add new tests for graphics-aria spec #38845
Closed
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
cookiecrook
commented
Mar 23, 2023
Bug: 1399631, 1413556 Change-Id: Id7f437d698928c5e62dddde07d849d3f4cc1e059 Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/4310688 Reviewed-by: Anders Hartvoll Ruud <[email protected]> Commit-Queue: Joey Arhar <[email protected]> Cr-Commit-Position: refs/heads/main@{#1120773}
…through a disabled form-associated custom element. Differential Revision: https://phabricator.services.mozilla.com/D173176 bugzilla-url: https://bugzilla.mozilla.org/show_bug.cgi?id=1818287 gecko-commit: a4ec639e1547c3d1e670b112dc1fcccbe1f8d0bf gecko-reviewers: edgar
Depends on D173068 Differential Revision: https://phabricator.services.mozilla.com/D173069 bugzilla-url: https://bugzilla.mozilla.org/show_bug.cgi?id=1823464 gecko-commit: 77479cf002be5addc60e3c39dca8457f71c4c21a gecko-reviewers: webdriver-reviewers, jdescottes
Differential Revision: https://phabricator.services.mozilla.com/D170580 bugzilla-url: https://bugzilla.mozilla.org/show_bug.cgi?id=1815937 gecko-commit: ac2fee8163f3edcab58cb66aaac4e7f801d40bb2 gecko-reviewers: smaug, devtools-reviewers, ochameau
…nt-size, for predictable whole-number baselines and sizing. Before this patch, Firefox fails this test's expectations on Linux, with a 1px sizing discrepancy between the input element in a vertical writing-mode vs. horizontal. That discrepancy is almost certainly due to the difference in baselines (central vs. ideographic) in vertical vs. horizontal writing-modes, and the fact that pixel-snapping that baseline might produce a +/-1px difference in block-position and block-size of the line, inside the form-field. This patch changes the test to use the Ahem font with a font-size of 30px, which should have a whole-pixel-valued central as well as ideographic baseline; so no pixel-snapping should be necessary, and the input element's size should be be consistent, just as the test expects it to be. This patch also makes one non-functional change, to explicitly declare the charset as UTF-8, per WPT best-practices from http://web-platform-tests.org/writing-tests/general-guidelines.html#character-encoding Differential Revision: https://phabricator.services.mozilla.com/D173191 bugzilla-url: https://bugzilla.mozilla.org/show_bug.cgi?id=1823516 gecko-commit: 563cd419882564f85e437550dc03fe0e3d29f261 gecko-reviewers: ntim
… if needed Depends on D172491 Differential Revision: https://phabricator.services.mozilla.com/D172492 bugzilla-url: https://bugzilla.mozilla.org/show_bug.cgi?id=1804086 gecko-commit: 0b351d3ec6c6d197c083ad30b0ed2b7a03bd6bba gecko-reviewers: smaug
Bug: 1371522 Change-Id: Ie60036576dc1d0bf6d8f945f34eab4ed0d3a6933 Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/4064159 Commit-Queue: Adithya Srinivasan <[email protected]> Reviewed-by: Jeremy Roman <[email protected]> Cr-Commit-Position: refs/heads/main@{#1120810}
…s preempted. This allows us to assert that NavigationApi::AbortOngoingNavigation() is only called when ongoing_navigate_event_ is non-null. This requires a slight reordering of the early-exit cases in NavigationApi::DispatchNavigateEvent(), but guarantees that ongoing_navigation_ only gets set non-null if a navigate event will definitely fire. Change-Id: If1213c9fd2500adf99427ea5115733f99ad49ea2 Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/4355034 Reviewed-by: Domenic Denicola <[email protected]> Commit-Queue: Nate Chapin <[email protected]> Auto-Submit: Nate Chapin <[email protected]> Cr-Commit-Position: refs/heads/main@{#1120859}
