt-pr: 17733
Automatic update from web-platform-tests
Fix decision of whether to drop y value for 2d translate
Spec requires dropping y value for 2d translate if it's zero (*). The
current implementation contains bug that:
1. It rounds the value before checking, which results in incorrect
dropping of '0.1px'
2. It calls DoubleValue() on calc() without type checking, while
DoubleValue() on calc() is meaningless when the calc() doesn't
resolve to a simple numeric value (e.g., 1px - 1%). This results in
incorrect dropping of some calc().
Both are fixed by this patch.
This is also a preparation for adding DCHECK in DoubleValue() to ensure
that it's called only when calc() can be resolved into a simple numeric
value.
Bug: 979895
Change-Id: Ie8b846729b91e55006485a25fab0b65533983eac
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1692202
Reviewed-by: Anders Hartvoll Ruud <andruud@chromium.org>
Commit-Queue: Xiaocheng Hu <xiaochengh@chromium.org>
Cr-Commit-Position: refs/heads/master@{#675708}
--
wpt-commits: d947446c473a627ee68835eab7dcb928d85488bd
wpt-pr: 17684
Automatic update from web-platform-tests
[docs] Fix the code blocks in pipes.py (#17738)
* [docs] Fix the code blocks in pipes.py
Some code blocks were previously not correctly indented or surrounded
with empty lines.
* Apply suggestions from code review
--
wpt-commits: 9d23e22e542d2722bdae1ca5a0b780349a12be79
wpt-pr: 17738
Automatic update from web-platform-tests
Include initial frame replication state on portal adoption.
This is needed so that checks that occur as part of navigation
don't get confused by the replication state (notably the origin)
of the portal guest contents being missing from the remote frame
in the host contents' renderer.
Since this state is pushed down into the blink::RemoteFrame,
we need InitializeCoreFrame to be called before we can set the
replication state. As a result WebRemoteFrame::CreateForAdoptedPortal
is created, which performs this initialization with the provided
HTMLPortalElement.
Included web test crashes before this CL.
Bug: 980937
Change-Id: I843192e5f5aa8e4e022a88d318fa4a5e1d46a3d9
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1677163
Commit-Queue: Jeremy Roman <jbroman@chromium.org>
Reviewed-by: Daniel Cheng <dcheng@chromium.org>
Reviewed-by: Nasko Oskov <nasko@chromium.org>
Reviewed-by: Lucas Gadani <lfg@chromium.org>
Cr-Commit-Position: refs/heads/master@{#675702}
--
wpt-commits: 7ab751c5295223c124723888edd1b9378d123b92
wpt-pr: 17624
Automatic update from web-platform-tests
[LayoutNG] Do not use out-of-flow cache if in inline container
This patch changes |NGOutOfFlowLayoutPart::LayoutCandidate()|
not to use a cached result if it has an inline container.
|NGLayoutResult::OutOfFlowPositionedOffset()| has an offset
relative to the inline formatting context block. When an out-
of-flow object has inline contianer, this offset may change
without setting |NeedsLayout()| to the out-of-flow object.
Bug: 981686
Change-Id: Ib8376f791af0b95f6c2b6e5c2361b6a887c74e69
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1692107
Commit-Queue: Emil A Eklund <eae@chromium.org>
Reviewed-by: Emil A Eklund <eae@chromium.org>
Cr-Commit-Position: refs/heads/master@{#675686}
--
wpt-commits: e2d6701e64fbf664b53a4db86c53bffadd9fb9cf
wpt-pr: 17685
Automatic update from web-platform-tests
[docs] Remove legacy substitution escape sequences
The `{% raw %}` and `{% endraw %}` markers were required to correctly
render some content when the documentation website was generated by
Jekyll. They have no meaning in the context of the recently-introduced
Sphinx-powered build system, and they are pass through to the rendered
output directly.
Remove the markers.
--
wpt-commits: 5267fe13c8a23fe05f952ffd7e790e4d9fe2dcd9
wpt-pr: 17737
Automatic update from web-platform-tests
[ci] Split Edge Canary into a separate job
Do not run Edge Canary and Edge Dev in the same job to prevent them from
overwriting each other. This also makes Edge setup the same as Safari
(two jobs for Safari & STP) on Azure Pipelines.
--
wpt-commits: 5a6c83e5332ea8c072cd088bb4aa3320bf8605a7
wpt-pr: 17731
Automatic update from web-platform-tests
[web-animations] Make handling of keyframes more spec compliant
WebAnimations spec has two keyframe concepts:
A) keyframes [1]
B) computed keyframes [2]
A key difference is that in A property values remain unresolved while in
B these properties are resolved (e.g., shorthands expand to long hands
etc.)
KeyframeEffect.getKeyframes() is expected to return (A) but our current
implementation only keeps around (B).
This CL introduces the following changes to bring us closer to the
specified behavior:
1. Introduce a new map <property, value> in StringKeyframe
representing (A)
2. CSS property pairs are added to this map as long as they parse
correctly [3]
3. Use the new map to produce the result of getKeyframes()
There is some additional special handling required for shorthand
properties because Blink CSS parser does not really produce any
shorthand values. As a shortcut this patch introduces a new type
of css value (`CSSKeyframeShorthandValue`) which represents a
shorthand by encapsulating all its relevant longhand property/value
pairs.
Note that we continue our old behavior for SVG and Presentation attrs
which will be switch over in follow up patch.
[1] https://drafts.csswg.org/web-animations/#keyframes-section
[2] https://drafts.csswg.org/web-animations/#calculating-computed-keyframes
[3] https://drafts.csswg.org/web-animations/#process-a-keyframes-argument step 8
TEST:
- web-animations/interfaces/{Animatable, Keyframes} => PASS previously failing tests
- Added two new test cases in above covering animated custom props
- external/wpt/web-animations/interfaces/KeyframeEffect/getKeyframes.html => New test to cover serialization
Bug: 816956
Change-Id: Icc8e0bc4a0ee3019ad7d2c566aacb9d3aee0ffe3
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1634393
Commit-Queue: Majid Valipour <majidvp@chromium.org>
Reviewed-by: Stephen McGruer <smcgruer@chromium.org>
Reviewed-by: Anders Hartvoll Ruud <andruud@chromium.org>
Cr-Commit-Position: refs/heads/master@{#675665}
--
wpt-commits: 631824bf82ea23e59604b443ab62c9b123eb25d5
wpt-pr: 17083
Automatic update from web-platform-tests
[wpt PR 17288] - Revert "Active Processing for ConvolverNode", a=testonly
Automatic update from web-platform-tests
Revert "Active Processing for ConvolverNode"
This reverts commit 499976c54a2b086bbde54ae137d0ac4adb3cd427.
Reason for revert:
Findit (https://goo.gl/kROfz5) identified CL at revision 668149 as the
culprit for flakes in the build cycles as shown on:
https://analysis.chromium.org/p/chromium/flake-portal/analysis/culprit?key=ag9zfmZpbmRpdC1mb3ItbWVyQwsSDEZsYWtlQ3VscHJpdCIxY2hyb21pdW0vNDk5OTc2YzU0YTJiMDg2YmJkZTU0YWUxMzdkMGFjNGFkYjNjZDQyNww
Sample Failed Build: https://ci.chromium.org/buildbot/chromium.mac/WebKit%20Mac10.13%20%28retina%29/15474
Sample Failed Step: webkit_layout_tests on ATI GPU on Mac Retina on Mac-10.13.6
Sample Flaky Test: external/wpt/webaudio/the-audio-api/the-convolvernode-interface/active-processing.https.html
Original change's description:
> Active Processing for ConvolverNode
>
> The ConvolverNode should output a single channel of silence when it is not
> actively processing. Active processing is already supported and just needed
> to have the convolver output a single channel of silence when it is
> constructed, before anything is connected to it.
>
> Bug: 971389
> Change-Id: Icc894cd7ea6dec60a1eccb0fa4a3aea68ac5b51f
> Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1644899
> Commit-Queue: Raymond Toy <rtoy@chromium.org>
> Reviewed-by: Hongchan Choi <hongchan@chromium.org>
> Cr-Commit-Position: refs/heads/master@{#668149}
Change-Id: Ie7fedee332ca167c3dfd4fd99142de89234ff03a
No-Presubmit: true
No-Tree-Checks: true
No-Try: true
Bug: 971389
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1653924
Cr-Commit-Position: refs/heads/master@{#668243}
--
wp5At-commits: 2477f79f3c0757225d06762aeb1d3b1eea6eb52a
wpt-pr: 17288
bugzilla-url: https://bugzilla.mozilla.org/show_bug.cgi?id=1558779
gecko-commit: 9f3894cee297f22e93ac8b76176ec52474ce5a00
gecko-integration-branch: central
gecko-reviewers: testonly
--
wpt-commits: 0968cc9522f12cf46972ede6423d3ab0742ebc24
wpt-pr: 17722
Automatic update from web-platform-tests
[docs] Consolidate instructions on server features (#17482)
* [docs] Consolidate instructions on server features
The file `server-side.md` duplicates much of the information available
in the `docs/` directory. Remove `server-side.md`, preserving its unique
content in the relevant files within the `docs/` directory.
--
wpt-commits: 8119bc10583682676a3db9806c82ed4044e88e13
wpt-pr: 17482
Automatic update from web-platform-tests
Deal with the virtualenv .Python link being broken
This happens relatively frequently on macOS when using Homebrew, as
Homebrew installs into versioned paths, so following an upgrade and
a cleanup of the old install, the old path no longer exists.
--
wpt-commits: 5ef31c71071b18b3164eb9eca0e991aab4313c54
wpt-pr: 17522
Automatic update from web-platform-tests
Ensure that the mozlog lock is released when starting test runner processes.
--
wpt-commits: af7f76ee8f480e3dc5c8b3c39edebc7a50ae6abf
wpt-pr: 17564
Automatic update from web-platform-tests
Add missing help for text-transform math test. (#17686)
--
wpt-commits: 7ed3166ac0fb3cd33c63db4846f522a0d30f8b3e
wpt-pr: 17686
Automatic update from web-platform-tests
Defer laying out atomic inline while traversing inline children
This patch reverts [1] so that, when in full layout mode, it
prevents |LayoutInlineChildren()| from doing any layout in
its |InlineWalker| loop.
The crash occurs because |DirtyLineBoxesForObject()| can
destroy |InlineTextBox| without unlinking them from the list,
assuming the tree was already destroyed. We could either:
a. Change |DirtyLineBoxesForObject()| to unlink |InlineBox|es
it destroys.
b. Prevent laying out any children in this loop.
This patch chooses the optoin b.
Laying out atomic inline in the descendant traversal loop of
|LayoutInlineChildren()| was delayed after the loop in [2],
then it was changed to use a temporary list in [3].
The temporary list was removed in [1] because the original
test (fast/block/inline-children-root-linebox-crash.html) no
longer crashes. But laying out in the descendant traversal
loop should still be avoided.
[1] r390765 https://codereview.chromium.org/1933633002
[2] https://trac.webkit.org/changeset/118248/webkit
[3] https://trac.webkit.org/changeset/125810/webkit
Bug: 981602
Change-Id: Ie37a611aa7f6dc9b015d449ac75570c356ce654e
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1690379
Reviewed-by: Emil A Eklund <eae@chromium.org>
Commit-Queue: Koji Ishii <kojii@chromium.org>
Cr-Commit-Position: refs/heads/master@{#675479}
--
wpt-commits: 7b1ce62b0f9796d78a30d2765189e42a1d0616e6
wpt-pr: 17669
Automatic update from web-platform-tests
[UserTimingL3] Always serialize detail
UserTimingL3 specs that the detail property of the measure method and
the performance mark constructor both need to serialize the detail
property:
In https://w3c.github.io/user-timing/#measure-method:
"Set entry's detail attribute as follows:
If startOrMeasureOptions is present and is a
PerformanceMeasureOptions object, set it to the result of calling
the StructuredSerialize algorithm on startOrMeasureOptions's detail.
Otherwise, set it as null."
In https://w3c.github.io/user-timing/#the-performancemark-constructor:
"Set entry's detail attribute to the result of calling the
StructuredSerialize algorithm on markOptions's detail."
In the current implementation of the performancemark constructor, it
serializes the detail only when the world of constructor and
the world of accessing the detail are different. To correct this, we
should serialize in both situations.
In the current implementation of the measure method, it does serialize
in both cases, but it swallows the exception. We should not swallow the
exception so as to "fail early, fail loudly"
Bug: 953857, 963926
Change-Id: I471c4d207d9c05bc9b9b8b65315c49b0aa73a672
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1620748
Commit-Queue: Liquan (Max) Gu <maxlg@chromium.org>
Reviewed-by: Nicolás Peña Moreno <npm@chromium.org>
Reviewed-by: Jeremy Roman <jbroman@chromium.org>
Cr-Commit-Position: refs/heads/master@{#675461}
--
wpt-commits: 09cbd6bc9ecc7349e9e38e6b2dd7e1d53dfb79ef
wpt-pr: 16937
Automatic update from web-platform-tests
Update MsEdgeDriver download to get driver binary based on channel info (#17540)
--
wpt-commits: 97a1a7874a9d0e97c340701b7b5e5f526fbfecfe
wpt-pr: 17540
Automatic update from web-platform-tests
Reland "[WPT] Reland Upload a test for sending mouse events with key pressed"
This reverts commit d91140eca4d13c992fe1803664adaf9605199610.
(i.e. relands c4dc34414b232ed5575d6a61837e2cfb32def045)
Original change's description:
> Revert "[WPT] Reland Upload a test for sending mouse events with key pressed"
>
> This reverts commit c4dc34414b232ed5575d6a61837e2cfb32def045.
>
> Reason for revert: actionsWithKeyPressed.html fails on MacOS
>
> Original change's description:
> > [WPT] Reland Upload a test for sending mouse events with key pressed
> >
> > When we have keyboard events and mouse events send together, we should
> > set key modifies in mouse events. This test first presses some keys
> > then send some mouse events. This test only works on Chrome.
> >
> > Bug: 606367
> >
> > TBR=nzolghadr@chromium.org
> >
> > Change-Id: Idad0bcca2c03f961f6c6af211d7f843f06e14fc1
> > Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1544170
> > Reviewed-by: Lan Wei <lanwei@chromium.org>
> > Commit-Queue: Lan Wei <lanwei@chromium.org>
> > Cr-Commit-Position: refs/heads/master@{#653672}
>
> TBR=lanwei@chromium.org
>
> # Not skipping CQ checks because original CL landed > 1 day ago.
>
> Bug: 606367
> Change-Id: I1569088b67060cf8fa749ec0dc725fe551a990c5
> Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1689030
> Commit-Queue: Lan Wei <lanwei@chromium.org>
> Reviewed-by: Lan Wei <lanwei@chromium.org>
> Cr-Commit-Position: refs/heads/master@{#674840}
TBR=lanwei@chromium.org
Bug: 606367
Change-Id: I06139386a7eb4d0b2323d48688a7b2cffbe0a583
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1690568
Reviewed-by: Robert Ma <robertma@chromium.org>
Commit-Queue: Robert Ma <robertma@chromium.org>
Cr-Commit-Position: refs/heads/master@{#675379}
--
wpt-commits: d73c675597c48e53b6fe1955a866d820518c2ee9
wpt-pr: 17678
Automatic update from web-platform-tests
bluetooth: Standalone Tests for characteristic/
This change refactors the Web Bluetooth WPT tests in the characteristic/
directory to be standalone window tests, as well as use the async/await
pattern. This makes the tests easier to format and easier to read.
The generated tests are not updated in this patch because they will
require updating the Python script that generates them.
Bug: 509038
Change-Id: I74d2158bbff3ef3b30d5286129cc3fc855e9a9da
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1682840
Reviewed-by: Reilly Grant <reillyg@chromium.org>
Commit-Queue: Ovidio de Jesús Ruiz-Henríquez <odejesush@chromium.org>
Cr-Commit-Position: refs/heads/master@{#675371}
--
wpt-commits: 175bb1c4268b76f5023708a11c9b48c694e7eae7
wpt-pr: 17578
Automatic update from web-platform-tests
WPT: Fix contain-style-counters-004.html with correct expectations.
contain-style-counters-004.html has spans as non-siblings of the
div that prints the number.
This patch adjusts the expectations to print 0, instead of 20. Note
that contain-style-counters-003.html already tests for siblings
relationship.
R=tabatkins@chromium.org, chrishtr@chromium.org
Change-Id: I0ca60bed403ba31844688b7f97db8c50628fcaf7
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1689072
Reviewed-by: Chris Harrelson <chrishtr@chromium.org>
Commit-Queue: vmpstr <vmpstr@chromium.org>
Cr-Commit-Position: refs/heads/master@{#675309}
--
wpt-commits: a09e43f71b85ea8dff2f77d58489ce95852e84f8
wpt-pr: 17652
Automatic update from web-platform-tests
Do not expose element, event, and layout-shift entries via getEntries
Currently, Performance's getEntriesByType, getEntriesByName, and
getEntries methods include buffered entries of types 'element', 'event',
and 'layout-shift'. We do not want to expose these to the performance
timeline now that the buffered flag is shipped. This CL removes them
from those methods and fixes/adds web_tests that check those methods.
Bug: 581518, 823744, 883483, 972830
Change-Id: I0d74794bcc205cded6684472cc56836e253856cf
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1689073
Reviewed-by: Steve Kobes <skobes@chromium.org>
Commit-Queue: Nicolás Peña Moreno <npm@chromium.org>
Cr-Commit-Position: refs/heads/master@{#675032}
--
wpt-commits: c13852e02350b4e86a8c348885933f9e9b83e807
wpt-pr: 17676
Automatic update from web-platform-tests
Null-check the CSS Transition's effect when doing updates
If you use the (still experimental) getAnimations() API to modify a
running CSS Transition to have a null effect, and then immediately start
a new transition without letting the old one finish, we would try to
unconditionally access and use the |effect| member. Add some null checks
to avoid this.
Bug: 964113
Change-Id: I26de63234b5249689e82596173ac354ff1ca6eee
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1690924
Reviewed-by: Yi Gu <yigu@chromium.org>
Commit-Queue: Stephen McGruer <smcgruer@chromium.org>
Cr-Commit-Position: refs/heads/master@{#675305}
--
wpt-commits: 9e1ff30db5610b44c74aae2394e4af095536a75c
wpt-pr: 17673
Automatic update from web-platform-tests
Worker: Simplify worker-interception.https.html
This CL removes unnecessary iframe creation, and adds test helper functions to
simplify the tests. This is just refactoring, and should work just like before.
Bug: 731604, 906991
Change-Id: I775567c64df0a7ed4b1931caf285c06f60b0b14b
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1683916
Commit-Queue: Hiroki Nakagawa <nhiroki@chromium.org>
Reviewed-by: Matt Falkenhagen <falken@chromium.org>
Cr-Commit-Position: refs/heads/master@{#673963}
--
wpt-commits: 1bac9e9d1715ca642d09ceb2a089db2b82c8714c
wpt-pr: 17593
Automatic update from web-platform-tests
Update entryType names to follow conventions
Currently, we have entryTypes 'firstInput', 'layoutShift', and
'largestContentfulPaint' in experimental code (disabled by default, at
most under Origin Trial). This CL updates the naming to follow the
conventions specified at
https://w3ctag.github.io/design-principles/#casing-rules, so they become
'first-input', 'layout-shift', and 'largest-contentful-paint'.
