Automatic update from web-platform-tests
Support sibling-index() in number interpolations
Use the CssToLengthConversionData in StyleResolverState to compute the
value in MaybeConvertValue() and invalidate using a new
TreeCountingChecker for conversion checking if the sibling-count() or
sibling-index() changes.
Bug: 405883453
Change-Id: I598d131e85d8a2e1f7745a34f3b9356cbbea1e82
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/6508866
Reviewed-by: Robert Flack <flackr@chromium.org>
Commit-Queue: Rune Lillesveen <futhark@chromium.org>
Cr-Commit-Position: refs/heads/main@{#1456304}
--
wpt-commits: 3b4afced7ecf45c9cb809041d3b0d8400cbb589f
wpt-pr: 52359
Differential Revision: https://phabricator.services.mozilla.com/D248856
Automatic update from web-platform-tests
[vt] Track animations that are pending a DOM callback after being skipped
Previously calling startViewTransition followed by a second
startViewTransition would correctly skip the first and wait on
its DOM callback to be triggered; however, since skipping a
transition removes it from the active list, starting a VT after an
explicit skip could result in the DOM callback of the first getting
lumped in with the second. This patch adds tracking of skipped
transitions with a pending DOM callback. These are considered
"active" from the perspective of blocking the replacement
transition's capture phase.
Bug: 414396338
Change-Id: Ife841566239e735db46f6006f2391d02d9cf183d
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/6513530
Commit-Queue: Kevin Ellis <kevers@chromium.org>
Reviewed-by: Steve Kobes <skobes@chromium.org>
Cr-Commit-Position: refs/heads/main@{#1456270}
--
wpt-commits: 8471745461ac2e9f057f13145c6f9b9f2aa87f32
wpt-pr: 52358
Differential Revision: https://phabricator.services.mozilla.com/D248855
Automatic update from web-platform-tests
Add test for xml serialization of attribute nodes
Signed-off-by: Simon Wülker <simon.wuelker@arcor.de>
--
wpt-commits: ef1952b4ac2e2b4a0c2b01e4df1e97d029edd4ad
wpt-pr: 52354
Differential Revision: https://phabricator.services.mozilla.com/D248854
Automatic update from web-platform-tests
Fix MapLocalToAncestor() & co for block fragmentation.
Get rid of the dependency on LayoutFlowThread and the other legacy
multicol classes for this.
Fix it so that LayoutBox::PhysicalLocation() returns the offset of the
first fragment relatively to the first fragment of its containing block.
No behavior changes unless content is inside block fragmentation. For
block fragmentation, this will affect the behavior of APIs such as
offsetLeft and offsetTop. Previously, when inside block fragmentation,
the offset would be in the internal "stitched" flow thread coordinate
space, where we pretend that all fragmented content is laid out into a
single tall strip. Then we'd convert this to real visual offsets with
help from the flow thread, in some cases, where in others (if both the
element and its offsetParent were inside block fragmentation), the
"stitched" flow thread coordinates would be exposed through APIs such as
offsetLeft and offsetTop.
Apart from this correctness fix, with this CL we're much closer to being
able to remove LayoutFlowThread & co, which are legacy layout objects
that were needed by the legacy layout engine, (and was still required by
the LayoutNG engine, due to this offset calculation stuff, but not by
actual layout).
In order to calculate the offset of the first fragment relatively to the
first fragment of its containing block inside block fragmentation, we
first need to actually finish fragmented layout, since the first
fragment of an element may not be in the same fragmentainer as the first
fragment of its containing block. In case of out-of-flow positioning an
element may even start in an earlier fragmentainer than its containing
block. The code for this machinery is in LayoutBoxUtils. Some of the
logic is copied from the BlockNode code that dealt with this (the
BlockNode parts will be removed when this feature eventually is soldered
to the engine - for now add DCHECKs to make sure that the code is
unreached).
LayoutTableColumn needed some additional fixing. Since table columns
don't generate fragments, there's some code in there to synthesize such
fragments. The new code is based on the logic in
LayoutTableColumn::DeprecatedLocationInternal() (which by the way had
bugs for vertical-rl writing-mode, which are fixed in the new code),
including some checks that never seem to trigger (add TODO).
Prefix old location functions in LayoutBox with "Deprecated", and add
some DCHECKs to ensure that they aren't reached. Also added such DCHECK
to a bunch of LayoutFlowThread, LayoutMultiColumnFlowThread,
LayoutMultiColumnSet functions.
This is behind a runtime-enabled feature flag called
LayoutBoxVisualLocation. For now set it to "experimental". Will flip it
to stable once the feature is complete. Since this CL affects the
behavior of offsetLeft and offsetTop when inside fragmentation, this is
an opportunity to get this to make more sense overall. First of all,
offsetWidth and offsetHeight should be a union of all fragments. It
still uses the stitched block-size. It's not 100% clear how this should
work exactly. See https://github.com/w3c/csswg-drafts/issues/11541
But it seems reasonable that we should be able to walk up the
offsetParent chain from one element `elm`, adding offsetLeft and
offsetTop at each step on the way, and then end up with the same result
as if using `elm.getBoundingClientRect()` (also for width and height).