unanimatable-properties.html tests that a few properties aren't animatable, but CSSTransitionDiscrete aims to make those animatable so this test is invalid. There is one property in this test which should remain unanimatable: transition-duration. In order to account for this, I've created a new WPT which tests that transition-duration, transition-delay, and transition-property are still not animatable. Bug: 1399631, 1413556 Change-Id: I091d358796559951436c7d63a181646af09472f4 Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/4349871 Reviewed-by: Robert Flack <[email protected]> Commit-Queue: Joey Arhar <[email protected]> Cr-Commit-Position: refs/heads/main@{#1120916}
This CL clears the <selectmenu>'s selection when the <selectmenu>'s value is changed to an invalid value. This brings <selectmenu>'s behaviour inline with <select>'s behaviour. BUG=1426875 TEST=selectmenu-value-selectedOption.tentative.html Change-Id: Idb22f47e97152cb2320d78fde786b898aaf294a4 Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/4362638 Reviewed-by: Joey Arhar <[email protected]> Reviewed-by: Stephen McGruer <[email protected]> Commit-Queue: Peter Kotwicz <[email protected]> Cr-Commit-Position: refs/heads/main@{#1120919}
Bug: 1371395 Change-Id: I57ea751df7264de50a518011c65b2eb78a67378d Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/4364324 Commit-Queue: Mingyu Lei <[email protected]> Auto-Submit: Yuzu Saijo <[email protected]> Reviewed-by: Mingyu Lei <[email protected]> Cr-Commit-Position: refs/heads/main@{#1121007}
Bumps [types-requests](https://github.com/python/typeshed) from 2.28.11.5 to 2.28.11.16. - [Release notes](https://github.com/python/typeshed/releases) - [Commits](https://github.com/python/typeshed/commits) --- updated-dependencies: - dependency-name: types-requests dependency-type: direct:production update-type: version-update:semver-patch ... Signed-off-by: dependabot[bot] <[email protected]>
Bumps [sphinx-argparse](https://github.com/ashb/sphinx-argparse) from 0.3.1 to 0.4.0. - [Release notes](https://github.com/ashb/sphinx-argparse/releases) - [Changelog](https://github.com/ashb/sphinx-argparse/blob/main/docs/changelog.rst) - [Commits](sphinx-doc/sphinx-argparse@0.3.1...v0.4.0) --- updated-dependencies: - dependency-name: sphinx-argparse dependency-type: direct:production update-type: version-update:semver-minor ... Signed-off-by: dependabot[bot] <[email protected]>
Differential Revision: https://phabricator.services.mozilla.com/D173107 bugzilla-url: https://bugzilla.mozilla.org/show_bug.cgi?id=1823586 gecko-commit: 2f9380a15086f5015e3b5ed303d1cc9556e09e4a gecko-reviewers: alwu
Differential Revision: https://phabricator.services.mozilla.com/D172700 bugzilla-url: https://bugzilla.mozilla.org/show_bug.cgi?id=1822521 gecko-commit: f215a7c1a760af42bd850696c641a13555cd5d56 gecko-reviewers: emilio
…dy-completed animations into the timeline's sampling-order list. We have an invariant that the mAnimationOrder LinkedList is a subset of the mAnimations hashset (omitting any animations that are hidden due to content-visibility). This patch corrects one case where we were incorrectly inserting an animation into the linked list when it wasn't present in the hashset (because the animation had completed). This patch also adds some documentation to mention this invariant, and some assertions to enforce it in several places. Differential Revision: https://phabricator.services.mozilla.com/D173333 bugzilla-url: https://bugzilla.mozilla.org/show_bug.cgi?id=1822907 gecko-commit: 74d3d5d35e7767882de82d40ca7aa4f3d85b488a gecko-reviewers: hiro
Also remove our internal .ini files that are unnecessarily generous in terms of fuzzy threshold and pass/fail allowance. Differential Revision: https://phabricator.services.mozilla.com/D173372 bugzilla-url: https://bugzilla.mozilla.org/show_bug.cgi?id=1823851 gecko-commit: f9805c7f23bf451130ffbba1f92d2ac95373d7b9 gecko-reviewers: layout-reviewers, TYLin