Bug: 823744, 965505, 581518
Change-Id: I989a59fb62b886ab305b0e0fb4be8f3556c1ca82
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1685975
Commit-Queue: Nicolás Peña Moreno <npm@chromium.org>
Reviewed-by: Timothy Dresser <tdresser@chromium.org>
Cr-Commit-Position: refs/heads/master@{#674490}
--
wpt-commits: f1a492a84b4cd1e4146f8f9d84ec34f902bebb4a
wpt-pr: 17644
Automatic update from web-platform-tests
Automate tap action on an element in an iframe (#17260)
--
wpt-commits: a9d6d1b058f9d37cabaf735ea60f9b87acc8a02b
wpt-pr: 17260
Automatic update from web-platform-tests
[tools] Remove use of deprecated API
The "message" parameter of Pytests's `raises` method has been deprecated
[1]. This causes a test failure in the current configuration. Since the
parameter was not previously serving the intended purpose (that is, it
was *not* verifying the message of the raised exception), it may be
removed without degrading coverage.
[1] https://docs.pytest.org/en/4.6-maintenance/deprecations.html#message-parameter-of-pytest-raises
--
wpt-commits: 8b642c10f67f0728f17d3284367d93c7480d290e
wpt-pr: 17648
Automatic update from web-platform-tests
Restore webidl2.js.headers.
It was added in #6685 and removed accidentally in #16864.
--
wpt-commits: 3387cf524436af6e29ad89b1b04ec94b8a76d4f9
wpt-pr: 17586
Automatic update from web-platform-tests
LayoutNG: Fix sizing and position of abspos tables
LayoutBox::ContainingBlockLogicalHeightForPercentageResolution was
adding padding but only for tables. NG already supplies padding so don't
do this table exception if the containing block is NG.
ng_absolute_utils relied on specified table width/height being the final
say, but tables use those as another min size. This patch makes
ng_absolute_utils treat table height/width as Auto for main sizing and
uses it as another min-height/width.
ResolveInlineLengthInternal assumes that anything with inline size Auto
will expand to fill the available space, but tables do not.
The 4-argument version of ComputeMinAndMaxContentContribution makes a
similar assumption to ng_absolute_utils -- if a node has a width set,
that's the final width. The 3-argument version of
ComputeMinAndMaxContentContribution is aware of this so bypasses the
4-argument version for tables and instead uses legacy preferred width
calculations, so I made ComputeMinAndMaxContentSizeForOutOfFlow do the
same.
The tests all at least partially failed in NG before this patch. After
this patch, NG matches legacy everywhere, except for the first 2 cases
of -002.html, which were broken in legacy and NG before this patch. It
is now correct in NG.
We partially fail 003 and 005, but identically to legacy layout.
I added more checks to fast/table/absolute-table-percent-lengths.html
and moved it to the wpt repository as
css/css-tables/absolute-tables-001.html .
Bug: 973390
Change-Id: I8cea275b2ca19492a501b84cd86456ed6745a4b8
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1671851
Commit-Queue: David Grogan <dgrogan@chromium.org>
Reviewed-by: Morten Stenshorne <mstensho@chromium.org>
Reviewed-by: Aleks Totic <atotic@chromium.org>
Cr-Commit-Position: refs/heads/master@{#675142}
--
wpt-commits: 8e3db7f3a765d1783f7e8c068c8bf90de61b1f66
wpt-pr: 17432
Automatic update from web-platform-tests
CSS Animations: WPTs for responsive animations
column-rule-color animations respond to changes in currentColor
and inherited column-rule-color that occur while the animation
is in progress.
column-width animations repond to changes in font-size
and inherited column-width that occur while the animation
is in progress.
Note this test fails in
Edge 18
Firefox 67 (partially)
Safari 12.1
Bug: 812908
Change-Id: Ic28819a7f4aa3c8c30b11d58ea2600563a182c07
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1689612
Reviewed-by: Robert Flack <flackr@chromium.org>
Commit-Queue: Eric Willigers <ericwilligers@chromium.org>
Cr-Commit-Position: refs/heads/master@{#675042}
--
wpt-commits: 3c5c83759bb5062de405b1c29257f89e884759fb
wpt-pr: 17658
Automatic update from web-platform-tests
Iframe can't be scrolled after scrolling='no' is modified to 'yes' by script.
When a LocalFrameView object corresponding to an iframe with scrolling attr
initial value 'no' is created, it will set CanHaveScrollbars to false. When
script modifies scrolling attr to 'yes', CanHaveScrollbars status should be
changed.
For example:
<iframe id='iframe_node' src='...' scrolling='no'></iframe>
<script>
var iframe = document.getElementById('iframe_node');
iframe.onload = function () {
iframe.scrolling = 'yes';
};
</script>
Bug: 979085
Change-Id: I07465df4d9cf38c37a61fa2606e17f2bc34777ec
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1679750
Reviewed-by: Mason Freed <masonfreed@chromium.org>
Reviewed-by: Daniel Cheng <dcheng@chromium.org>
Commit-Queue: Mason Freed <masonfreed@chromium.org>
Cr-Commit-Position: refs/heads/master@{#675037}
--
wpt-commits: 1ae7eb52c8153973f16d6ebf3a8ba363199e8895
wpt-pr: 17619
Automatic update from web-platform-tests
Add Preload + SRI image destination tests
This CL adds all of the Preload + SRI tests for the as=image
destination. Chrome is not implementing Preload + SRI on non-{script,
style} destinations immediately, but it is good to add the tests for
these destinations so that the future implementation's success can
easily be tracked and measured.
R=kouhei@chromium.org, yhirano@chromium.org
Bug: 981419
Change-Id: I7e04c4ff9df400942782ab224ea5148f9874969f
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1689511
Reviewed-by: Yutaka Hirano <yhirano@chromium.org>
Commit-Queue: Dominic Farolino <dom@chromium.org>
Cr-Commit-Position: refs/heads/master@{#674980}
--
wpt-commits: 550b9db5975f88173a5c9ceb1d576376288f18d7
wpt-pr: 17655
Automatic update from web-platform-tests
Add more combinations for scrollIntoView (#17647)
This patch added more combinations of writing-mode and css direction for scrollIntoView.
Now we have tested: horizontal-tb, horizontal-tb with rtl, vertical-lr, vertical-lr with rtl and vertical-rl.
The scroll bar position of vertical-rl with rtl is not consistency in browsers, and the specification doesn't indicate the position. And scroll bar position will affect the scroll coordinate. So this test is pending.
--
wpt-commits: c9cb3807e0031f551aa4c188923640f6cebe793a
wpt-pr: 17647
Automatic update from web-platform-tests
Document::CheckComplete should be nop when called from RequestResource
Otherwise,
1 when the request is served from the memory cache, it conflicts with
a ScriptForbiddenScope (already covered).
2 when the request is a synchronous XHR call,
ResourceLoadObserver::DidFinishLoading is called synchronously,
which means Document.readyState change would be observable
synchronously (A NEW PROBLEM).
The ResponseSource parameter in ResourceLoadObserver::DidFinishLoading
was introduced to fix 1, but with this change it's not needed. Remove
it.
Bug: 977486
Change-Id: I103032600c66837475850c87d8a3823a85e5b27f
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1687014
Commit-Queue: Yutaka Hirano <yhirano@chromium.org>
Reviewed-by: Kouhei Ueno <kouhei@chromium.org>
Cr-Commit-Position: refs/heads/master@{#674947}
--
wpt-commits: 1d7610e955878a6fccd01cd545639c5629c237ee
wpt-pr: 17620
Automatic update from web-platform-tests
[Native File System] Remove remove and removeRecursively operations
These operations are no longer part of the spec. This CL removes their
implementation.
Bug: 968553
Change-Id: If2600a55bd724bf2fee537b36a3bece8fd5c1313
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1680173
Commit-Queue: Eero Häkkinen <eero.hakkinen@intel.com>
Reviewed-by: Marijn Kruisselbrink <mek@chromium.org>
Reviewed-by: Chris Palmer <palmer@chromium.org>
Cr-Commit-Position: refs/heads/master@{#674922}
--
wpt-commits: 8327a0378e5cbd6f31db6511bd2f4707ca2b1c20
wpt-pr: 17654
Automatic update from web-platform-tests
[Native File System] Implement removeEntry operation
The removeEntry operation is to substitute the remove and
removeRecursively operations according to the spec.
Bug: 968553
Change-Id: I3f99f51e643fe0d7c50859d6b971faf8446c3b47
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1680172
Reviewed-by: Kentaro Hara <haraken@chromium.org>
Reviewed-by: Chris Palmer <palmer@chromium.org>
Reviewed-by: Marijn Kruisselbrink <mek@chromium.org>
Commit-Queue: Eero Häkkinen <eero.hakkinen@intel.com>
Auto-Submit: Eero Häkkinen <eero.hakkinen@intel.com>
Cr-Commit-Position: refs/heads/master@{#674911}
--
wpt-commits: 6974bcc39ed029ab62c719378d8dbc66ad7495d5
wpt-pr: 17545
Automatic update from web-platform-tests
Re-enable RTCRtpSender-getStats.https.html tests.
This was temporarily disabled to unblock a roll.
A helper function used by the test is updated to reflect the fact that
audioLevel is not always present.
Bug: 736403
Change-Id: Ia2d7b51d58f23023bf5894fbf2707ba155d9ac9a
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1688954
Reviewed-by: Armando Miraglia <armax@chromium.org>
Commit-Queue: Henrik Boström <hbos@chromium.org>
Cr-Commit-Position: refs/heads/master@{#674856}
--
wpt-commits: 7de7666607b4e4ea60a04c601a4c91b3cb7d673a
wpt-pr: 17643
Automatic update from web-platform-tests
Revert "[WPT] Reland Upload a test for sending mouse events with key pressed"
This reverts commit c4dc34414b232ed5575d6a61837e2cfb32def045.
Reason for revert: actionsWithKeyPressed.html fails on MacOS
Original change's description:
> [WPT] Reland Upload a test for sending mouse events with key pressed
>
> When we have keyboard events and mouse events send together, we should
> set key modifies in mouse events. This test first presses some keys
> then send some mouse events. This test only works on Chrome.
>
> Bug: 606367
>
> TBR=nzolghadr@chromium.org
>
> Change-Id: Idad0bcca2c03f961f6c6af211d7f843f06e14fc1
> Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1544170
> Reviewed-by: Lan Wei <lanwei@chromium.org>
> Commit-Queue: Lan Wei <lanwei@chromium.org>
> Cr-Commit-Position: refs/heads/master@{#653672}
TBR=lanwei@chromium.org
Bug: 606367
Change-Id: I1569088b67060cf8fa749ec0dc725fe551a990c5
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1689030
Commit-Queue: Lan Wei <lanwei@chromium.org>
Reviewed-by: Lan Wei <lanwei@chromium.org>
Cr-Commit-Position: refs/heads/master@{#674840}
--
wpt-commits: 59d845bb6f7c9fb271e91262f659e7a8a3e70d9f
wpt-pr: 17641
Automatic update from web-platform-tests
Enable LayoutInstabilityMoveDistance by default.
This CL will be followed by a CL to remove the conditional logic
for move distance in LayoutShiftTracker.
Change-Id: I9ecb10a33d4c79830e240e96913da4bec652e96b
Bug: 981338
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1686713
Reviewed-by: Steve Kobes <skobes@chromium.org>
Commit-Queue: Bryan McQuade <bmcquade@chromium.org>
Cr-Commit-Position: refs/heads/master@{#674819}
--
wpt-commits: 218bb5da6f686abf472c91c4dcfe1c3133487f9d
wpt-pr: 17640
Automatic update from web-platform-tests
Adding multipart/byteranges to the never sniffed list. (#17557)
Multipart range requests would allow bypassing CORB since the response
will have type multipart/byteranges instead of the resource's
original MIME type.
Blocking multipart responses is compatible with the web because the
renderer can only see into the result of a fetch for a multipart file
when the request is made with CORS. Media tags only make single-range
requests which will not have the multipart type.
Bug: 802836
Change-Id: Ib4ff928fdf9c604055f0d5ec52fcb5bee578099f
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1678394
Commit-Queue: Kristina Nelson <krstnmnlsn@google.com>
Reviewed-by: Charlie Reis <creis@chromium.org>
Reviewed-by: Łukasz Anforowicz <lukasza@chromium.org>
Cr-Commit-Position: refs/heads/master@{#673851}
--
wpt-commits: c085fba4ff2ad1341e308b453f2c7207d4b27e4e
wpt-pr: 17557
Automatic update from web-platform-tests
[webnfc] Introduce AbortSignal in NFCPushOptions.
This CL added the AbortSignal member in NFCPushOptions and use it
to cancel unhappened push operation. Tests have also been adjuested
accordingly.
Original change in the Spec:
https://github.com/w3c/web-nfc/pull/165
Bug: 520391
Change-Id: Ic69fde2bc343c65ed5c337dcfdbdeca97b456ba5
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1680418
Commit-Queue: Donna Wu <donna.wu@intel.com>
Reviewed-by: Reilly Grant <reillyg@chromium.org>
Cr-Commit-Position: refs/heads/master@{#674723}
--
wpt-commits: 845988417926b440697dc5d66ba5fa8891cbea52
wpt-pr: 17550
Automatic update from web-platform-tests
Reland "Fix first line style caching issues"
This is a reland of 228ee33446fc20cef8fcc94503a91c16bbcfc6fb
Compared to the original CL, this relanded (again) CL has the
following changes:
- Remove the lazy update of first line image observers which was not
needed and didn't work (causing reentrancy of
UpdateFirstLineImageObservers),
- In LayoutObject::UpdateFirstLineImageObservers(), hold the first
line style's reference locally because UpdateImageObservers may
call LayoutBlockFlow::ImageChanged() which invalidates the cached
first line style.
Original change's description:
> Fix first line style caching issues
>
> 1. Cache conflict issue:
>
> Previously calls to Element:: Element::CachedStyleForPseudoElement()
> with different second parameters caused cache conflict. The second
> call to it might get unwanted result that was cached with a different
> second parameter.
>
> Now remove the second parameter, and the function gets and caches the
> pseudo style only based on its own computed style. If a caller wants
> a pseudo style based on another style, the caller should call
> Element::UncachedStyleForPseudoElement() and manage cache by itself.
>
> 2. Under-invalidation on change
>
> Cached first line style needs invalidation in two cases:
> a) when the ::first-line style changes, we should invalidate all
> cached first line styles affected.
> b) when style of an inline element (which inherits ::first-line
> style) changes, as the inline element's style has higher priority
> than the inherited ::first-line style, even if the style doesn't
> change, we need to invalidate the cached
> kPseudoIdFirstLineInherited style if the new style is different
> from the inherited ::first-line style.
>
> Bug: 979054
>
> Change-Id: Icffe68d5136349d12cce6e92d1d3bf153c28903a
> Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1678878
> Reviewed-by: Rune Lillesveen <futhark@chromium.org>
> Commit-Queue: Xianzhu Wang <wangxianzhu@chromium.org>
> Cr-Commit-Position: refs/heads/master@{#673574}
Bug: 979054
Change-Id: If7211c585df77964789140277839a8f1c3822a15
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1685672
Commit-Queue: Xianzhu Wang <wangxianzhu@chromium.org>
Reviewed-by: Fredrik Söderquist <fs@opera.com>
Cr-Commit-Position: refs/heads/master@{#674509}
--
wpt-commits: bc0eab5e5ed10c689f6f08028a2060daddb3e32a
wpt-pr: 17610
Automatic update from web-platform-tests
[PW] Make PW canvas transform account for device scale factor
Right now when the zoom-for-dsf feature is not enabled (currently on
Mac only), the recording canvas has its own logic to handle device
scale factor. Given an example, when we run chrome on Linux with
--force-device-scale-factor = 2, the transformation matrix for the
recording canvas has a scale of (2, 2). But when this is on Mac, the
transformation matrix has a scale of (1, 1). This causes a big problem
when call setTransform() canvas API because it changes the
transformation matrix in an unexpected way without accounting for the
device scale factor.
In this CL, we give the PaintRenderingContext2D a new member which
is specifically the device scale factor on a platform where the
zoom-for-dsf is not enabled. Then whenever a setTransform is called,
we need to account for the device scale factor.
Bug: 970783, 716231
Change-Id: Idaca018a9afab656f868079dd8dd9324327da4b4
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1682659
Reviewed-by: Stephen McGruer <smcgruer@chromium.org>
Reviewed-by: Fernando Serboncini <fserb@chromium.org>
Commit-Queue: Xida Chen <xidachen@chromium.org>
Cr-Commit-Position: refs/heads/master@{#674488}
--
wpt-commits: 6e93013acee5454f1a7313f42af409cf8bd46e65
wpt-pr: 17603
Automatic update from web-platform-tests
Fix pen pointer capture in iframe
When set/release/check pointercapture, we checked whether it's touch
type pointer. This is because touch like events will route to local frame
root's PME, while mouse like events will route event to hit tested frame.
However, it didn't check whether pen type events are send like touch.
This CL makes set/release/has pointercapture always check whether pointer
is active in the capture element's frame, if not, then check the local frame
root. So we can avoid checking the pointer type.
1.For touch-like events, pointer active state is not available in inner frame and
we will always fall into checking local frame root.
2.For mouse-like events:
* When trying to set capture to a frame that has pointer active,
|IsPointerIdActiveOnFrame| will return true and pointer capture is allowed.
* When pointer is active in the outer frame, and trying to set to an element in
inner frame; or pointer is active in the inner frame, and trying to set to an
element in outer frame. The element's frame's |IsPointerIdActiveOnFrame| will
failed and also |RootFramePointerActiveInCurrentFrame| will check locat root
frame's |element_under_pointer_| and forbid the request because target frame
dismatch.
This will fail the case that mouse down at local frame root and move to a inner
frame. Because we are moving inside and no "leave" event send to outer frame.
Outer frame still has info of pointer is active, and did not update the
|element_under_pointer_|. But this case is failed without this change as well.
We are not going to address this with this change.
The test has an expected file for the second test failing on 'pen'. It's
PEM.GetMouseCaptureTarget returns the target for kMouseId, but for layout
test, pen events is send as mouse-like but has different id.
It's unnecessary to fix this now as we have direct manipulation enabled
on all platform, we're not likely to have mouse-like pen event for capturing.
It'll be fixed once we have pen event on test driver work as current pen
flow. Our Q3 plan that having pointerevent states per page should also fix
this.