Will follow up on this.
Many tests that deal with offsetLeft and offsetTop inside block
fragmentation have to be updated because of this change. Also add some
new tests for the improved getClientRects() behavior for table columns
and also for content that overflows the fragmentainers (due to being
tall and monolithic).
Add another PrintContextTest for something that has never worked before:
handling root element margins correctly for PDF URL fragment target
rectangles (this is a corner case all right).
Remove some unit tests (or parts) that are no longer valid because of
this change.
Bug: 40855022
Fuchsia-Binary-Size: shut up.
Change-Id: I0e1fc6cf7717b74b904c825e519ee4694faf87f7
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/6507711
Reviewed-by: Ian Kilpatrick <ikilpatrick@chromium.org>
Commit-Queue: Morten Stenshorne <mstensho@chromium.org>
Cr-Commit-Position: refs/heads/main@{#1456237}
--
wpt-commits: 0a5a7abb801d347601814cd6ffaa1f85ecaa434c
wpt-pr: 52350
Differential Revision: https://phabricator.services.mozilla.com/D248852
Automatic update from web-platform-tests
Fix WPT by correcting spin button coordinates
The test
`input-events-spin-button-click-on-number-input-delete-document.html`
was failing because the calculated coordinates for clicking the spin-up
button were incorrect, causing the pointer events to miss the target.
The coordinates have been adjusted to accurately target the spin-up
button, ensuring the test now passes consistently.
Bug: 366783813
Change-Id: Iccb6dbc3e09b8e92f34d77accf09d409b0963248
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/6497956
Commit-Queue: Tanu Jain <tanujain@microsoft.com>
Reviewed-by: Utkarsh Pathak <utpathak@microsoft.com>
Cr-Commit-Position: refs/heads/main@{#1456100}
--
wpt-commits: 16a6ce620e84ed527800524cf34cb72f6af31bfe
wpt-pr: 52340
Differential Revision: https://phabricator.services.mozilla.com/D248845
Automatic update from web-platform-tests
Printing test for :root block-start margin taller than page.
This is already working in Blink, but I was about to introduce a
crasher, and luckily discovered it in time, even though there was no
test coverage for it (but now there is!)
Change-Id: I60d43d17d67fd1c1da21c692914dd3b6cbae5709
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/6509313
Commit-Queue: Ian Kilpatrick <ikilpatrick@chromium.org>
Reviewed-by: Ian Kilpatrick <ikilpatrick@chromium.org>
Cr-Commit-Position: refs/heads/main@{#1455817}
--
wpt-commits: 44ab3987ef82a45f729f6b1eeb2f12c1a831356b
wpt-pr: 52344
Differential Revision: https://phabricator.services.mozilla.com/D248842
Automatic update from web-platform-tests
cc/vt: Another speculative fix for view-transition-types-mutable-no-document-element-crashtest.html
This patch does the following:
- Wait for the compositor to be ready before running a VT test.
This passes with 500 runs (100 iterations / 5 each).
R=kevers@chromium.org
Bug: 413305967
Change-Id: Ibe37db724777d885e812f1fd2c6f3d41a1c861b5
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/6506166
Commit-Queue: Vladimir Levin <vmpstr@chromium.org>
Reviewed-by: Kevin Ellis <kevers@chromium.org>
Cr-Commit-Position: refs/heads/main@{#1455716}
--
wpt-commits: d8b209ae839eadb200ffd2ccd60bdcd23c652ccc
wpt-pr: 52339
Differential Revision: https://phabricator.services.mozilla.com/D248841
Automatic update from web-platform-tests
ScopedVT: Add checks for scope elements that are not rendered.
(1) If the scope element has no ComputedStyle, do not create a pseudo
element tree. This can happen if the scope element is inside the light
DOM of a shadow host, but does not appear in the flat tree.
(2) If the scope element has no LayoutObject, don't look for tagged
participants inside it. This happens if scope has display:none style.
Tag discovery uses paint order, so it requires a PaintLayer.
Bug: 412265459
Change-Id: I848cd438b043dbdba35a66177ba7ed6256e6cc44
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/6505605
Reviewed-by: Kevin Ellis <kevers@chromium.org>
Commit-Queue: Steve Kobes <skobes@chromium.org>
Cr-Commit-Position: refs/heads/main@{#1455654}
--
wpt-commits: c11e6a428b9f95b1d514cda540e5c190aba4606f
wpt-pr: 52338
Differential Revision: https://phabricator.services.mozilla.com/D248835
Automatic update from web-platform-tests
[carousel] Fix crashing on ::scroll-button without scroller activation
First, put scroll button in disabled state, when it's created on a
non-scroller originating element.