Bug: 970595
Change-Id: I7f9f390d1526eefee4657f7b859edf5ad4d163ee
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1649519
Commit-Queue: Ella Ge <eirage@chromium.org>
Reviewed-by: David Bokan <bokan@chromium.org>
Reviewed-by: Mustaq Ahmed <mustaq@chromium.org>
Reviewed-by: Navid Zolghadr <nzolghadr@chromium.org>
Cr-Commit-Position: refs/heads/master@{#674487}
--
wpt-commits: fbf38bd7d9dbc2f074d9110112d563ab47efd015
wpt-pr: 17358
Automatic update from web-platform-tests
[css-text] Hang sequence of white spaces that may otherwise overflow
When a item overflows, we try to find a safe position that breaks the
item to avoid it. However, we should consider that sequence of white
space characters might have to be hanged, honoring the white-space
property.
We manage that case correctly the overflowing text is part of a
previously broken (kSuccess result from BreakText) item. However, if
there is an item that starts with a white space sequence and it
overflows, we always try to break it instead of hanging.
This patch address the issue by detecting such overflowing item,
verifying that it has a white space sequence and the white-space
property's value allows hanging it. In that case, we should rewind
the iterator, so that the further item_results are removed and skip
the rest of the overflowing handling logic.
Bug: 973358
Change-Id: I0e2ad5c5740f4ce592dd83b8ff3d224dbb45117e
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1662451
Reviewed-by: Koji Ishii <kojii@chromium.org>
Commit-Queue: Javier Fernandez <jfernandez@igalia.com>
Cr-Commit-Position: refs/heads/master@{#674430}
--
wpt-commits: 6dd8159628d492944540af129cf90e4bd692420c
wpt-pr: 17376
Automatic update from web-platform-tests
Reland: Cleaning up ElementTiming and EventTiming observer buffer behavior.. (#17608)
Original land in crrev.com/c/1654876
Reverted in crrev.com/c/1680091
1) Remove methods that handle the buffers separately
2) Remove only buffering until onload
3) Fixing up wpt tests associated with both
Bug: 969123
Change-Id: Ie63b37d9ff118007462c635c745a4eacaa4be16e
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1684677
Commit-Queue: Emily Hanley <eyaich@chromium.org>
Reviewed-by: Steve Kobes <skobes@chromium.org>
Reviewed-by: Nicolás Peña Moreno <npm@chromium.org>
Cr-Commit-Position: refs/heads/master@{#674080}
--
wpt-commits: 83a739f36b26abff6d2f0ecf27038701072bac6f
wpt-pr: 17608
Automatic update from web-platform-tests
Update WebXR TestAPI and tests to match new spec (#17630)
Now that WebXR is more mature, the test API is being revisited with
input from other UAs. This change updates existing methods and tests
to match the new API, which largely involves:
* FakeXRDeviceInit now requires views
* FakeXRDeviceInit *CAN* support an initial viewerOrigin
* setXRPresentationFrameData is now setViews and set/clearViewerOrigin
* XRTest is now navigator.xr.test [1]
As part of this, webxr_test_constants were cleaned up and expanded to
help minimize boilerplate code in the tests. Where possible tests that
were previously calling setXRPresentationFrameData to ensure that the
fake device was "tracking" were updated to instead use a constant that
sets these values. Future work (after the API update is complete) will
identify if there are any cases where an initial requestAnimationFrame
could be removed.
Most changes (especially to vr/ tests) should be fairly mechanical, but
now that the eye offsets expected/set by the test are actually being
plumbed through, both xr/xrReferenceSpace_originOffset tests had to have
their expected matrices updated. Two tests had to be updated due to the
discovery of 981003.
Other changes to update to the Test API after this change should be
mostly additive and should actually already exist in the internal tests.
[1] XRTest currently still exists as a workaround until 944987 is merged
Also due to the fact that accessing xr blocks VR, the WebVR tests had to
be updated to set a special flag to tell the test API to not set the
object on XR. It's unlikely that this behavior will be able to be
removed after 944987, but to allow XRTest to be removed these tests have
had it migrated to navigator.vr.test
Bug:979312
Change-Id: I19ccffd921728ec3799252693d945b7b3f99d757
--
wpt-commits: 3f8ee365a7cacc091dc1dd1d0b5f21a89241c894
wpt-pr: 17630
Automatic update from web-platform-tests
Split out the manual parts of webauthn idlharness test (#17417)
* Remove manual parts of webauthn idlharness test
* Remove duplicated object
* Remove WebAuthentication interface
* Reinstate deleted challengeBytes
* Remove unneeded conditional
--
wpt-commits: cd18d4954c5184ecfb7a210aa939081c5e2552e8
wpt-pr: 17417
Automatic update from web-platform-tests
Add test of direction:rtl for ScrollIntoView (#17622)
* Add test of direction:rtl for ScrollIntoView
Test the behavior of ScrollIntoViewOptions in direction:rtl page.
* Fixed code review issues
Fixed code review issues.
--
wpt-commits: b7e9c990780705608441f5ae14ffb67e1aba7799
wpt-pr: 17622
Automatic update from web-platform-tests
Fix expectation for css/cssom-view/scrollIntoView-vertical-rl-writing-mode.html (#17621)
- Improve the code to make it more readable
- Better handle browser inconsistencies
--
wpt-commits: 5b224fe81f10d807bd47711d5a92f6754beb3b8a
wpt-pr: 17621
Automatic update from web-platform-tests
[webnfc] Adjust returned error for NFC HW not supported case.
According to the Spec, this CL adjusted some error types for NFC
hardware not supported and permission not allowed cases. Original
changes in Spec are listed as follows.
aa50e6e193https://github.com/w3c/web-nfc/pull/236
Bug: 520391
Change-Id: Ib4b017c0f3fcb71b030700f6b89a6ba3d76ae749
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1677736
Commit-Queue: Donna Wu <donna.wu@intel.com>
Reviewed-by: Kinuko Yasuda <kinuko@chromium.org>
Reviewed-by: Reilly Grant <reillyg@chromium.org>
Reviewed-by: Rijubrata Bhaumik <rijubrata.bhaumik@intel.com>
Cr-Commit-Position: refs/heads/master@{#674390}
--
wpt-commits: 213380b1fbb9305d64b72baedee710035689a50d
wpt-pr: 17519
Automatic update from web-platform-tests
Add tests for new math values of CSS text-transform (#16922)
* Modify mathvariant-transforms.py to generate CSS tests for "text-transform: math-*" (other than math-auto).
* Modify mathvariant-transforms.py to generate tests for single-char <mi> and text-transform: math-auto.
* Add a test to verify 'text-transform: math-auto' on text nodes of different lengths.
* Move text-transform-math tests into their own subdirectory.
* Update meta help to point to mathml core spec.
--
wpt-commits: 1af7e47d15ec171bb181fc5e3309bcfb52dd8f01
wpt-pr: 16922
Automatic update from web-platform-tests
mojo: Move meta API methods for receivers to $
Moves bindHandle(), createProxy(), and close() to the "$" member of the
generated receiver classes.
Bug: 968369
Change-Id: I0549bd23221343afda5a2733d601f6c56bdc1b57
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1639036
Reviewed-by: Ken Rockot <rockot@google.com>
Reviewed-by: calamity <calamity@chromium.org>
Commit-Queue: Giovanni Ortuño Urquidi <ortuno@chromium.org>
Cr-Commit-Position: refs/heads/master@{#674309}
--
wpt-commits: f83c9a52301a18819bd317b137b6c15553a1f2f2
wpt-pr: 17584
Automatic update from web-platform-tests
Reland "Insert a generated item after preserved spaces following a forced break"
This is a reland of 6b0ca8ce33e71d570fce11ae6fb5f898ddf053e4
The root cause of the regression that motivated the revert was that in the
original commit I removed a Failure entry by mistake from the TestExpectations
file.
The removed entry was added precisely for the word-break-break-all-004.html
test only for Mac platforms. This was the test that regressed, since the
original path didn't provide any behavior change affecting the mentioned
test.
So this CL tries to reland the original patch, but restoring the Failure entry
for the word-break-break-all-004.html test in Mac platforms.
Original change's description:
> Insert a generated item after preserved spaces following a forced break
>
> We already had to implement a special logic to deal with leading
> preserved white spaces due some optimizations we have in the line
> breaking code to give always breaking opportunities before spaces.
>
> However, the presence of forced breaks may cause some of the following
> spaces becoming leading preserved spaces in the next line.
>
> This patch applies the same logic we had for such leading spaces at
> the text's first line. It also needs to define additional cases in the
> function that evaluates the need of adding these generated breaking
> opportunity items.
>
> Bug: 976195
> Change-Id: If39e7ba8a535c15d33d6021c6cd04da9590b1131
> Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1667305
> Commit-Queue: Javier Fernandez <jfernandez@igalia.com>
> Reviewed-by: Koji Ishii <kojii@chromium.org>
> Cr-Commit-Position: refs/heads/master@{#671656}
TBR=kojii@chromium.org
Bug: 976195
Change-Id: Id55acef369248ec30f5cfa68cbb59051bf7af8f3
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1683234
Reviewed-by: Javier Fernandez <jfernandez@igalia.com>
Commit-Queue: Javier Fernandez <jfernandez@igalia.com>
Cr-Commit-Position: refs/heads/master@{#674305}
--
wpt-commits: 93483b4118a0af7fbb6ac3f1c4dc67fa9fbdf916
wpt-pr: 17618
Automatic update from web-platform-tests
Ensure that WebXR mojom types are properly assigned in JS
In JS the types aren't created via a standard "Constructor" like method
but instead by a map with the corresponding fields set. This updates a
few types that were being incorrectly constructed.
Bug: 980759
Change-Id: I1a6e8f033b89f08ffb92e6c1abf8a84631bf4649
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1686603
Commit-Queue: Alexander Cooper <alcooper@chromium.org>
Auto-Submit: Alexander Cooper <alcooper@chromium.org>
Reviewed-by: Bill Orr <billorr@chromium.org>
Cr-Commit-Position: refs/heads/master@{#674304}
--
wpt-commits: 9bbeb647445cd6858de91ee45ab4a05cb165e504
wpt-pr: 17617
Automatic update from web-platform-tests
[ElementTiming] Check opacity of image before exposing
Before CompositeAfterPaint, we would just rely on painting being skipped
when an image has opacity:0. Now, we need to explicitly check because
paint is no longer being skipped.
Bug: 959949, 879270
Change-Id: I57590798e7f76c4c63a6062f145f95fa1eec6c3a
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1680867
Reviewed-by: Chris Harrelson <chrishtr@chromium.org>
Commit-Queue: Nicolás Peña Moreno <npm@chromium.org>
Cr-Commit-Position: refs/heads/master@{#674252}
--
wpt-commits: 1e05bb271ae12717e383a6e3da625127cb88dcd0
wpt-pr: 17562
Automatic update from web-platform-tests
Move a pointercapture pointer test to non-wpt pointerevent tests.
This test initially added to wpt but the behavior is not clearly defined
in spec. So move to non-wpt pointerevents test.
See discussion in https://github.com/w3c/pointerevents/issues/291TBR=nzolghadr@chromium.org
Change-Id: Ic871c57526681d790d6e68081f5a4bf8316e8494
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1684464
Reviewed-by: Ella Ge <eirage@chromium.org>
Commit-Queue: Ella Ge <eirage@chromium.org>
Cr-Commit-Position: refs/heads/master@{#674191}
--
wpt-commits: 060265cb71f8cbe6e9a042e245232a0fabb07048
wpt-pr: 17614
Automatic update from web-platform-tests
[LayoutNG] Include negative margin beyond the right edge
This patch changes the line breaking behavior so that a
negative inline margin beyond the right edge can be included
in the line. Before this change, negative margins could move
the position back only when it's within the available width.
Legacy and WebKit don't do this, while Edge and Gecko do.
However, LayoutNG cannot render a page legacy and WebKit
render the same as other browser due to a bug in min-content.
Instead of copying the min-content bug in LayoutNG, this
patch fixes the page by matching to Edge and Gecko.
Bug: 979894
Change-Id: I07eedbb75f5c3b43e48b826bef18d95f1af872c9
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1684243
Commit-Queue: Koji Ishii <kojii@chromium.org>
Reviewed-by: Emil A Eklund <eae@chromium.org>
Cr-Commit-Position: refs/heads/master@{#674177}
--
wpt-commits: 9aacd72c9f396544fc93a1419fa8c9bfe02c015a
wpt-pr: 17612
Automatic update from web-platform-tests
Add manual test for print()
For https://github.com/whatwg/html/pull/4690.
--
wpt-commits: 8e53d2a89fe610b2ce480877c60bca88c949cf60
wpt-pr: 17439
Automatic update from web-platform-tests
[LayoutNG] Fix Google Docs font picker menu.
LayoutNG lacks a mechanism that's present in the legacy engine, that
handles situations where a descendant adds a scrollbar, which affects
the inline-size of an ancestor (fit-content sizing).
This CL only fixes the issue when this happens to an out-of-flow
positioned ancestor, which is enough to fix the Google Docs issue.
Bug: 978979
Change-Id: Ib60bbe244d8ee2cd80f3209b3322a097738841d3
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1685388
Commit-Queue: Morten Stenshorne <mstensho@chromium.org>
Commit-Queue: Emil A Eklund <eae@chromium.org>
Reviewed-by: Emil A Eklund <eae@chromium.org>
Cr-Commit-Position: refs/heads/master@{#674123}
--
wpt-commits: 4a9bf4d17af6f7527af4558369bcb4b709ee280c
wpt-pr: 17611
Automatic update from web-platform-tests
Don't consider fieldsets as replaced content.
They should be treated as regular block containers, as far as sizing is
concerned. The LayoutNG out-of-flow positioning machinery treated
fieldsets as replaced (since it was told to), and ended up with an
incorrect inline-size.
A "side-effect" of this fix is that now that fieldsets are no longer
treated as replaced content, they can can also fragment inside multicol
containers or paged media - just like any other block-level container.
This is good. There should be no reason to prohibit fragmentainer
breaks inside fieldsets. Added a test for it, to assert the new
behavior.
Bug: 979523
Change-Id: I2fd73f8324d3207c5d12c500b5632f631468ab34
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1685096
Reviewed-by: Ian Kilpatrick <ikilpatrick@chromium.org>
Reviewed-by: Koji Ishii <kojii@chromium.org>
Commit-Queue: Morten Stenshorne <mstensho@chromium.org>
Cr-Commit-Position: refs/heads/master@{#674096}
--
wpt-commits: 727763fa27f9d8d36c1c337dc2ac52ac092c61f0
wpt-pr: 17602
Automatic update from web-platform-tests
[LargestContentfulPaint] Add attributes and checks
This CL adds checks so that we do not expose rendering timestamp of
images which do not pass the Timing Allow Origin check. It also adds
attributes found in:
https://wicg.github.io/largest-contentful-paint/#sec-performance-largest-contentful-paint-candidate
The ImagePaintTimingDetector did not plumb the ImageResourceContent to
the ImageRecord, so this is fixed to allow LCP Calculator to make use of
it.
Bug: 965505
Change-Id: I019658dc6d130c8cea336f1ca70de6a32beebbb0
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1674588
Reviewed-by: Chris Harrelson <chrishtr@chromium.org>
Commit-Queue: Nicolás Peña Moreno <npm@chromium.org>
Cr-Commit-Position: refs/heads/master@{#674085}
--
wpt-commits: f7dbdc0bb7a3566b1b28e1cb49178ba9d71b1bc6
wpt-pr: 17500
Automatic update from web-platform-tests
Revert "Reland "Fix first line style caching issues""
This reverts commit 3bce815a57b12cd5be22ecd0ba838c417f6179fd.
Reason for revert: crbug.com/980421
Original change's description:
> Reland "Fix first line style caching issues"
>
> This is a reland of 228ee33446fc20cef8fcc94503a91c16bbcfc6fb
>
> Compared to the original CL, this relanded CL removes the lazy update
> of first line image observers which was not needed and didn't work
> (causing reentrancy of UpdateFirstLineImageObservers).
>
> Original change's description:
> > Fix first line style caching issues
> >
> > 1. Cache conflict issue:
> >
> > Previously calls to Element:: Element::CachedStyleForPseudoElement()
> > with different second parameters caused cache conflict. The second
> > call to it might get unwanted result that was cached with a different
> > second parameter.
> >
> > Now remove the second parameter, and the function gets and caches the
> > pseudo style only based on its own computed style. If a caller wants
> > a pseudo style based on another style, the caller should call
> > Element::UncachedStyleForPseudoElement() and manage cache by itself.
> >
> > 2. Under-invalidation on change
> >
> > Cached first line style needs invalidation in two cases:
> > a) when the ::first-line style changes, we should invalidate all
> > cached first line styles affected.
> > b) when style of an inline element (which inherits ::first-line
> > style) changes, as the inline element's style has higher priority
> > than the inherited ::first-line style, even if the style doesn't
> > change, we need to invalidate the cached
> > kPseudoIdFirstLineInherited style if the new style is different
> > from the inherited ::first-line style.
> >
> > Bug: 979054
> >
> > Change-Id: Icffe68d5136349d12cce6e92d1d3bf153c28903a
> > Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1678878
> > Reviewed-by: Rune Lillesveen <futhark@chromium.org>
> > Commit-Queue: Xianzhu Wang <wangxianzhu@chromium.org>
> > Cr-Commit-Position: refs/heads/master@{#673574}
>
> Bug: 979054
> Change-Id: Ic0c05a4a3d36401f7ae18c41cad81e7613bb3cbc
> Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1682283
> Commit-Queue: Xianzhu Wang <wangxianzhu@chromium.org>
> Reviewed-by: Fredrik Söderquist <fs@opera.com>
> Cr-Commit-Position: refs/heads/master@{#673756}
TBR=wangxianzhu@chromium.org,fs@opera.com,futhark@chromium.org
Change-Id: I59fe9947deba7267e69f5b76c9ff6b3c0251b8c6
No-Presubmit: true
No-Tree-Checks: true
No-Try: true
Bug: 979054
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1685415
Reviewed-by: Xianzhu Wang <wangxianzhu@chromium.org>
Commit-Queue: Xianzhu Wang <wangxianzhu@chromium.org>
Cr-Commit-Position: refs/heads/master@{#674074}
--
wpt-commits: fef95ecd2b3d59504eb4b5405db5c2995ab32710
wpt-pr: 17609
Automatic update from web-platform-tests
[Trusted Types] Support Trusted Types on eval.
Modify V8's eval callback to take both CSP and Trusted Types into
account. This should not modify Blink's behaviour in case TT is not present.
Bug: 940927, 980127
Change-Id: Ic11956cd455d352f2ba9d3fd9e4a00bb12e8d852
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1591274
Commit-Queue: Daniel Vogelheim <vogelheim@chromium.org>
Reviewed-by: Andy Paicu <andypaicu@chromium.org>
Reviewed-by: Mike West <mkwst@chromium.org>
Cr-Commit-Position: refs/heads/master@{#674038}
--
wpt-commits: 4ed0465b108be61f184a896b7cb9e2c9356fd1d9
wpt-pr: 17525
Automatic update from web-platform-tests
[LayoutNG] Keep trucated fragments for `text-overflow: ellipsis`
This patch keeps truncated fragments for `text-overflow:
ellipsis` as "hidden for paint", but still uses them for
when querying layout information such as `offsetWidth` or
editing.
Bug: 947593, 978392
Change-Id: Ic9d6888728339dc3ea37672b1ba004cddd295216
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1678915
Reviewed-by: Emil A Eklund <eae@chromium.org>
Commit-Queue: Emil A Eklund <eae@chromium.org>
Commit-Queue: Koji Ishii <kojii@chromium.org>
Cr-Commit-Position: refs/heads/master@{#673777}
--
wpt-commits: 424e3bce2245105a515ba8f5d97e6852bb69e973
wpt-pr: 17588
Automatic update from web-platform-tests
Reland "Fix first line style caching issues"
This is a reland of 228ee33446fc20cef8fcc94503a91c16bbcfc6fb
Compared to the original CL, this relanded CL removes the lazy update
of first line image observers which was not needed and didn't work
(causing reentrancy of UpdateFirstLineImageObservers).
Original change's description:
> Fix first line style caching issues
>
> 1. Cache conflict issue:
>
> Previously calls to Element:: Element::CachedStyleForPseudoElement()
> with different second parameters caused cache conflict. The second
> call to it might get unwanted result that was cached with a different
> second parameter.
>
> Now remove the second parameter, and the function gets and caches the
> pseudo style only based on its own computed style. If a caller wants
> a pseudo style based on another style, the caller should call
> Element::UncachedStyleForPseudoElement() and manage cache by itself.
>
> 2. Under-invalidation on change
>
> Cached first line style needs invalidation in two cases:
> a) when the ::first-line style changes, we should invalidate all
> cached first line styles affected.
> b) when style of an inline element (which inherits ::first-line
> style) changes, as the inline element's style has higher priority
> than the inherited ::first-line style, even if the style doesn't
> change, we need to invalidate the cached
> kPseudoIdFirstLineInherited style if the new style is different
> from the inherited ::first-line style.
>
> Bug: 979054
>
> Change-Id: Icffe68d5136349d12cce6e92d1d3bf153c28903a
> Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1678878
> Reviewed-by: Rune Lillesveen <futhark@chromium.org>
> Commit-Queue: Xianzhu Wang <wangxianzhu@chromium.org>
> Cr-Commit-Position: refs/heads/master@{#673574}
Bug: 979054
Change-Id: Ic0c05a4a3d36401f7ae18c41cad81e7613bb3cbc
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1682283
Commit-Queue: Xianzhu Wang <wangxianzhu@chromium.org>
Reviewed-by: Fredrik Söderquist <fs@opera.com>
Cr-Commit-Position: refs/heads/master@{#673756}
--
wpt-commits: f05fad8e3fee427059ab8040badee942c2ac8555
wpt-pr: 17582
Automatic update from web-platform-tests
[LayoutNG] Resolve auto inline margins on root.
The root HTML element is in a unique situation in LayoutNG, in that its
parent is a regular block container that is laid out by the legacy
engine. In legacy, block container layout expects the children to set
their own margin. Make sure that we do this.
Bug: 978702
Change-Id: I84b8e36bd706e3c01096cd9bd75a0747ddf265ff
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1683973
Reviewed-by: Koji Ishii <kojii@chromium.org>
Reviewed-by: Ian Kilpatrick <ikilpatrick@chromium.org>
Commit-Queue: Morten Stenshorne <mstensho@chromium.org>
Cr-Commit-Position: refs/heads/master@{#673733}
--
wpt-commits: 16680001ee64a84651c44311a605cc7b2bcc18bf
wpt-pr: 17587
Automatic update from web-platform-tests
CSS math-script-level and math-style: Update meta help links. (#17604)
--
wpt-commits: eda082d3fb516a0bf522cbc4198999a605d7a438
wpt-pr: 17604
Automatic update from web-platform-tests
- Update help meta tags. (#17601)
- Minor markup and documentation tweaks.
--
wpt-commits: ec636b58e82b466c201fb89ea7265092e9f05284
wpt-pr: 17601
Automatic update from web-platform-tests
[css-text] Correct inconsistencies between text and assertion in line-break:anywhere tests (#17597)
--
wpt-commits: 708c7910e71c408d2ff7a1093a2b82bd3415449e
wpt-pr: 17597
Automatic update from web-platform-tests
[cssom] Add tests for CSSGroupingRule
These tests are somewhat contentious due an inconsistency in the CSS
specification:
https://github.com/w3c/csswg-drafts/issues/3528
--
wpt-commits: 814db842c492f8c480749b671e6bc31c83c68645
wpt-pr: 17276
Automatic update from web-platform-tests
[docs] Relocate instruction for reviewers
This information is not relevant for first-time contributors. Including
it in the same document as the instructions for new contributors risks
confusing or intimidating the less experienced audience.
Relocate the instructions to a dedicated document that can be discovered
by those who specifically wish to learn about reviewing submissions.
--
wpt-commits: 53bf566b00aa0efd9b80496d5d488a338b16ec1c
wpt-pr: 17371
Automatic update from web-platform-tests
Revert "Fix first line style caching issues"
This reverts commit 228ee33446fc20cef8fcc94503a91c16bbcfc6fb.
Reason for revert: Clusterfuzz found several bugs caused by the CL.
Original change's description:
> Fix first line style caching issues
>
> 1. Cache conflict issue:
>
> Previously calls to Element:: Element::CachedStyleForPseudoElement()
> with different second parameters caused cache conflict. The second
> call to it might get unwanted result that was cached with a different
> second parameter.
>
> Now remove the second parameter, and the function gets and caches the
> pseudo style only based on its own computed style. If a caller wants
> a pseudo style based on another style, the caller should call
> Element::UncachedStyleForPseudoElement() and manage cache by itself.
>
> 2. Under-invalidation on change
>
> Cached first line style needs invalidation in two cases:
> a) when the ::first-line style changes, we should invalidate all
> cached first line styles affected.
> b) when style of an inline element (which inherits ::first-line
> style) changes, as the inline element's style has higher priority
> than the inherited ::first-line style, even if the style doesn't
> change, we need to invalidate the cached
> kPseudoIdFirstLineInherited style if the new style is different
> from the inherited ::first-line style.
>
> Bug: 979054
>
> Change-Id: Icffe68d5136349d12cce6e92d1d3bf153c28903a
> Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1678878
> Reviewed-by: Rune Lillesveen <futhark@chromium.org>
> Commit-Queue: Xianzhu Wang <wangxianzhu@chromium.org>
> Cr-Commit-Position: refs/heads/master@{#673574}
TBR=wangxianzhu@chromium.org,fs@opera.com,futhark@chromium.org
Change-Id: Ief1eb5749624f877e04d7bff19f42c6f94844b70
No-Presubmit: true
No-Tree-Checks: true
No-Try: true
Bug: 979054, 979961, 979942, 979940
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1682282
Reviewed-by: Xianzhu Wang <wangxianzhu@chromium.org>
Commit-Queue: Xianzhu Wang <wangxianzhu@chromium.org>
Cr-Commit-Position: refs/heads/master@{#673594}
--
wpt-commits: ff6212c69e2519a94fd62d6f1e06dd0dac6152d3
wpt-pr: 17581
Automatic update from web-platform-tests
Fix first line style caching issues
1. Cache conflict issue:
Previously calls to Element:: Element::CachedStyleForPseudoElement()
with different second parameters caused cache conflict. The second
call to it might get unwanted result that was cached with a different
second parameter.
Now remove the second parameter, and the function gets and caches the
pseudo style only based on its own computed style. If a caller wants
a pseudo style based on another style, the caller should call
Element::UncachedStyleForPseudoElement() and manage cache by itself.
2. Under-invalidation on change
Cached first line style needs invalidation in two cases:
a) when the ::first-line style changes, we should invalidate all
cached first line styles affected.
b) when style of an inline element (which inherits ::first-line
style) changes, as the inline element's style has higher priority
than the inherited ::first-line style, even if the style doesn't
change, we need to invalidate the cached
kPseudoIdFirstLineInherited style if the new style is different
from the inherited ::first-line style.
Bug: 979054
Change-Id: Icffe68d5136349d12cce6e92d1d3bf153c28903a
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1678878
Reviewed-by: Rune Lillesveen <futhark@chromium.org>
Commit-Queue: Xianzhu Wang <wangxianzhu@chromium.org>
Cr-Commit-Position: refs/heads/master@{#673574}
--
wpt-commits: c0a3e93389bdcc9e8ad12d3988e4568d48b78c9d
wpt-pr: 17561
Automatic update from web-platform-tests
test correct exception when attempting to draw a closed ImageBitmap to a canvas. (#17211)
* add test for drawImage of closed canvas
* make consistent with existing imageBitmap tests
--
wpt-commits: f0ed5874274b40b1fbe14f8ab6a072ef3df12a6e
wpt-pr: 17211
Automatic update from web-platform-tests
Ship `referer` header length limitation. (#17204)
This patch adds web platform tests to verify the behavior of the
referrer header and `document.referrer` after flipping the length
limitiation flag to on-by-default.
Intent to Ship: https://groups.google.com/a/chromium.org/forum/#!msg/blink-dev/lckJ5OAkUNo/ooVNdvhLAgAJ
Bug: 959757
Change-Id: I385f8f12f20109c50ededb9018a08cd665cb6868
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1646784
Commit-Queue: Mike West <mkwst@chromium.org>
Reviewed-by: Jochen Eisinger <jochen@chromium.org>
Cr-Commit-Position: refs/heads/master@{#667039}
--
wpt-commits: 9e059910c901218756454b7979c6f95a6f306854
wpt-pr: 17204
Automatic update from web-platform-tests
HTML: applet no longer exists
For HTML purposes that's "worse" than obsolete.
In the same directory there's historical.html to ensure it won't be added again.
Fixes#17559.
--
wpt-commits: c30f7af7ebc3269ba6790122272692c12c920f02
wpt-pr: 17570
Automatic update from web-platform-tests
Fix crash when setting invalid @font-face descriptors via CSSOM.
When @font-face rules are parsed normally (as part of a stylesheet),
the descriptors are parsed with CSSParserToken::ParseAsAtRuleDescriptorID,
which is allowed to return kInvalid.
However, when using CSSOM, the parsing takes a different path that first
creates a CSSPropertyID, then converts it to a AtRuleDescriptorID using
CSSPropertyIDAsAtRuleDescriptor, which *isn't* allowed to return kInvalid.
Fix by allowing kInvalid, and gracefully handling this in
CSSPropertyParser::ParseFontFaceDescriptor.
Note that this is the only callsite for CSSPropertyIDAsAtRuleDescriptor.
BUG=977953
Change-Id: I42081dd4bae3c828843be3ca994ff4f48d06a7f4
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1676629
Reviewed-by: Rune Lillesveen <futhark@chromium.org>
Commit-Queue: Anders Hartvoll Ruud <andruud@chromium.org>
Cr-Commit-Position: refs/heads/master@{#673262}
--
wpt-commits: 3487a399f3a1603d97388a5452530a89c10e47e1
wpt-pr: 17543
Automatic update from web-platform-tests
Cleanup types in vr_service.mojom
This changes various vectors in the vr_service mojom to use the correct
gfx.mojom.geometry type. It's largely mechanical, but the change from
array<float, 16> to gfx::Transform did introduce the need to transpose
the transforms being passed across that boundary.
For WMR, this resulted in a consistent way to convert from Matrix4x4 to
gfx::Transform, so some manual usages were consolidated to use a helper
function.
Bug: 732403
Change-Id: I1eefaa90c7b7d5ea6fa03c01522422bd37f7290a
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1669891
Commit-Queue: Alexander Cooper <alcooper@chromium.org>
Reviewed-by: Daniel Cheng <dcheng@chromium.org>
Reviewed-by: Bill Orr <billorr@chromium.org>
Reviewed-by: danakj <danakj@chromium.org>
Reviewed-by: Klaus Weidner <klausw@chromium.org>
Cr-Commit-Position: refs/heads/master@{#673193}
--
wpt-commits: c48a7037e8059f9a331d311ba711de2da4abf625
wpt-pr: 17477
Automatic update from web-platform-tests
Added a link to spec in 6 CSS2/values tests (#17514)
--
wpt-commits: a3f623801e5fef0bb8a806fbc8119c32d57240d1
wpt-pr: 17514
Automatic update from web-platform-tests
Added a link to spec in 13 CSS2/values tests (#17513)
--
wpt-commits: 4c26654f77551d686bbc255cee95089b12aa6907
wpt-pr: 17513
Automatic update from web-platform-tests
Added a link to spec to numbers-units-00[5-6] tests (#17512)
--
wpt-commits: 95a1d50e544d1894110449e207f8e8cc92324dc8
wpt-pr: 17512
Automatic update from web-platform-tests
Added a link to spec to numbers-units-00[2-3] tests (#17510)
--
wpt-commits: 78f9cce31d3a905598773b0486f9567e8a4e6193
wpt-pr: 17510
Automatic update from web-platform-tests
Added a link to spec to numbers-units-004 test (#17508)
--
wpt-commits: 288514517e5012ad4459c63e8b37e6dcff3fd561
wpt-pr: 17508
Automatic update from web-platform-tests
Added another link to spec to anumbers-units-001 test (#17507)
--
wpt-commits: 73f28194f388630a856827e63e286883e6c42a12
wpt-pr: 17507
Automatic update from web-platform-tests
[LayoutNG] Fix for incorrect Legacy containing block oof removal
When containing block changes from being an OOF container
to not being one, it needs to remove the positioned descendants
from its list of positioned objects.
If an OOF object remains in the positioned objects list,
it will cause NGOutOfLayoutPart::SweepLegacyDescendants
to go into an infinite loop and run out of memory. This
was fixed to just trigger a DCHECK in a recent CL.
LayoutInline::StyleDidChange did not perform OOF descendant
removal correctly. LayoutInline OOF descendants are stored in
first non-anonymous block on parent chain, (container's container)
which is not the same as ContainingBlock().
Added a method that computes container's container to
LayoutObject, and made LayoutInline call it.
This fix also works for fixed, because
fixed containers are always absolute containers.
Bug: 977930
Change-Id: I9f88fe676ec67d9d6751f3502507d773ed978dc8
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1680123
Commit-Queue: Aleks Totic <atotic@chromium.org>
Reviewed-by: Ian Kilpatrick <ikilpatrick@chromium.org>
Cr-Commit-Position: refs/heads/master@{#673156}
--
wpt-commits: a8405825541d608fd26f28b0af36e6b5c42fd70a
wpt-pr: 17542
Automatic update from web-platform-tests
Test that font-family does not match against postscript names. (#17004)
The spec states "A font family name only specifies a name given to a set
of font faces, it does not specify an individual face." [1] however a
postscript name uniquely identifies a specific face rather than a family
and so should not be matched by font-family.
[1] https://drafts.csswg.org/css-fonts-3/#font-family-prop
--
wpt-commits: 79f8156a104bb55409d84b0f9ffedc13c55f26f4
wpt-pr: 17004
Automatic update from web-platform-tests
[css-ui] Test for appearance/-webkit-appearance and initial value
--
wpt-commits: 9d3498a34109f0a36c5f181c6f18d75607f9e566
wpt-pr: 14979
Automatic update from web-platform-tests
[LayoutNG] Fix hit-testing on floats in line box
This patch fixes hit-testing floats when they are included in
a line box, and extend the height of the line box.
LayoutNG places floats within a line box, and make its bounds
to include floats. Hit-testing code checks floating objects
only if foreground hit-testing did not succeeded.
This patch changes foreground hit-testing on line box to fail
if there's a float at the location. It should be ok because
line box does not have foreground objects, but this is not
the most efficient way to fix this problem. We might want to
revisit either where to put such floats, or how to hit-test
floats.
Bug: 977856
Change-Id: I687dd44ddfbf7ebcb25b38a9d90470ea939cb053
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1680074
Commit-Queue: Emil A Eklund <eae@chromium.org>
Reviewed-by: Emil A Eklund <eae@chromium.org>
Cr-Commit-Position: refs/heads/master@{#672911}
--
wpt-commits: 7baea31d05c76d6d37975bb08304088f7f50d480
wpt-pr: 17544
Automatic update from web-platform-tests
Revert "[PW] Fix paint worklet canvas clip bound calculation"
This reverts commit aa7c84555637cabd3fdb0c03d781cc6f72b9efa2.
Reason for revert: virtual/scalefactor200/external/wpt/css/css-paint-api/hidpi/canvas-transform.https.html test fails on WebKit Mac10.13 (retina).
This test consistently fails starting from https://ci.chromium.org/p/chromium/builders/ci/WebKit%20Mac10.13%20%28retina%29/16135
Original change's description:
> [PW] Fix paint worklet canvas clip bound calculation
>
> Right now when we initialize the paint canvas for paint worklet with a
> scale factor or browser zoom, we do not record the zoom factor in the
> canvas transform. As a result, when computing the clip bound we will
> apply the wrong transformation matrix and thus sometimes the contents
> can be completely clipped out while it should not.
>
> This CL fixes the problem and adds a layout test.
>
> Bug: 970783
> Change-Id: Ie3709ea5845ca2b405245d4b36f9fe8f99052bad
> Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1678287
> Reviewed-by: Stephen McGruer <smcgruer@chromium.org>
> Commit-Queue: Xida Chen <xidachen@chromium.org>
> Cr-Commit-Position: refs/heads/master@{#672722}
TBR=xidachen@chromium.org,smcgruer@chromium.org
Change-Id: I6c6710f76ba204c2db9d90be56845c04ff072a54
No-Presubmit: true
No-Tree-Checks: true
No-Try: true
Bug: 970783
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1679696
Reviewed-by: Sergey Poromov <poromov@chromium.org>
Commit-Queue: Sergey Poromov <poromov@chromium.org>
Cr-Commit-Position: refs/heads/master@{#672870}
--
wpt-commits: 0804e0fcb5033df46553c9e245ffd4abbf122cf2
wpt-pr: 17553
Automatic update from web-platform-tests
[LayoutNG] Fixed crash issue when LI + zero height checkbox
List marker would resolve the bfc block_offset, but it didn't continue
to resolve its parent's. So it's possible that the bfc block_offset of
LI isn't resolved after layout list marker which will cause crash issue.
This patch make LI resolve it after layout list marker.
Bug: 969741
Change-Id: I1e1f2500217c51f24cb2633002d9a78ca1088d6c
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1645135
Commit-Queue: cathie chen <cathiechen@igalia.com>
Reviewed-by: Ian Kilpatrick <ikilpatrick@chromium.org>
Reviewed-by: Morten Stenshorne <mstensho@chromium.org>
Reviewed-by: Koji Ishii <kojii@chromium.org>
Cr-Commit-Position: refs/heads/master@{#672853}
--
wpt-commits: 296089d51afdc8ee3c0c905d89d3ddebf7633af8
wpt-pr: 17183
Automatic update from web-platform-tests
Add Preload + SRI web platform tests for script + style destinations
More preload destinations will be added after this CL, as well as the
main Preload + SRI implementation, to keep the changes here minimal.