Second, don't do anything when receiving events on such scroll button.
For now, the second part is needed due to the incorrect ability of
a scroll button to be focusable while disabled, this will be fixed
soon, but the second part is still needed for future proof, as stated
in the comment in the code.
Fixed: 415199295
Change-Id: I9440f025e0ce321442ea4b612d9455034ddd49a8
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/6508371
Reviewed-by: Rune Lillesveen <futhark@chromium.org>
Commit-Queue: Daniil Sakhapov <sakhapov@chromium.org>
Cr-Commit-Position: refs/heads/main@{#1455635}
--
wpt-commits: 75b6a093948d9867c2b1110ec34750317db52cf5
wpt-pr: 52336
Differential Revision: https://phabricator.services.mozilla.com/D248832
Automatic update from web-platform-tests
Fix a DCHECK in incremental style updates.
We would not always update the ComputedStyle HasAuthor*() flags
when they needed to go from unset to set. This could have caused
wrong appearance to be chosen.
Fixed: 40934009
Change-Id: I3358db8b43ce2ad1bf7d255190514f2706586f54
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/6494523
Reviewed-by: Anders Hartvoll Ruud <andruud@chromium.org>
Auto-Submit: Steinar H Gunderson <sesse@chromium.org>
Commit-Queue: Anders Hartvoll Ruud <andruud@chromium.org>
Cr-Commit-Position: refs/heads/main@{#1455561}
--
wpt-commits: 892e915a5a07e3e7b2436437e8ab8b4736b24f9b
wpt-pr: 52328
Differential Revision: https://phabricator.services.mozilla.com/D248824
Automatic update from web-platform-tests
Added tests for Ignore Letter Spacing for Cursive Scripts
--
Added tests for Ignore Letter Spacing for Cursive Scripts
--
Added tests for Ignore Letter Spacing for Cursive Scripts
--
incorporating comments
--
wpt-commits: 6eca21500200996e302e373dd51367bbff74dc2a, f692284a6ff8c2376e334dafd0b2b41a641bfe20, de3951770281eca0079b2e4846d880edd861e1c0, 4464aa3a97c6b99415bf96a66e22b4ff452ff0f4
wpt-pr: 52296
Differential Revision: https://phabricator.services.mozilla.com/D248820
Automatic update from web-platform-tests
Implement Trusted types document write sinks
Implements the Document.write algorithm covering
Trusted HTML.
Part of #36258
Signed-off-by: Tim van der Lippe <tvanderlippe@gmail.com>
--
wpt-commits: d3287f10104c970fb7216606a93ace0cf7c16867
wpt-pr: 52318
Differential Revision: https://phabricator.services.mozilla.com/D248819
Automatic update from web-platform-tests
Fix a test to declare line-height with Ahem. (#52313)
This fixes the test align-self-static-position-005.html and its reference
from recent PR https://github.com/web-platform-tests/wpt/pull/52260 to declare
a line-height of `1` when specifying the Ahem font, per best practices at
https://web-platform-tests.org/writing-tests/ahem.html
Per that documentation: when using Ahem in a WPT, "An explicit (i.e., not
normal) line-height should also always be used")
This is needed in order for the test to pass in Firefox, since Gecko uses a
larger-than-1 value as the "normal" line-height for Ahem.)
--
wpt-commits: dd8bc79937cc33e720381c68b37af6254e301e8a
wpt-pr: 52313
Differential Revision: https://phabricator.services.mozilla.com/D248816
Automatic update from web-platform-tests
Allow <input> in <select> with wrapper tags
This patch makes <input> get parsed as expected in <select> if there is
another tag in between the <select> and <input>, such as
<select><div><input>, in order to ensure compatibility with the legacy
behavior where <input> adds a closing </select> tag.
Bug: 402429384
Change-Id: Iff21245d0555844aa636eb002866bb3ed5af4d42
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/6435161
Reviewed-by: Mason Freed <masonf@chromium.org>
Commit-Queue: Joey Arhar <jarhar@chromium.org>
Cr-Commit-Position: refs/heads/main@{#1455251}
--
wpt-commits: 598e8867a9abe53512f2baffad1192d554c69ced
wpt-pr: 52314
Differential Revision: https://phabricator.services.mozilla.com/D248814
Automatic update from web-platform-tests
Check CSP for `javascript:` URLs
Also update a WPT test to fail-fast if the iframe incorrectly
evaluates the `eval`. Before, it would run into a timeout if
the implementation is correct. Now we reject the promise
when an exception is thrown.
Requires servo/rust-content-security-policy#6
Signed-off-by: Tim van der Lippe <tvanderlippe@gmail.com>
--
wpt-commits: 79b8e2d8223eb930cd9a3c21bace3a365f5eafd7
wpt-pr: 52193
Differential Revision: https://phabricator.services.mozilla.com/D248813