R=kouhei@chromium.org, yhirano@chromium.org, yoavweiss@chromium.org
Bug: 677022
Change-Id: I596ac006874fb02bcd330422bd6496a05d388d45
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1669954
Commit-Queue: Dominic Farolino <dom@chromium.org>
Reviewed-by: Yutaka Hirano <yhirano@chromium.org>
Reviewed-by: Mike West <mkwst@chromium.org>
Cr-Commit-Position: refs/heads/master@{#672808}
--
wpt-commits: f0c2701197a8dc5682b0361698b73a0be9fd983c
wpt-pr: 17426
Automatic update from web-platform-tests
[PW] Fix paint worklet canvas clip bound calculation
Right now when we initialize the paint canvas for paint worklet with a
scale factor or browser zoom, we do not record the zoom factor in the
canvas transform. As a result, when computing the clip bound we will
apply the wrong transformation matrix and thus sometimes the contents
can be completely clipped out while it should not.
This CL fixes the problem and adds a layout test.
Bug: 970783
Change-Id: Ie3709ea5845ca2b405245d4b36f9fe8f99052bad
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1678287
Reviewed-by: Stephen McGruer <smcgruer@chromium.org>
Commit-Queue: Xida Chen <xidachen@chromium.org>
Cr-Commit-Position: refs/heads/master@{#672722}
--
wpt-commits: a5f7dc57600b83a47626507dd30216b0fa447b66
wpt-pr: 17533
Automatic update from web-platform-tests
Recalculate descendant styles when display changes to custom layout
Previously this only happened if the old display was custom layout,
but not when changing from a standard display to a custom one.
This was a typo, the same condition was checked twice for old_style,
but the second occurrence should check new_style instead.
The change matters when there are inline children that should be
blockified by the custom layout, without the patch they would remain
inline.
BUG=978990
TEST=external/wpt/css/css-layout-api/layout-child-inlines-dynamic.https.html
Change-Id: I73bcf679209dd4bc4a0d7a49f4345ee699bbf37e
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1678484
Reviewed-by: Rune Lillesveen <futhark@chromium.org>
Commit-Queue: Oriol Brufau <obrufau@igalia.com>
Cr-Commit-Position: refs/heads/master@{#672718}
--
wpt-commits: b26fcfbbc09d2fb3bfa1b9bbc643f7d4ffb971b0
wpt-pr: 17536
Automatic update from web-platform-tests
[LargestContentfulPaint] Fix flaky test observe-image.html
The timestamp captured at onload does not seem to always be smaller than
the render timestamp. This CL changes the lower bound so the test is no
longer flaky.
Bug: 978587
Change-Id: I48c058ac0abb684cb4406617500356d0aae019e8
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1678836
Commit-Queue: Nicolás Peña Moreno <npm@chromium.org>
Reviewed-by: Yoav Weiss <yoavweiss@chromium.org>
Cr-Commit-Position: refs/heads/master@{#672691}
--
wpt-commits: 0a02aa2599f1914a56b8557f29ec579ca5e6e701
wpt-pr: 17528
Automatic update from web-platform-tests
Query WebMediaPlayer when checking for presence of video frame.
ReadyState has never been suficient to test whether a video frame is
actually present. Instead we need to know when the frame has actually
been delivered to the compositor.
This patch updates HTMLVideoElement to query a new WebMediaPlayer method
which exposes the compositor's knowledge of the video frame.
Note: This patch also fixes some spurious logs and HTMLVideoElement tests
that were silently doing nothing.
BUG=974012,974190
TEST=new unittest.
TBR=eae
Change-Id: I1df54491c1fd7c2246ce0f60bc6e794b8b84476a
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1659490
Commit-Queue: Dale Curtis <dalecurtis@chromium.org>
Reviewed-by: Mounir Lamouri <mlamouri@chromium.org>
Cr-Commit-Position: refs/heads/master@{#672682}
--
wpt-commits: b6b0d9b33bd45d468f7835cb79b7de318123aead
wpt-pr: 17348
Automatic update from web-platform-tests
Web MIDI: reflect spec's [SecureContext] to relevant .idl files
This is a follow up to disallow Web MIDI uses on insecure origins.
After this change, the Web MIDI related interfaces won't appear
on insecure origins.
https://github.com/WebAudio/web-midi-api/issues/183https://github.com/WebAudio/web-midi-api/issues/199
Bug: 924471
Change-Id: I61d55d5ba192a6cbcea9e29f97c4bfe11ac4b983
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1673491
Commit-Queue: Takashi Toyoshima <toyoshim@chromium.org>
Reviewed-by: Kent Tamura <tkent@chromium.org>
Cr-Commit-Position: refs/heads/master@{#672474}
--
wpt-commits: 12b6e03e14f991cc20a6b80dc7c4dc09d78ba9bf
wpt-pr: 17520
--HG--
rename : testing/web-platform/tests/feature-policy/reporting/midi-report-only.html => testing/web-platform/tests/feature-policy/reporting/midi-report-only.https.html
rename : testing/web-platform/tests/feature-policy/reporting/midi-report-only.html.headers => testing/web-platform/tests/feature-policy/reporting/midi-report-only.https.html.headers
rename : testing/web-platform/tests/feature-policy/reporting/midi-reporting.html => testing/web-platform/tests/feature-policy/reporting/midi-reporting.https.html
rename : testing/web-platform/tests/feature-policy/reporting/midi-reporting.html.headers => testing/web-platform/tests/feature-policy/reporting/midi-reporting.https.html.headers
Automatic update from web-platform-tests
[LayoutNG] Don't run the OOFLayoutPart if we have any adjoining floats.
If we have adjoining floats we'll revisit the node to correctly place
floats. During the first pass we may not have floats present within
lineboxes, causing NGContainerFragmentBuilder::GetChildOffset to fail.
This makes sure we don't run the OOFLayoutPart for this condition.
There is a followup cleanup patch I'll send out for review later which
speculatively places floats avoiding this condition also:
https://chromium-review.googlesource.com/c/chromium/src/+/1619047
Bug: 976397
Change-Id: I10db8207357572521fc9103134cb6fbf885c4a94
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1676727
Commit-Queue: Morten Stenshorne <mstensho@chromium.org>
Reviewed-by: Morten Stenshorne <mstensho@chromium.org>
Reviewed-by: Koji Ishii <kojii@chromium.org>
Cr-Commit-Position: refs/heads/master@{#672423}
--
wpt-commits: 5a42c001e74045dd9347ce9870fb7a8f17e6286a
wpt-pr: 17504
Automatic update from web-platform-tests
[WebUSB] Fix usage of WTF::Vector as a bit vector
When using WTF::Vector as a replacement for WTF::BitVector the bits
should be set using assignment rather than with insert() and EraseAt().
The latter changes the length of the bit vector as it is used.
Bug: 977638
Change-Id: I59390412013c97c7c06aa7239cab75a2557e8f07
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1673131
Commit-Queue: Reilly Grant <reillyg@chromium.org>
Reviewed-by: Ovidio de Jesús Ruiz-Henríquez <odejesush@chromium.org>
Cr-Commit-Position: refs/heads/master@{#672358}
--
wpt-commits: a18f55bcc133a4638fd77c67d0bc41e48c70d15d
wpt-pr: 17468
Automatic update from web-platform-tests
Set default font size for OffscreenCanvas to 10px
There's no hard standard on this, so we are matching the behavior
of detached canvas.
Bug: 977876
Change-Id: I84f8e32000f290d6826722863bb6fb16116e2f18
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1675911
Commit-Queue: Fernando Serboncini <fserb@chromium.org>
Reviewed-by: Yi Xu <yiyix@chromium.org>
Cr-Commit-Position: refs/heads/master@{#672198}
--
wpt-commits: 79f12a79675c0b084737d43b1af2cb85f97bbff9
wpt-pr: 17496
Automatic update from web-platform-tests
Indexed DB: store/index keyPath property should return same instance
On IDBObjectStore and IDBIndex instances, the keyPath property says is
defined in the spec with: "if this attribute returns an object
(specifically an Array), it returns the same object instance every
time it is inspected". Web platform tests for this were missing.
Since the property is 'any' in Web IDL (since it could be a string or
Array) it can't use the [SameObject] extended attribute. And in the
Blink IDL, it can't use the [SaveSameObject] hint since that requires
[SameObject]. Instead, use the [CachedAttribute] hint and never dirty
the cache.
Bug: 977048
Change-Id: Ic4ede31a3a01e7a8a90c4f54eb8a989a673fcf22
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1674495
Commit-Queue: Daniel Murphy <dmurph@chromium.org>
Reviewed-by: Daniel Murphy <dmurph@chromium.org>
Cr-Commit-Position: refs/heads/master@{#672137}
--
wpt-commits: 1fd5380edd2390b2584c2c51754749d68916a525
wpt-pr: 17480
Automatic update from web-platform-tests
Individual WPT tests for each backdrop-filter
This CL adds a specific, individual WPT test for each type of
backdrop-filter effect.
If you find yourself here, looking for the test for blur() or
a multi-filter (combined) test, see crbug.com/978481.
Bug: 831485
Change-Id: I065fd94dbfe5ef0813ffe90f07878247147d3261
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1674708
Auto-Submit: Mason Freed <masonfreed@chromium.org>
Reviewed-by: Robert Ma <robertma@chromium.org>
Commit-Queue: Mason Freed <masonfreed@chromium.org>
Cr-Commit-Position: refs/heads/master@{#672131}
--
wpt-commits: d4a8ad3cb52de06a90b991dd0107f5ffbe206125
wpt-pr: 17501
Automatic update from web-platform-tests
Fetch: Origin outside of CORS
Ensure we have more coverage for how to set the Origin header outside of
CORS. And also how its value is impacted by Referrer Policy.
Fetch change: https://github.com/whatwg/fetch/pull/908.
Co-authored-by: Junior Hsu <cuveehsu@gmail.com>
Co-authored-by: Anne van Kesteren <annevk@annevk.nl>
--
wpt-commits: ef44bff0adaa07f2e420a0cbc1bc493cd5786656
wpt-pr: 14260
Automatic update from web-platform-tests
math-script-level-004: Fix expectation for the font-size values. (#17534)
When the MATH table has scriptScriptPercentScaleDown the spec says 0.5041 = 0.71^2 should be used as the default value.
The scale80-0-scaleup test incorrectly uses a hardcoded value of 0.71 instead.
This patch fixes this issue and introduces new variables for the fallback values of scriptPercentScaleDown and scriptScriptPercentScaleDown
to clarify the expectation.
--
wpt-commits: 677e4ac410c6538a76c72d704768dc9c0e8bc0c5
wpt-pr: 17534
Automatic update from web-platform-tests
KV Storage: add timeout=long and reenable disabled test
Like several IndexedDB tests, KV Storage can cause timeouts unless given more time.
BUG=936615
Change-Id: I8069d77088625b32867a5ebcc9ce69c4810a18ac
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1670621
Reviewed-by: Joshua Bell <jsbell@chromium.org>
Commit-Queue: Domenic Denicola <domenic@chromium.org>
Cr-Commit-Position: refs/heads/master@{#671576}
--
wpt-commits: 88bac6b712530867081709212b365a6a01feee04
wpt-pr: 17421
Automatic update from web-platform-tests
KV Storage: update to match IDL
Along with https://chromium-review.googlesource.com/c/chromium/src/+/1670572, this aligns the KV Storage implementation with https://github.com/WICG/kv-storage/pull/68, which uses Web IDL to define the API. The observable changes are to:
* Enumerability of methods
* Adding @@toStringTag (affecting Object.prototype.toString.call)
This includes web platform tests that abuse the current idlharness.js infrastructure, plus some ad-hoc hand-written tests that we expect to be generated by future versions of idlharness.js once the relevant Web IDL pull requests are merged. It removes the existing API surface tests and helpers in favor of idlharness.js.
Bug: 931263
Change-Id: I9205d1a8b3040617cbb6200f825ba9ad250e61c5
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1672137
Commit-Queue: Domenic Denicola <domenic@chromium.org>
Reviewed-by: Joshua Bell <jsbell@chromium.org>
Cr-Commit-Position: refs/heads/master@{#671975}
--
wpt-commits: 72da3e36e3657a358c4e74b495e582262b40a016
wpt-pr: 17443
Automatic update from web-platform-tests
Fix font size expectation (#17524)
The local variable should match the actual font-size.
--
wpt-commits: 5fd60735ebebf21db46f272587052c360ebd7615
wpt-pr: 17524
Automatic update from web-platform-tests
[docs] Simplify recommendation for advanced usage
The steps listed in the "Tips & Tricks" section largely summarize the
instructions in the preceeding sections with two exeptions:
1. the recommendation to rebase (which lacks a definition)
2. the warning against modifying the branch following submission (which
conflicts with a prior section)
Remove the steps and expand the recommendation of Pro Git.
--
wpt-commits: 4b93a2e9c9f904265ae755291ca253045adb3f66
wpt-pr: 17369
Automatic update from web-platform-tests
Merge pull request #17292 from gsnedders/ahem-part2b
Add ahem.css to many more tests
--
wpt-commits: daf75e3bc7829fc4de06bff7136009210aafe3e5
wpt-pr: 17292
Automatic update from web-platform-tests
[Navigation Timing] secureConnectionStart should be non-zero
The value of secureConnectionStart should be non-zero for secure
context navigations. It wasn't. This CL makes sure that it is!
Bug: 977519
Change-Id: Ia12e2842bc061e687512f1ed94d02c07f4dd42ee
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1672068
Reviewed-by: Nicolás Peña Moreno <npm@chromium.org>
Commit-Queue: Yoav Weiss <yoavweiss@chromium.org>
Cr-Commit-Position: refs/heads/master@{#671799}
--
wpt-commits: 87e1dabfa2ef2cf1d1946f428d19973b9bb35ddc
wpt-pr: 17438
Automatic update from web-platform-tests
Add more tests for fraction gaps. (#17486)
These tests check the gap parameters from https://mathml-refresh.github.io/mathml-core/#fractions-mfrac
The existing frac-parameters-1 and frac-parameters-2 tests use mspace elements as numerator/denominator so the box returned by getBoundingClientRect() perfectly encloses the visual rendering of these elements.
These versions rely on reftests to check when mtext elements are used as numerator/denominator with a font typo ascent/descent different from the actual glyph's bounding box ascent/descent. In each case, three glyphs are tested (centered, above or below the baseline).
These tests pass on gecko/webkit but currently fail with the chromium-mathml build for glyphs that are not centered on the baseline.
--
wpt-commits: ecc43f56d11b2485f17bf57c0db2031778d3eca9
wpt-pr: 17486
Automatic update from web-platform-tests
Bring XRRay.matrix and XRRay constructors up to spec
Minor update to XRRay.matrix algorithm to recompute internal matrix
if it was detached.
Modification to XRRay's constructors to throw InvalidStateError in
case direction normalization cannot be performed.
Update WPT tests to account for the change.
Bug: 957763
Change-Id: I7e65b660de013badbc2cff249d5a75a5fa75ab96
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1671871
Reviewed-by: Brandon Jones <bajones@chromium.org>
Commit-Queue: Piotr Bialecki <bialpio@chromium.org>
Cr-Commit-Position: refs/heads/master@{#671779}
--
wpt-commits: 766f40ade665477bdc44b5070a5d7cf533748552
wpt-pr: 17462
Automatic update from web-platform-tests
[wpt PR 16992] - Revert "Frame freezing on load.", a=testonly
Automatic update from web-platform-tests
Revert "Frame freezing on load."
This reverts commit 0daefdcf34ba9a8c21ab907b2159c5652169a45b.
Reason for revert:
DCHECK(!GetDocument()->LoadEventStillNeeded());
in local_frame.cc is failing at
https://ci.chromium.org/p/chromium/builders/ci/Mac10.13%20Tests%20%28dbg%29
Original change's description:
> Frame freezing on load.
>
> Move load event checks to renderer side so it is consistent with the
> page lifecycle spec. Add additional resource that is slow to load
> to ensure the load event is waited for correctly before freezing.
>
> BUG=907125
>
> Change-Id: I8f45ddffcf3ce88f6244307833325a54e1c2d395
> Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1625464
> Reviewed-by: Kentaro Hara <haraken@chromium.org>
> Reviewed-by: Avi Drissman <avi@chromium.org>
> Commit-Queue: Dave Tapuska <dtapuska@chromium.org>
> Cr-Commit-Position: refs/heads/master@{#662340}
TBR=avi@chromium.org,haraken@chromium.org,dtapuska@chromium.org
# Not skipping CQ checks because original CL landed > 1 day ago.
Bug: 907125
Change-Id: I4e27e30519359e2a3ef0d2c92a0fc9d88ac78aff
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1628499
Reviewed-by: Yuki Shiino <yukishiino@chromium.org>
Reviewed-by: Kentaro Hara <haraken@chromium.org>
Commit-Queue: Yuki Shiino <yukishiino@chromium.org>
Cr-Commit-Position: refs/heads/master@{#663009}
--
wp5At-commits: 138d2e938d493a5c8435025162759c2e34b3b1d1
wpt-pr: 16992
bugzilla-url: https://bugzilla.mozilla.org/show_bug.cgi?id=1554083
gecko-commit: 6b5ce266625e08bf06a40e6fd2b9b419a9f7f3ed
gecko-integration-branch: central
gecko-reviewers: testonly
--
wpt-commits: 87c8ca1b6f9debb046752ca92d6cea7af6e9f17c
wpt-pr: 17463
Automatic update from web-platform-tests
[wpt PR 17233] - Active Processing for ConvolverNode, a=testonly
Automatic update from web-platform-tests
Active Processing for ConvolverNode
The ConvolverNode should output a single channel of silence when it is not
actively processing. Active processing is already supported and just needed
to have the convolver output a single channel of silence when it is
constructed, before anything is connected to it.
Bug: 971389
Change-Id: Icc894cd7ea6dec60a1eccb0fa4a3aea68ac5b51f
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1644899
Commit-Queue: Raymond Toy <rtoy@chromium.org>
Reviewed-by: Hongchan Choi <hongchan@chromium.org>
Cr-Commit-Position: refs/heads/master@{#668149}
--
wp5At-commits: 6531eab802391a24dabca973cc4329abbeadd896
wpt-pr: 17233
bugzilla-url: https://bugzilla.mozilla.org/show_bug.cgi?id=1558611
gecko-commit: 3b9b543d1c8db5573155dc6ff44e65f871ae6ec9
gecko-integration-branch: central
gecko-reviewers: testonly
--
wpt-commits: 9e43b67f4fec9e0a1ffe2d91a5b4fd0c9b5ad705
wpt-pr: 17461
Automatic update from web-platform-tests
Revert "Insert a generated item after preserved spaces following a forced break"
This reverts commit 6b0ca8ce33e71d570fce11ae6fb5f898ddf053e4.
Reason for revert:
Broke external/wpt/css/css-text/word-break/word-break-break-all-004.html on many Mac bots:
https://ci.chromium.org/p/chromium/builders/ci/Mac10.10%20Tests/44291https://ci.chromium.org/p/chromium/builders/ci/Mac10.11%20Tests/39449https://ci.chromium.org/p/chromium/builders/ci/Mac10.12%20Tests/24278https://ci.chromium.org/p/chromium/builders/ci/Mac10.13%20Tests/14909https://ci.chromium.org/p/chromium/builders/ci/WebKit%20Mac10.13%20%28retina%29/16025
Original change's description:
> Insert a generated item after preserved spaces following a forced break
>
> We already had to implement a special logic to deal with leading
> preserved white spaces due some optimizations we have in the line
> breaking code to give always breaking opportunities before spaces.
>
> However, the presence of forced breaks may cause some of the following
> spaces becoming leading preserved spaces in the next line.
>
> This patch applies the same logic we had for such leading spaces at
> the text's first line. It also needs to define additional cases in the
> function that evaluates the need of adding these generated breaking
> opportunity items.
>
> Bug: 976195
> Change-Id: If39e7ba8a535c15d33d6021c6cd04da9590b1131
> Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1667305
> Commit-Queue: Javier Fernandez <jfernandez@igalia.com>
> Reviewed-by: Koji Ishii <kojii@chromium.org>
> Cr-Commit-Position: refs/heads/master@{#671656}
TBR=jfernandez@igalia.com,kojii@chromium.org
Change-Id: I418bb0d30fce7f9ac5724f76c0a854bcb070aa5c
No-Presubmit: true
No-Tree-Checks: true
No-Try: true
Bug: 976195
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1674249
Reviewed-by: Dan Beam <dbeam@chromium.org>
Commit-Queue: Dan Beam <dbeam@chromium.org>
Cr-Commit-Position: refs/heads/master@{#671703}
--
wpt-commits: 3d97450f2c48758db3f76f3360ff7c3e4069c320
wpt-pr: 17465
Automatic update from web-platform-tests
Add test checking SharedArrayBuffers are transferred over MessageChannels
SharedArrayBuffers in Chrome don't end up getting passed correctly when
it crosses a document boundary. This is because of the simplistic
Cluster Agent Key implementation. SharedArrayBuffers are currently
passed correctly provided the MessageChannel is all inside the same
document. (a document and itself, or a document and an associated worker).
BUG=798572
Change-Id: I012b2fbdae52ebee7d26034caa996f2e0cf929ae
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1672777
Reviewed-by: Ben Kelly <wanderview@chromium.org>
Commit-Queue: Dave Tapuska <dtapuska@chromium.org>
Cr-Commit-Position: refs/heads/master@{#671672}
--
wpt-commits: b3418ad6bf7b49e59025e5278e1291587c6fc10e
wpt-pr: 17454
Automatic update from web-platform-tests
Insert a generated item after preserved spaces following a forced break
We already had to implement a special logic to deal with leading
preserved white spaces due some optimizations we have in the line
breaking code to give always breaking opportunities before spaces.
However, the presence of forced breaks may cause some of the following
spaces becoming leading preserved spaces in the next line.
This patch applies the same logic we had for such leading spaces at
the text's first line. It also needs to define additional cases in the
function that evaluates the need of adding these generated breaking
opportunity items.
Bug: 976195
Change-Id: If39e7ba8a535c15d33d6021c6cd04da9590b1131
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1667305
Commit-Queue: Javier Fernandez <jfernandez@igalia.com>
Reviewed-by: Koji Ishii <kojii@chromium.org>
Cr-Commit-Position: refs/heads/master@{#671656}
--
wpt-commits: 9348328991a301000225e6da1010755eceaf1389
wpt-pr: 17393
Automatic update from web-platform-tests
[docs] Add detail on writing Python handlers (#17386)
Extend the documentation on writing Python handlers to more completely
describe the feature. Introduce a short example to concretely
demonstrate its usage. Apply similar changes to the corresponding
documentation in the `wptserve` directory for parity.
--
wpt-commits: acbefd606a1b9284b96d21a4a332d0202e36d531
wpt-pr: 17386
Automatic update from web-platform-tests
[Azure Pipelines] put parenthesis pack in step names
This reverts drive-by changes for consistency in commit
e75855649360298870a40ed78874bd4ab1470c20, which actually left the step
names inconsistent.
--
wpt-commits: 42ea2f906ab2367dc4403fb5cf8f7d476f9e0e5c
wpt-pr: 17409
Automatic update from web-platform-tests
Improve failing condition of a test (#17452)
So it doesn't time out.
--
wpt-commits: 1f512077ca0998e0af60ff727eea8c8ba5d78491
wpt-pr: 17452
Automatic update from web-platform-tests
Revert "[serve] Make pywebsocket honor log_level config"
This reverts commit 7edf9eabfae3550a4dd2a48625e68b02a734de15.
--
Fix pywebsocket logging properly
The previous attempt (#17124) called pywebsocket._configure_logging,
which had the unwanted side effect of resetting lots of logging setup in
the pywebsocket subprocess (e.g. mozlog configured by wptrunner -- the
log messages produced by pywebsocket lost its structure and became a
simple STDERR text captured by mozlog.capture).
Instead of calling pywebsocket._configure_logging, the original issue
(log level in config.json not honoured by pywebsocket) should be fixed
by setting the log level on the root logger instead, which will set up
all loggers (including those in third-party libraries) correctly. This
is done only on the entry point of `wpt serve` and won't affect
wptrunner which has its own logging setup.
--
wpt-commits: b693b1b954a8828dc5a0c69319e234b9e3b2d76d, 6e7ac6df08e23d9c8e7aa3223f387f71beb39b9c
wpt-pr: 17433
Automatic update from web-platform-tests
Handle null ::first-letter text layout object.
Ideally, UpdateFirstLetterPseudoElement() will remove the ::first-letter
pseudo element from the DOM when there is no text which can contain the
first letter. Hence, we should not end up in a situation where we try to
AttachLayoutTree for the FirstLetterPseudoElement with a null
first_letter_text.
However, the layout tree is different in AttachLayoutTree() than in
UpdateFirstLetterPseudoElement() in the case where there already was a
::first-letter pseudo element, since we detached it right before
AttachLayoutTree. This means we may have different results for
FirstLetterTextLayoutObject() at those two points if the method has
bugs.
Such a bug was found here where a list item marker inside an anonymous
block had an out-of-flow sibling, but we were not able to traverse out
of the anonymous block to find the first letter text.
Add a DCHECK to let us trigger cases where FirstLetterTextLayoutObject()
is buggy, but also add if-tests to make sure we don't crash.
Bug: 977044
Change-Id: Id80f8aad5c04678d64927e7a7ca4d60cdb6a4b8e
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1672845
Reviewed-by: Morten Stenshorne <mstensho@chromium.org>
Commit-Queue: Rune Lillesveen <futhark@chromium.org>
Cr-Commit-Position: refs/heads/master@{#671629}
--
wpt-commits: 53ebe42e282c9cd4c49d8e615b30eee06c2211a0
wpt-pr: 17447
Automatic update from web-platform-tests
[shapedetection] Add some shapedetection tests
Cover following two checkpoints:
- If the ImageBitmapSource is an HTMLVideoElement object whose readyState attribute
is either HAVE_NOTHING or HAVE_METADATA then reject the Promise with a new
DOMException whose name is InvalidStateError.
- If the ImageBitmapSource argument is an HTMLCanvasElement whose bitmap’s origin-clean
flag is false, then reject the Promise with a new DOMException whose name is SecurityError.
spec: https://wicg.github.io/shape-detection-api/#image-sources-for-detection
Change-Id: I0522d0340d3cb0291df6be65dfc1ab99037b30f7
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1666766
Reviewed-by: Reilly Grant <reillyg@chromium.org>
Commit-Queue: Wanming Lin <wanming.lin@intel.com>
Cr-Commit-Position: refs/heads/master@{#671578}
--
wpt-commits: ec977dea05df9ce55c03a842ffe05a1bf6ee399a
wpt-pr: 17392
Automatic update from web-platform-tests
[Native File System] Create FileSystemFileWriterImpl to mirror the renderer side
This is the initial cl of the refactor of NativeFileSystemFileHandleImpl to
mirror the split on the renderer side with the handle and the writer as
two separate entities.
It is the initial implementation for FileSystemFileWriterImpl. It includes
only one no-op method for now: close().
This sets the stage for atomic writes as close() will be the explicit
termination of a write, which allows us to do the atomic operation.
Bug: 968550
Change-Id: I0af81efa46789f18c4a8b0bca3ab89c5703b798c
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1662709
Commit-Queue: Olivier Yiptong <oyiptong@chromium.org>
Reviewed-by: Daniel Cheng <dcheng@chromium.org>
Reviewed-by: Marijn Kruisselbrink <mek@chromium.org>
Auto-Submit: Olivier Yiptong <oyiptong@chromium.org>
Cr-Commit-Position: refs/heads/master@{#671539}
--
wpt-commits: 3cb5a99e5521936fb8819de8aaba806050b84184
wpt-pr: 17364
Automatic update from web-platform-tests
Add mandatory capabilities parameter in the body for the webdriver new session command (#17212)
--
wpt-commits: 693543f7296ef79cc4ec534422c5a0bb433992dd
wpt-pr: 17212
Automatic update from web-platform-tests
[css-flexbox] Clamp the cross-size by the min/max size
...when computing the main-axis size of a flex item.
As per this change:
https://drafts.csswg.org/css-flexbox/#change-2018-used-cross-sizeR=dgrogan@chromium.org, eae@chromium.org
Bug: 972576
Change-Id: I11a1ac14e2e0e44e39fcd45d4568874da2dcfce8
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1672014
Auto-Submit: Christian Biesinger <cbiesinger@chromium.org>
Reviewed-by: David Grogan <dgrogan@chromium.org>
Reviewed-by: Emil A Eklund <eae@chromium.org>
Commit-Queue: Christian Biesinger <cbiesinger@chromium.org>
Commit-Queue: Emil A Eklund <eae@chromium.org>
Cr-Commit-Position: refs/heads/master@{#671485}
--
wpt-commits: cc0f6ffd9998bdc180d3c70354656f438d99f701
wpt-pr: 17434
Automatic update from web-platform-tests
Don't skip fragment clip if there are other clips for composited layers
Composited layer under multicol is still not fully correctly
supported. Currently we don't allow fragmentation of composited layers,
and skip fragment clip for them. However, if there are other clips below
the fragment clip, we should not skip fragment clip to ensure the other
clips are correctly applied. This is still not fully correct (see the
comment in the code for details), but the result is less broken in many
cases. Hopefully we'll fully fix the issue in
CompositeAfterPaint+LayoutNG-box-fragments.
Bug: 976187
Change-Id: Ica9acc71f0d1204d6380f1bcac3e2159631d63e9
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1670413
Commit-Queue: Xianzhu Wang <wangxianzhu@chromium.org>
Reviewed-by: Chris Harrelson <chrishtr@chromium.org>
Cr-Commit-Position: refs/heads/master@{#671480}
--
wpt-commits: 1c33536fe6e7d2a6a0323bdb849a5a80c33513a6
wpt-pr: 17429
Automatic update from web-platform-tests
[LayoutNG] Fix |ComputeMinMaxSize| not to clear |NeedsLayout|
This patch fixes |ComputeMinMaxSize| not to clear |NeedsLayout|
when all of following conditions are met:
1. Needs |MinMaxSize|
2. Is an orthogonal flow root.
3. Is legacy layout.
4. Change is applied that dirties the layout.
In this case, |NGBlockNode::ComputeMinMaxSize()| calls
|Layout()|, which goes to |RunLegacyLayout()|, which calls
|LayoutObject::LayoutIfNeeded()|. However, its constraint
space is intermediate that it does not update
|CachedLayoutResult|.
Then when laying it out, because |NeedsLayout()| is cleared,
existing out-of-date |CachedLayoutResult| is used.
This patch fixes:
1. Not to pre-layout orthogonal roots if it has
|CachedLayoutResult|.
2. Set |NeedsLayout()| if |RunLegacyLayout()| clears it, but
|CachedLayoutResult| was not updated.
Bug: 976859
Change-Id: Ic2d979a3c4f9479040bfd0e54cf4283b60ed52b5
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1670146
Reviewed-by: Christian Biesinger <cbiesinger@chromium.org>
Reviewed-by: Emil A Eklund <eae@chromium.org>
Commit-Queue: Emil A Eklund <eae@chromium.org>
Cr-Commit-Position: refs/heads/master@{#671460}
--
wpt-commits: 47b08c15495e05fa73e243736742851109198af1
wpt-pr: 17428
Automatic update from web-platform-tests
Remove unused BroadcastChannel in portal-activate-data.html.
It was probably left behind during recent WPT refactoring.
Change-Id: Ia993654248303c56559cb8cf755d11944213048a
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1671947
Commit-Queue: Jeremy Roman <jbroman@chromium.org>
Commit-Queue: Adithya Srinivasan <adithyas@chromium.org>
Auto-Submit: Jeremy Roman <jbroman@chromium.org>
Reviewed-by: Adithya Srinivasan <adithyas@chromium.org>
Cr-Commit-Position: refs/heads/master@{#671414}
--
wpt-commits: 674c9f5b6c076746d8823ee215181b9b471cd1a4
wpt-pr: 17431
Automatic update from web-platform-tests
[docs] Deprecate some requirement flags (#15964)
Reference: https://github.com/web-platform-tests/wpt/issues/5293
Also update the example to use non-deprecated flags.
--
wpt-commits: a56df4a639b5e1c278037d53ce461f66e06c902a
wpt-pr: 15964
Automatic update from web-platform-tests
[docs] Reorganize information on Ahem
WPT maintains a document dedicated to configuring systems to run the
tests. Relocate the instructions on installing Ahem to this document,
and reference them from their original location (an explanation of how
to use the font to author tests).
--
wpt-commits: 49e92561e87e4e1dd7a798dc74f69378b4526878
wpt-pr: 16339
Automatic update from web-platform-tests
[docs] Enable Sphinx's "nit-picky" mode
The `-n` flag to Sphinx is documented as "warn about all missing
references." Enable that flag and correct the missing references which
were previously tolerated.
--
[docs] Incorporate wptserve documentation
While Sphinx is capable of including the contents of files which exist
outside of the documentation root (via the `include` directive), it
interprets such text in the context of the including file. That makes
the mechanism inappropriate for including documentation that references
still other files using relative paths.
To address this, create a symbolic link to the `wptserve` source
directory. Fix or ignore infractions in the newly-incorporated files.
--
wpt-commits: 929bb86fff798cac3b94af7a2549480fac40a84e, a54dccec285341423d3e956bcbdbaac5b0e45ab8
wpt-pr: 17418
Automatic update from web-platform-tests
Worker: Deflake worker-interception.https.html
This is a follow-up CL for
https://chromium-review.googlesource.com/c/chromium/src/+/1657855/10#message-de50523b6c1fc520cf1a7c2edf3033d4e2139103
worker-interception.https.html is flaky on Firefox. It looks like service worker
unregistration is unexpectedly canceled. Each test case doesn't explicitly
terminate (shared) workers, so they may prolong the lifetime of the registration
being unregistered, and the next test case may unexpectedly reuse it. This
results in failures in setup_service_worker().
To fix it, this CL separates service worker scope per test. Note that explicit
worker termination should not work in this case because shared workers are
out-of-process workers and it's not possible to terminate them
deterministically.
Bug: n/a
Change-Id: I16c32d6477c8b1a4d1a246b33a8e21747b4fec66
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1663459
Reviewed-by: Matt Falkenhagen <falken@chromium.org>
Commit-Queue: Hiroki Nakagawa <nhiroki@chromium.org>
Cr-Commit-Position: refs/heads/master@{#670831}
--
wpt-commits: 8ad6f6a2ebb5b861f17c1b8084b264c71e6b965b
wpt-pr: 17395
Automatic update from web-platform-tests
Optimize cubic bezier solver.
Newton's method requires an accurate guess to get good convergence.
This patch introduces a linear interpolation of a crude spline as a
starting point for the iterative solver. Previously the starting point
t = x was used. The patch also introduces a tighter tolerance for
Newton's method than the fallback, and is justified by Newton's method
having quadratic convergence characteristics. Speed testing indicates
that the solver is roughly 50% faster, since the number of iterations
is typically less even at the higher accuracy requirement.
Bug: 827560
Change-Id: I89e3cc9061823269aabb16697e2d167751aa52b1
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1643973
Reviewed-by: Ian Vollick <vollick@chromium.org>
Reviewed-by: Robert Flack <flackr@chromium.org>
Reviewed-by: Xida Chen <xidachen@chromium.org>
Commit-Queue: Kevin Ellis <kevers@chromium.org>
Cr-Commit-Position: refs/heads/master@{#670995}
--
wpt-commits: dbe15f82985f2e2a978255665f388c7dcdd425b1
wpt-pr: 17349
Automatic update from web-platform-tests
Expect actionsWithKeyPressed.html to fail on Mac
Related to issue #17296. This PR does not fix the underlying issue, but
only unblocks the infrastructure tests on macOS.
--
Expect elementPosition.html to timeout
For Chrome on macOS
--
wpt-commits: 2e5b9fb08c42966831690bd01b9b6410efc479fa, fe2232a673785d77c1b23509c5ae52c325378918
wpt-pr: 17415
Automatic update from web-platform-tests
[Navigation Timing] Replace TAO restrictions with same-origin ones
The NavigationTiming specification changed [1] to replace TAO based opt-in
to get information about unloadStart, unloadEnd and redirectCount with
same origin restrictions.
Our implementation was not aligned with the old spec language, but
isn't aligned with the new one either.
This CL aligns the implementation with the new language.
Change-Id: Ic4f3c7061e7b9a40f960f0bf26ef26823d5fe743
Bug: 977147
[1] https://github.com/w3c/navigation-timing/pull/108
Change-Id: Ic4f3c7061e7b9a40f960f0bf26ef26823d5fe743
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1669355
Commit-Queue: Yoav Weiss <yoavweiss@chromium.org>
Reviewed-by: Nicolás Peña Moreno <npm@chromium.org>
Cr-Commit-Position: refs/heads/master@{#671229}
--
wpt-commits: 94f51553f1413cf2118b9bf89762bfad16d8a18a
wpt-pr: 17412
Automatic update from web-platform-tests
Service worker: self.serviceWorker (#17285)
* Tests for self.serviceWorker
* Don't need `self`
* Add read only test
* Adding comments & test the object doesn't change
* Swap order
--
wpt-commits: 3725067ef0328c998be2ba93dbaeb0579586fccd
wpt-pr: 17285
Automatic update from web-platform-tests
[webnfc] Remove incorrect tests (#17425)
Per spec https://heycam.github.io/webidl/#es-DOMString,
non string value will be converted to string via `toString()`
method, therefore there's no `TypeError` to be thrown.
--
wpt-commits: 3a00a57af5f449a72f0e1b89dae6f1bd15ad8fcd
wpt-pr: 17425
Automatic update from web-platform-tests
Worker: Remove duplicate service worker interception tests for shared workers
worker-interception.https.html has more test coverage, and is well-formatted.
Bug: n/a
Change-Id: I29640ffa781c1b90b1faa56eab4f771c1a3c0a84
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1669137
Auto-Submit: Hiroki Nakagawa <nhiroki@chromium.org>
Commit-Queue: Matt Falkenhagen <falken@chromium.org>
Reviewed-by: Matt Falkenhagen <falken@chromium.org>
Cr-Commit-Position: refs/heads/master@{#671183}
--
wpt-commits: 982f12dc86098e5b32ffec455520a5eb114c27af
wpt-pr: 17414
Automatic update from web-platform-tests
[docs] Increase strictness of build process (#17401)
Use the `-W` flag to interpret Sphinx warnings as errors, causing the
build to fail in the presence of either. This makes various authoring
mistakes more visible. Perhaps most importantly: it highlights instances
of "orphaned" content.
Fix all details which were previously reported as warnings and thus
permitted.
--
wpt-commits: 9c6304b71c5ee090ebc1eef02ab32b70444d9704
wpt-pr: 17401
Automatic update from web-platform-tests
Allow var() references in -internal-visited-* properties.
Previously, -internal-visited-* properties didn't get any parsing behavior,
because they were assumed to be used during style application only. In
other words, it was assumed that the parsing would happen _before_ e.g.
color would expand to -internal-visited-color.
This is (of course) not true if the value has a var() reference, since
the actual parsing then happens computed-value time.
To fix this, this CL gives all -internal-visited-* properties parsing
behavior.
The "pending substitution value" adds complexity here, since the value
itself contains the shorthand it came from. This means that if e.g. the
border shorthand is parsed with a value that contains a var() reference,
the corresponding -internal-border-*-color property will point to the
same pending substitution value, that contains the "border" shorthand.
When resolving that pending substitution value computed-value time,
we expand the shorthand into its longhands, which of course are unaware
of the -internal-visited-* context. To fix this, this CL adds a way to
get the unvisited counterpart from a visited property, and uses the
unvisited CSSPropertyID as they key for the pending substitution cache.
Bug: 974665, 976913
Change-Id: I1a168bfe188eaccad7feac6c065f8e447ee56dbe
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1669290
Reviewed-by: Rune Lillesveen <futhark@chromium.org>
Commit-Queue: Anders Hartvoll Ruud <andruud@chromium.org>
Cr-Commit-Position: refs/heads/master@{#670896}
--
wpt-commits: be93a1b93e08be1bd161208aaa6ab662c1c2d1c1
wpt-pr: 17411
Automatic update from web-platform-tests
WebSocket: stop removing an iframe in onerror from crashing
There was an issue that if the iframe owning a WebSocket was
synchronously removed in the onerror handler then it would crash the
render process. This was caused by unconditionally calling Close()
without checking the current state.
Fix it.
Also add a web platform test.
Bug: 974667
Change-Id: Iaeb8f96cf600a2c052f06b5b59f4444b02615e83
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1663681
Reviewed-by: Yutaka Hirano <yhirano@chromium.org>
Commit-Queue: Adam Rice <ricea@chromium.org>
Cr-Commit-Position: refs/heads/master@{#670061}
--
Set a long timeout
--
wpt-commits: 433e1f06259a79c6cb8bf695bc03ffdfdfe4e7e7, 28ad65497bccb301707e341a9d36f9d144e2793e
wpt-pr: 17380
Automatic update from web-platform-tests
Implement basic LargestContentfulPaint API
This CL adds a basic implementation of the LargestContentfulPaint API.
The implementation is done behind a flag, and is missing attribution
information and special handling for images.
Explainer: https://github.com/WICG/LargestContentfulPaint
Intent to Implement:
https://groups.google.com/a/chromium.org/forum/#!topic/blink-dev/WVqgYtyaGJk
Bug: 965505
Change-Id: Ic29604349dc60de2808bdef3df4684e81e1845d7
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1646634
Commit-Queue: Nicolás Peña Moreno <npm@chromium.org>
Reviewed-by: Chris Harrelson <chrishtr@chromium.org>
Cr-Commit-Position: refs/heads/master@{#670911}
--
wpt-commits: 93ba864d402d941a2670681d535e33723e5296c0
wpt-pr: 17214
Automatic update from web-platform-tests
Enable PauseExecutionContextOnBackgroundFreeze
Thie change enables (experimentally) pausing the execution context
when a document is frozen.
The WPT tests written required that promise resolution continued after
the page was frozen. This isn't the case when
PauseExecutionContextOnBackgroundFreeze is enabled so rewrite the tests
to poll for the correct state.
BUG=957589
Change-Id: If104235fd7281d88b5d86a3f0b89337828e7f4e7
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1662633
Reviewed-by: Shubhie Panicker <panicker@chromium.org>
Reviewed-by: Alexander Timin <altimin@chromium.org>
Commit-Queue: Dave Tapuska <dtapuska@chromium.org>
Cr-Commit-Position: refs/heads/master@{#670883}
--
wpt-commits: 81c649c98aa73ac14416491271ed001ffe9e4bbe
wpt-pr: 17363
Automatic update from web-platform-tests
[Azure] Fix logging params
This makes the logging behaviour consistent across all jobs on Azure:
info level tbpl logging to stdout, which is also consistent with
Taskcluster setup.
Notably, the previous configuration for "all tests: Edge" accidentally
specified --log-tbpl twice, causing duplicate log outputs, which is
fixed in this commit.
--
Disable version check of EdgeDriver
--
wpt-commits: 7dd31d2b2f9ebca522b50af958c18208d08683ae, 5303a4e841969e0b5991c45c276027a671e3ba20
wpt-pr: 17404
Automatic update from web-platform-tests
Refactor SRI script tests to prepare for Preload + SRI
This CL refactors the SRI script tests and what script resources they
consume to remove duplication, and factor out + generalize some of the
helpers, such as the script URL prefixes.
Before this CL, there were multiple script files whose contents were
irrelevant to the tests. The scripts would be served differently due
to `.headers` files. This CL consolidates all of the script resources
into a single `script.js` that the test runner can use while demanding
specific headers for tests. This file and factored-out URL prefixes in
`sri-test-helpers.sub.js` will be used by Preload + SRI tests, so this
CL makes writing those tests cleaner.
Bug: 677022
Change-Id: Ie6cc312f9d81de4d5f75925810b7ce88a61910a5
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1667326
Reviewed-by: Yutaka Hirano <yhirano@chromium.org>
Reviewed-by: Kouhei Ueno <kouhei@chromium.org>
Commit-Queue: Dominic Farolino <dom@chromium.org>
Cr-Commit-Position: refs/heads/master@{#670809}
--
wpt-commits: 1dd0bc9ffd9b1433cb2aa1e20f972fd477c65ad4
wpt-pr: 17394
Automatic update from web-platform-tests
location.hash = '' should always result in an empty fragment, not no-fragment
So far in |setHash|, old fragment identifier and new fragment identifier are
compared ignoring nullity.
When fragment has never been set, |old_fragment_identifier| is null.
In this situation, |location.hash = ''| is called,
|new_fragment_identifier| is not null empty value.
|EqualIgnoringNullity| treats these as the same, |setHash| does nothing.
So, this patch compare old and new fragment by == operator
not |EqualIgnoringNullity|.
Bug: 615300
Change-Id: Id509bc5066eeee83236ecb4ee816ecb90540a260
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1647298
Commit-Queue: Daniel Cheng <dcheng@chromium.org>
Reviewed-by: Daniel Cheng <dcheng@chromium.org>
Cr-Commit-Position: refs/heads/master@{#670795}
--
wpt-commits: a3e67aed3509dc5255a029f913c5785be4e67aed
wpt-pr: 17293
Automatic update from web-platform-tests
[LayoutNG] Fix inserting exclusion crash.
When placing a new area into the closed area list there is almost
always an area at LayoutUnit::Min().
However if there was a float placed at LayoutUnit::Min() it is possible
to get into a state where this isn't the case.
Bug: 972904
Change-Id: I1c787ea8e4f5d87ae97023efd9cdb994c2b065ce
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1664930
Commit-Queue: Ian Kilpatrick <ikilpatrick@chromium.org>
Reviewed-by: Morten Stenshorne <mstensho@chromium.org>
Cr-Commit-Position: refs/heads/master@{#670726}
--
wpt-commits: 9a183439c7fd1a22298c49a86be1e293d460ea75
wpt-pr: 17391
Automatic update from web-platform-tests
SMS API: Clean up SMS API mojom interface
This CL cleans up the SMS API mojom interface based on the suggestions
on the following issue link. No behavioral changes for the SMS API
mojom interface.
Bug: 968584
Change-Id: Iba707dd9a94f75cb28b7409f930e75b0f6f2f792
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1654272
Commit-Queue: Jun Cai <juncai@chromium.org>
Reviewed-by: Reilly Grant <reillyg@chromium.org>
Reviewed-by: Tom Sepez <tsepez@chromium.org>
Reviewed-by: Kinuko Yasuda <kinuko@chromium.org>
Reviewed-by: Philip Jägenstedt <foolip@chromium.org>
Cr-Commit-Position: refs/heads/master@{#670694}
--
wpt-commits: 43ef8451989ca9be29d55ca88cb80a1ab90030a1
wpt-pr: 17304
Automatic update from web-platform-tests
[LayoutNG] Fix DCHECK(NeedsLayout) for inline oof fixed container.
NGOutOfFlowLayoutDescendant.inline_container was not being set
correctly for an Element if:
- There is an OOF containing block between Element and
its Container(). This can only happens if Element is
position:fixed.
This caused Element not to be laid out after it got dirty (but
its containing block did not).
The fix is to set inline_container when Element's
NGOutOfFlowDesecendantCandidate gets propagated up the OOF
containing block chain.
For a while, I was afraid that this fundamentally broke
the concept of inline_container. I could not come up
with a counterexample that broke this fix.
Bug: 974760
Change-Id: Id16a0057f0aefe183c30c53244dd5c46108f093c
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1666809
Commit-Queue: Aleks Totic <atotic@chromium.org>
Reviewed-by: Emil A Eklund <eae@chromium.org>
Reviewed-by: Koji Ishii <kojii@chromium.org>
Cr-Commit-Position: refs/heads/master@{#670634}
--
wpt-commits: b6343fc17a3035a0c2170f44fb63df2356aa2cb9
wpt-pr: 17390
Automatic update from web-platform-tests
[css-flexbox] Also skip non-flexbox items with percentages
The same percentage issue happens with regular blocks. See the
linked bug and associated testcase.
R=dgrogan@chromium.org, eae@chromium.org
Bug: 973924
Change-Id: I588353f0dcce4a446c08a8aa0f186c9ec2ee8949
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1667684
Commit-Queue: Christian Biesinger <cbiesinger@chromium.org>
Auto-Submit: Christian Biesinger <cbiesinger@chromium.org>
Reviewed-by: David Grogan <dgrogan@chromium.org>
Cr-Commit-Position: refs/heads/master@{#670606}
--
wpt-commits: 6515e798b30a5d3053bec1faae2f8c366fdbcba9
wpt-pr: 17396
Automatic update from web-platform-tests
[css-ui] Test serialization: `-webkit-appearance` (#17221)
* [css-ui] Test serialization: `-webkit-appearance`
* fixup! [css-ui] Test serialization: `-webkit-appearance`
Change API used to set style in order to actually test the internals
referenced by the test title.
* fixup! [css-ui] Test serialization: `-webkit-appearance`
Introduce tests for property reference
* fixup! [css-ui] Test serialization: `-webkit-appearance`
Remove duplicate assertion
* fixup! [css-ui] Test serialization: `-webkit-appearance`
Comprehensive tests for property references have been added since these
tests were written. Remove the excess assertions in order to improve the
focus of these tests.
* fixup! [css-ui] Test serialization: `-webkit-appearance`
Add equivalent tests for unprefixed `appearance` property
* fixup! [css-ui] Test serialization: `-webkit-appearance`
--
wpt-commits: 79787d6d6eff255fbe6efdbe386e92363d2b6303
wpt-pr: 17221
Automatic update from web-platform-tests
Adding a test for backdrop filter plus mask
This tests a backdrop filter element that also contains a (non-border
rect) mask or clip. It currently fails.
Bug: 974857
Change-Id: Ied65cc96deb2d02634b62d8743bff02588c94411
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1662499
Commit-Queue: Chris Harrelson <chrishtr@chromium.org>
Reviewed-by: Chris Harrelson <chrishtr@chromium.org>
Auto-Submit: Mason Freed <masonfreed@chromium.org>
Cr-Commit-Position: refs/heads/master@{#670587}
--
wpt-commits: 1ace0eaddd369c7063e0a34ac91a2f361020b5bf
wpt-pr: 17366
Automatic update from web-platform-tests
Return show promise instead of the resolved value of show promise (#17389)
--
wpt-commits: e0cfe65d6e5338f501dfe0494e6ce2cc666ee978
wpt-pr: 17389
Automatic update from web-platform-tests
Worker: Add service worker interception tests for shared workers
One test case for shared workers is failing because it's not allowed to create a
dedicated worker on a shared worker.
Bug: n/a
Change-Id: I4f7139adacdccb0f16d7afedd7eaab065c4bea2c
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1657855
Commit-Queue: Hiroki Nakagawa <nhiroki@chromium.org>
Reviewed-by: Matt Falkenhagen <falken@chromium.org>
Cr-Commit-Position: refs/heads/master@{#669615}
--
wpt-commits: 3671eb9b321c60898a79c76d20a97807925ebe1e
wpt-pr: 17333
Automatic update from web-platform-tests
Initial implementation of std-toast
Intent to Implement: https://groups.google.com/a/chromium.org/forum/?utm_source=digest&utm_medium=email#!topic/blink-dev/Gl7FIKM5IFw
BUG=972945
Follow-ups will align with std-switch in elements and de-duplicate the .eslintrc.js files between toast and kv-storage.
Change-Id: Ib5837ccf8c9ea03fd2f1f2bd581e443a2936f693
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1635867
Commit-Queue: Jack Steinberg <jacksteinberg@chromium.org>
Commit-Queue: Fergal Daly <fergal@chromium.org>
Reviewed-by: Kouhei Ueno <kouhei@chromium.org>
Cr-Commit-Position: refs/heads/master@{#670383}
--
wpt-commits: 17de974a80b0d9a6ff48dc24796441057afad5a0
wpt-pr: 17146
Automatic update from web-platform-tests
Removing initial Element Timing implicit registration heuristic.
The Element Timing specification originally included a provision for reporting
timings of elements that were automatically detected as 'important'. This would
allow useful information from the Element Timing API to be reported through the
Performance Timeline even if site authors neglected to mark their important
elements with the 'elementtiming' attribute.
The automatic detection of 'important' elements is, necessarily, a heuristic
and, until consensus for the correct heuristic had been reached, we were
treating any image that occupied greater than %15 of the viewport as
'automatically important'. As the Element Timing specification is still being
refined, the exact threshold and heuristic is still being investigated. Until
agreement on an appropriate heuristic is reached, implicit registration will be
disabled.
This CL removes our %15 threshold and disables implicit registration. There was
also WPT coverage to test that all browsers would respect the %15 threshold so
that test is being removed too.
Change-Id: I14847533402f5a68bb3de63a8f9acbd004c5a8e4
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1643379
Reviewed-by: Chris Harrelson <chrishtr@chromium.org>
Reviewed-by: Nicolás Peña Moreno <npm@chromium.org>
Commit-Queue: Tom McKee <tommckee@chromium.org>
Cr-Commit-Position: refs/heads/master@{#670319}
--
wpt-commits: f76b2e5e2b1dd5d73cf2351c57a303d4a67eb729
wpt-pr: 17176
Automatic update from web-platform-tests
[html] Add test for history traversal (#16478)
* [html] Add test for history traversal
* fixup! [html] Add test for history traversal
* fixup! [html] Add test for history traversal
--
wpt-commits: 7d8eefc2e4793df438d1f120737078c0bb67979e
wpt-pr: 16478
Automatic update from web-platform-tests
[IndexedDB] Promote test isolation (#16436)
* [IndexedDB] Promote test isolation
Use distinct database names for each test so that failure to clean up
from one test does not influence the result of another.
* fixup! [IndexedDB] Promote test isolation
--
wpt-commits: 23776c16a2db7b8df7374ec7c1e97a0cec003280
wpt-pr: 16436
Automatic update from web-platform-tests
[cookie-store] Promote stability in test results (#16415)
* [cookie-store] Promote stability in test results
Allow service worker activation to proceed in the absence of the API
under test. This prevents the workers in failing implementations from
entering the "redundant" state, which can interfere with the harness's
ability to detect all tests.
* fixup! [cookie-store] Promote stability in test results
* fixup! [cookie-store] Promote stability in test results
--
wpt-commits: 199b5fa5c003e348490000bc8147e0f2bb3fbaf0
wpt-pr: 16415
Automatic update from web-platform-tests
[csp] Wait for all tests to be defined (#15631)
Configure the test harness to delay completion until all
asynchronously-defined tests are available.
--
wpt-commits: 159c7c2f98576d2288c12aee86264356cbaa1be1
wpt-pr: 15631
Automatic update from web-platform-tests
Prevent leaking Sec-CH-/Sec-Fetch- Request Headers on HTTPS Downgrade Redirects.
Currently various Sec-CH- and Sec-Fetch- request headers are added through the codebase and correctly check if the request target url is trustworthy. If the destination is not trustworthy then the headers are not added. However in the event a request redirects from a trustworthy to non-trustworthy destination, such as https => http, there may be pre-existing headers present. Since these headers would not have been added to the non-trustworthy request we need to remove these extra headers from the request.
This change adds a helper call to the UrlLoader class in the network service to remove any request headers prefixed with "sec-ch-" or "sec-fetch-" when a downgrade redirect is detected. In addition to this unit test cases for the helper and end to end WPT cases have been added to validate the scenario and prevent future regressions.
Bug: 964053
Change-Id: I109c3ec1b3a05f6341c3c4adbd1a8da1274fd0d6
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1647354
Commit-Queue: Brandon Maslen <brandm@microsoft.com>
Reviewed-by: Matt Menke <mmenke@chromium.org>
Reviewed-by: Mike West <mkwst@chromium.org>
Cr-Commit-Position: refs/heads/master@{#667415}
--
wpt-commits: 376e2a8250c3824b794ec518ab320d8a82cedf95
wpt-pr: 17200
Automatic update from web-platform-tests
[LayoutNG] Skip "replaced % resolution" quirk on ortho-roots.
This is about an undocumented quirk. I raised
https://github.com/whatwg/quirks/issues/46 for it.
LayoutNG currently cannot handle this situation, because we don't store
the inline-size to use in replaced percentage resolution (only the
block-size). However, I find it hard to believe that there'd be any
specific expectations to how this is supposed to work if a table cell
has a definite block-size, and a descendant establishes an orthogonal
writing-mode root. So just ignore it and use regular percentage size
resolution in these cases.
Bug: 967069
Change-Id: Iac61abe80f4f949017cc161855b2223ec4503812
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1664075
Reviewed-by: Ian Kilpatrick <ikilpatrick@chromium.org>
Reviewed-by: Christian Biesinger <cbiesinger@chromium.org>
Commit-Queue: Morten Stenshorne <mstensho@chromium.org>
Cr-Commit-Position: refs/heads/master@{#670126}
--
wpt-commits: 361912dbe4ef692ea256b6f5bc0d74c0668ccdf5
wpt-pr: 17381
Automatic update from web-platform-tests
Integrating ScrollTimeline with main thread WebAnimations.
Changes made:
- ScrollTimeline interface is enabled under its own ScrollTimeline flag.
- Animation class now has member variable timeline_ of type AnimationTimeline instead of
DocumentTimeline.
- Document and scroll linked animations are now attached to their document global document
timeline. TODO: reconsider this approach.
Bug: 916117
Change-Id: I5f0c3d22fdb0c9b9763f5a2400c2e3d5af32e896
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1597286
Commit-Queue: Olga Gerchikov <gerchiko@microsoft.com>
Reviewed-by: Kevin Ellis <kevers@chromium.org>
Reviewed-by: Emil A Eklund <eae@chromium.org>
Reviewed-by: Andrey Kosyakov <caseq@chromium.org>
Reviewed-by: Stephen McGruer <smcgruer@chromium.org>
Reviewed-by: Majid Valipour <majidvp@chromium.org>
Cr-Commit-Position: refs/heads/master@{#670086}
--
wpt-commits: 31373e2726073d95b05e09ed112efac950b562b5
wpt-pr: 16755
Automatic update from web-platform-tests
CSP handling of 'self' inside sandbox iframe was incorrect.
The correct handling of 'self' depended on the timing of when the
ExecutionContext's sandbox flags were applied. If a CSP policy for 'self'
was applied after the origin was derived to be opaque it would fail.
BUG=973991
Change-Id: Ie6f384b02aa3dea3ce48c3052f06b7bd8fb2d3ea
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1658764
Commit-Queue: Dave Tapuska <dtapuska@chromium.org>
Reviewed-by: Mike West <mkwst@chromium.org>
Reviewed-by: Tom Sepez <tsepez@chromium.org>
Cr-Commit-Position: refs/heads/master@{#670074}
--
wpt-commits: b50766de4c9059ed41559315e0f12d602c8c9d59
wpt-pr: 17326
Automatic update from web-platform-tests
[Azure Pipelines] Use more verbose TBPL logging for full runs (#17379)
This is to help diagnose problems if they occur, such as this failure
to run all of the tests:
https://github.com/web-platform-tests/wpt/pull/17219#issuecomment-501272482
--
wpt-commits: 3857b199e7ee7867550b268df05664e2ef055600
wpt-pr: 17379
Automatic update from web-platform-tests
Add CEReactions and CustomElementCallbacks to toggleAttribute
Previously the IDL definition for toggleAttribute doesn't have
those modifiers, so they behave differently than setAttribute when
called on custom elements (it won't fire attributeChangedCallback, etc)
Bug: 976177
Change-Id: I7d8107db8d07a2e2ec856305c56a178339c5e05c
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1663881
Commit-Queue: Rakina Zata Amni <rakina@chromium.org>
Commit-Queue: Kent Tamura <tkent@chromium.org>
Reviewed-by: Kent Tamura <tkent@chromium.org>
Auto-Submit: Rakina Zata Amni <rakina@chromium.org>
Cr-Commit-Position: refs/heads/master@{#670060}
--
wpt-commits: f6881fbb28940a254944e62b994b282b176bb030
wpt-pr: 17377
Automatic update from web-platform-tests
[LayoutNG] Fix NeedsLayout failure when CR/FF is dirty
This patch fixes |ClearNeedsLayout| for CR (U+000D) and FF
(U+000C). These were overlooked when |ClearNeedsLayout| was
moved from |NGInlineItemsBuilder| to |NGLineBreaker| and
|NGInlineLayoutAlgorithm|.
Bug: 971811
Change-Id: Ie3d9d3d11e35a7ed3458bb8b96d2fa68bf9c37a9
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1663478
Reviewed-by: Xiaocheng Hu <xiaochengh@chromium.org>
Commit-Queue: Koji Ishii <kojii@chromium.org>
Cr-Commit-Position: refs/heads/master@{#670010}
--
wpt-commits: 95372aa3ca35565b56fc6d9d5e0653a4fff8a6e8
wpt-pr: 17373
Automatic update from web-platform-tests
[LayoutNG] Fix DCHECK failure for table cell width quirk
This patch fixes a DCHECK failure when table cell width
calculation quirk logic[1] found an out-of-flow image.
The logic was in |BreakLine| because it may need to change
|can_break_after| of the last item. By calling
|HandleAtomicInline| in trailing item phase, the logic can
be moved to |HandleAtomicInline|.
With this change, |HandleAtomicInline| may choose to break
before the item, similar to |HandleText|.
[1] https://quirks.spec.whatwg.org/#the-table-cell-width-calculation-quirk
Bug: 968317
Change-Id: Ib4a71ca62107d2601449ba589bb2771b378ab38b
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1660296
Reviewed-by: Morten Stenshorne <mstensho@chromium.org>
Reviewed-by: Emil A Eklund <eae@chromium.org>
Commit-Queue: Koji Ishii <kojii@chromium.org>
Cr-Commit-Position: refs/heads/master@{#669968}
--
wpt-commits: 2b29767a9f53e3a7a142b6c4cf72407dfe88506d
wpt-pr: 17357
Automatic update from web-platform-tests
[LayoutNG] Fix "body fills html" quirks.
We had a few issues related to quirky body elements. We were
stretching the body when it was OOF-positioned, float, or an atomic
inline when we weren't supposed to.
Bug: 967215
Change-Id: I640b1a2f6c2ab30d071e04c245a34dd1e30b19fd
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1662839
Reviewed-by: Christian Biesinger <cbiesinger@chromium.org>
Reviewed-by: Aleks Totic <atotic@chromium.org>
Commit-Queue: Ian Kilpatrick <ikilpatrick@chromium.org>
Cr-Commit-Position: refs/heads/master@{#669954}
--
wpt-commits: 90c18764adad351a28069adbfade18d6dd4949e7
wpt-pr: 17367
Automatic update from web-platform-tests
[docs] Build (but do not deploy) for pull requests
Although the documentation website should not be rebuilt in response to
pull requests, refusing to simply build the site may potentially hide
errors during the review process.
Update the documentation generation script to build the site as part of
the pull request, provided a relevant file has been modified. With this
applied, if the patch under review regresses the script, the contributor
and reviewers are immediately notified.
--
wpt-commits: 4be1e7e7b78712303076819ef8a6b3f055ebbb3c
wpt-pr: 17365
Automatic update from web-platform-tests
Link to testdriver docs from the main docs site (#17362)
* Link to testdriver docs from the main docs site
`testdriver.md` and `testdriver-tutorial.md` aren't correctly indexed anywhere
because of https://github.com/bocoup/wpt-docs/issues/1 .
This PR manually adds them to the ToC in "Writing Tests" as per user requests.
* fixup! sorted
--
wpt-commits: da3b6e4dfafe250800c490cee949b2b405364ecd
wpt-pr: 17362
Automatic update from web-platform-tests
[sms] Move SMS Receiver API to Navigator
This change moves the SMSReceiver API under navigator.
Details on its usage can be found here[1]. Implementation
of the AbortController will come in a following CL.
[1] https://github.com/samuelgoto/sms-receiver#sms-receiver-api-1
Bug: 960546
Change-Id: Iba9882db05e0d153d6f37e1481d6237738542949
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1626018
Reviewed-by: Kentaro Hara <haraken@chromium.org>
Reviewed-by: Reilly Grant <reillyg@chromium.org>
Commit-Queue: Ayu Ishii <ayui@chromium.org>
Cr-Commit-Position: refs/heads/master@{#669779}
--
wpt-commits: 94bff4ec536738245d29502e40516a6509db5c74
wpt-pr: 17049
Automatic update from web-platform-tests
[ElementTiming] Add intersectionRect for text entries
This CL adds support for intersectionRect for PerformanceElementTiming
entries referring to text elements. The rect is different from the one
used by Largest Contentful Paint because it does not get projected to
the top-level frame, as that is not acceptable from security/privacy
perspective. However the rect is only computed when needed.
Bug: 942033, 969174
Change-Id: I1f09288b17a73c84daf8cc55996b78e44739a042
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1658630
Reviewed-by: Chris Harrelson <chrishtr@chromium.org>
Reviewed-by: Liquan (Max) Gu <maxlg@chromium.org>
Commit-Queue: Nicolás Peña Moreno <npm@chromium.org>
Cr-Commit-Position: refs/heads/master@{#669777}
--
wpt-commits: ca62107b234fb0aa8d6044e47c915be3fe42981f
wpt-pr: 17328
Automatic update from web-platform-tests
[LayoutNG] Fix negative margins on inline boxes
This patch fixes computing size of inline boxes that have
negative inline margins.
It looks like we didn't have enough test coverage when
`NGFragmentGeometry` was introduced.
Bug: 971622
Change-Id: Icd0f5300ef76a873ec25b8d96756884da954d184
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1662249
Reviewed-by: Aleks Totic <atotic@chromium.org>
Commit-Queue: Koji Ishii <kojii@chromium.org>
Cr-Commit-Position: refs/heads/master@{#669699}
--
wpt-commits: c81dc02d6b1b4a5a5925d6cb27f16af7b897cb59
wpt-pr: 17356
Automatic update from web-platform-tests
[LayoutNG] Fixed crash issue list-style-type change
The crash reason is that the NGInlineItem is obsolete after change
list-style-type. This patch added ListStyleTypeChanged to make sure
marker is fully updated and re_CollectInlines for marker.
Bug: 966750
Change-Id: I8c51bc90d31344fb36627fe29608b2b576d85228
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1660399
Commit-Queue: cathie chen <cathiechen@igalia.com>
Reviewed-by: Koji Ishii <kojii@chromium.org>
Cr-Commit-Position: refs/heads/master@{#669613}
--
wpt-commits: dbaf4c77b6bf7ffc8d030a40f98045e8832ba4de
wpt-pr: 17352
Automatic update from web-platform-tests
Revert "Ignore lint failures until we import a rev with them fixed" (#17289)
This reverts https://github.com/web-platform-tests/wpt/pull/13396.
These exact lines are also in the "Duplicate filename where there's no
actual merging" section of the file.
The change came from https://hg.mozilla.org/mozilla-central/rev/f23487d9ab99d486fe3ab7f261e4d4e0bf1172ae
where those lines were present, although they were present in WPT when
that change was upstreamed.
--
wpt-commits: 4a075835b37e1cdd90094e4766d4c9842fe64480
wpt-pr: 17289
Automatic update from web-platform-tests
Reland tests from "Change ICE connection state on transceiver changes"
This reverts the rest of commit 2229f484784131f7c655c650b389cfd78fe3e739.
Reason for revert: Fixed the reason the test was flaky.
Original change's description:
> Revert "Change ICE connection state on transceiver changes"
>
> This reverts commit 0882bebe06b38aa5d4e155a5cebc11e7af08abd2.
>
> Reason for revert: Causing flaky failures of
> external/wpt/webrtc/RTCPeerConnection-iceConnectionState.https.html
> on Linux Tests and WebKit Linux Trusty Leak
>
> https://ci.chromium.org/p/chromium/builders/ci/Linux%20Tests/78387
> https://ci.chromium.org/p/chromium/builders/ci/WebKit%20Linux%20Trusty%20Leak/34892
>
> Original change's description:
> > Change ICE connection state on transceiver changes
> >
> > This ensures that if the PC iceConnectionState should change
> > because unused transports are discarded, the state is updated.
> >
> > Bug: chromium:966798
> >
> > Change-Id: I09d945f5e70eec813f33c3131fbe889825613652
> > Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1632254
> > Reviewed-by: Henrik Boström <hbos@chromium.org>
> > Commit-Queue: Harald Alvestrand <hta@chromium.org>
> > Cr-Commit-Position: refs/heads/master@{#664206}
>
> TBR=hta@chromium.org,hbos@chromium.org
>
> Change-Id: I2cbae9f1327e7aacb7022d2a823ce1e954c91b62
> No-Presubmit: true
> No-Tree-Checks: true
> No-Try: true
> Bug: chromium:966798
> Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1635208
> Reviewed-by: Aaron Gable <agable@chromium.org>
> Commit-Queue: Aaron Gable <agable@chromium.org>
> Cr-Commit-Position: refs/heads/master@{#664388}
Bug: chromium:966798
Change-Id: I2e9887bc62942393a2d39042da9bfc85bbd98c91
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1636049
Commit-Queue: Harald Alvestrand <hta@chromium.org>
Reviewed-by: Henrik Boström <hbos@chromium.org>
Cr-Commit-Position: refs/heads/master@{#668750}
--
wpt-commits: 8a474834dc86e8713b1f7e966d6d5aa8fd69c9a6
wpt-pr: 17082
Automatic update from web-platform-tests
Add tests for movablelimits (#17354)
Tests taken from Webkit.
--
wpt-commits: 79cf5fb8d18ef01ba174de2ebf841f782f342095
wpt-pr: 17354
Automatic update from web-platform-tests
Convert Long Tasks API idlharness test to use `idl_test` (#17344)
Because the setup function is allowed to fail, this makes the setup a
bit simpler.
This is to make it easier to add 'hr-time' as a dependency for
https://github.com/web-platform-tests/wpt/pull/17305.
--
wpt-commits: e5167a1ffdd432f1fadd721d63fd1cf42bd67e96
wpt-pr: 17344
Automatic update from web-platform-tests
Don't round hit test location in HitTestLocation::RectForPoint
HitTestLocation rounds the hit test location to the closest 1x1 rect
for no good reason, causing some confusing hit test results. Hence,
this patch removes such rounding.
Bug: 974314
Change-Id: I48266a7ec8bffbeb746085c5d8835d3f300dd4b1
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1661297
Reviewed-by: Emil A Eklund <eae@chromium.org>
Reviewed-by: Philip Rogers <pdr@chromium.org>
Commit-Queue: Xiaocheng Hu <xiaochengh@chromium.org>
Cr-Commit-Position: refs/heads/master@{#669579}
--
wpt-commits: 5689f6b102fa8055740ff66dfe6b8c3b9a808c08
wpt-pr: 17351
Automatic update from web-platform-tests
[webnfc] Introduce IDLs for NFCErrorEvent and NFCErrorEventInit.
This CL introduces IDLs for NFCErrorEvent and NFCErrorEventInit
interfaces, and updates the baseline of relevant wpt tests.
http://w3c.github.io/web-nfc/#dom-nfcerrorevent
The corresponding spec changes were introduced in
https://github.com/w3c/web-nfc/pull/184
BUG=520391
Change-Id: Ia8af8b49f53be7161b7e7b0f4e45f617c4991473
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1652968
Commit-Queue: Leon Han <leon.han@intel.com>
Reviewed-by: Leon Han <leon.han@intel.com>
Reviewed-by: Reilly Grant <reillyg@chromium.org>
Reviewed-by: Rijubrata Bhaumik <rijubrata.bhaumik@intel.com>
Reviewed-by: Kentaro Hara <haraken@chromium.org>
Cr-Commit-Position: refs/heads/master@{#669482}
--
wpt-commits: 4100d1d51b885185c50902397f7d203d89efc24c
wpt-pr: 17263
Automatic update from web-platform-tests
[Azure Pipelines] avoid parenthesis in job names
The Azure Pipelines integration with GitHub Checks produces checks on
the form "$pipelineName ($jobName)", leading to check names like
"Azure Pipelines (tools/wptrunner/ unittests (macOS))". Use colons as
separator instead, like the existing "wpt.fyi hook" jobs.
Also change some step names for consistency.
--
wpt-commits: e75855649360298870a40ed78874bd4ab1470c20
wpt-pr: 17265
Automatic update from web-platform-tests
Add more direction tests (#17345)
* Add more direction tests
Put these in a new directory.
* Make direction.html use testharness.js
--
wpt-commits: fdf8ff01960e031fe0381133f5a39e1b8e372454
wpt-pr: 17345
Automatic update from web-platform-tests
[trusted types] Check text node insertions into <script> elements.
insertAdjacentText into <script> elements can be used to bypass
Trusted Types checks. This change fixed that hole by adding TT checks
to related Node API methods, and changing insertAdjacentText to use
the public methods.
See additional discussion on the root issue in
https://github.com/WICG/trusted-types/issues/133
Bug: 739170
Change-Id: Ie26b2aca2d60c7e14a0118c6fd957f32928afabf
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1547746
Reviewed-by: Mason Freed <masonfreed@chromium.org>
Commit-Queue: Daniel Vogelheim <vogelheim@chromium.org>
Cr-Commit-Position: refs/heads/master@{#669210}
--
wpt-commits: 90a794e8e93a5e22e7cb3b2f8c79d751989fa7b0
wpt-pr: 17230
Liquan(Max) Gu
Xiaocheng Hu
Tab Atkins Jr
joanmarie
Jean-Francois Paradis
Domenic Denicola
Ian Clelland
Ian Kilpatrick
Robert Ma
Jeremy Roman
Koji Ishii
Mike Pennisi
autofoolip
Majid Valipour
Findit
Sam Sneddon
Josh Matthews
Frédéric Wang
Mustapha Jaber
moz-wptsync-bot
Ovidio Henriquez
Vladimir Levin
Nicolás Peña Moreno
Stephen McGruer
Hiroki Nakagawa
Eric Willigers
Lan Wei
Eric Willigers
Tsuyoshi Horo
Ms2ger
David Grogan
Kagami Sascha Rosylight
George Steel
Jinfeng Ma
Dominic Farolino
cathiechen
Yutaka Hirano
Eero Häkkinen
Henrik Boström
Lan Wei
Bryan McQuade
Blink WPT Bot
Donna Wu
Xianzhu Wang
Xida Chen
Ella Ge
Javier Fernandez
Gérard Talbot
Luke Bjerring
L. David Baron
Giovanni Ortuño Urquidi
Alex Cooper
Chris Harrelson
Emily Hanley
Timothy Gu
Morten Stenshorne
Daniel Vogelheim
Suman Nelson Kancherla
Alice Boxhall
Florian Rivoal
Oliver Dunk
Alexey Shvayka
aardgoose
Anne van Kesteren
Anders Hartvoll Ruud
Aleks Totic
Jack Steinberg
Robert Flack
Simon Pieters
Sergey Poromov
Oriol Brufau
Dale Curtis
Mustaq Ahmed
Takashi Toyoshima
Reilly Grant
Fernando Serboncini
Joshua Bell
Mason Freed
Domenic Denicola
Makoto Shimazu
Rob Buis
Andreea Pavel
pyup-bot
Yoav Weiss
Piotr Bialecki
Emil A Eklund
Mugdha Lakhani
Yuki Shiino
Raymond Toy
Dan Beam
Dave Tapuska
Philip Jägenstedt
Emilio Cobos Álvarez
Rune Lillesveen
Wanming Lin
Olivier Yiptong
George Roman
Christian Biesinger
Kevin Ellis
Jake Archibald
Adam Rice
Jihwan Marc Kim
Jun Cai
Danyao Wang
Tom McKee
Brandon Maslen
Olga Gerchikov
Rakina Zata Amni
Kouhei Ueno
Ayu Ishii
Harald Alvestrand
kaixinjxq
Leon Han