…web-platform-tests#38723) WebKitGTK tests have started to timeout on the CI again. The issue is caused because there is a 2-3 hour maximum time allowed for each test chunk to run and we are hitting the time limit. In the past we tried to fix this by raisin this timeout or the number of chunks See commits: ac16b46 and 6f7c2fd and a218d10 This time instead of rising the time limit or the number of chunks let's try to run the tests in parallel. Using 4 jobs gives 4x speed-up even on a machine that only has 1 core because we avoid waiting serially for tests that are slow or timeout. For more details see: web-platform-tests#38723 (comment)
Previously the promise resolution steps would only happen when Agent::PerformMicrotaskCheckpoint was called. This was incorrect because there were other microtask checkpoints that either call into the EventLoop directly or use a v8::MicrotaskScope block to run tasks. Ensure all calls to running a microtask checkpoint end up calling to the Agent::RejectPromises callback. This code is behind an enabled feature flag for safety. Bug: 1027230 Change-Id: I557e9ad092eed51e8e1ae20b589f71f5ddc27b5f Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/4355716 Commit-Queue: Dave Tapuska <[email protected]> Reviewed-by: Scott Haseley <[email protected]> Cr-Commit-Position: refs/heads/main@{#1121123}
As per spec change <ray-size> is now optional, defaulting to <closest-side>. Spec: https://drafts.fxtf.org/motion/#valdef-ray-ray-size Change-Id: I739ead2552528b5fdb755a2b1ab13bc4c4510b5b Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/4360295 Reviewed-by: Anders Hartvoll Ruud <[email protected]> Commit-Queue: Daniil Sakhapov <[email protected]> Cr-Commit-Position: refs/heads/main@{#1121144}
This is done per the resolution in w3c/csswg-drafts#7797 (comment) This is a direct revert of commit https://crrev.com/f4b3a78fc4 which explicitly dropped input while rendering was blocked for a view transition capture. Reverting the above CL indirectly implements the desired behavior from the CSSWG resolution. The reason is that the UA stylesheet gives the ::view-transition pseudo width/height: 100% which is viewport filling. Pseudo elements already route input events to their owning DOM element, in this case the document :root. Authors cannot change this as their pseudo styles get applied only once the view transition animation starts, see [1]. [1] https://source.chromium.org/chromium/chromium/src/+/main:third_party/blink/renderer/core/view_transition/view_transition_style_tracker.cc;l=1116;drc=f599ac6c54b827d30f79a8f15bf1e2661af8925a Bug: 1424821,1368757 Change-Id: I4a0efa5180c69225cb34661c5cea3e8aba95a4d1 Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/4362639 Reviewed-by: Ian Vollick <[email protected]> Commit-Queue: David Bokan <[email protected]> Reviewed-by: Khushal Sagar <[email protected]> Cr-Commit-Position: refs/heads/main@{#1121175}
…latform-tests#38946) If an element's painting exceeds viewport bounds, clip it down to max texture size or twice the viewport bounds by painting the subset which is closest to the snapshot root. This involves the following 2 changes: 1) A new ClipNode is introduced to paint the desired subrect when rendering this element in paint, raster and compositing. All these stages should render only this subset. 2) The object-view-box on the replaced element is still set to the ink overflow rect. So at paint time, we adjust the replaced content rect to the actual subset that will be painted based on the clipping above. [email protected], [email protected], [email protected] Bug: 1409713 Change-Id: If81bc67ad6ffe49d68e4470a6717b704edafcd3f Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/4330193 Reviewed-by: Xianzhu Wang <[email protected]> Commit-Queue: Khushal Sagar <[email protected]> Reviewed-by: Philip Rogers <[email protected]> Auto-Submit: Khushal Sagar <[email protected]> Cr-Commit-Position: refs/heads/main@{#1121126} Co-authored-by: Khushal Sagar <[email protected]>
…rs" (web-platform-tests#39114) This reverts commit 80faa9503f457b4a09fa31912eafc7eefd585568. Reason for reland: Performance regression fixed. See diff with PS#1. The original patch used an expensive condition to decide whether to install an AnchorElementObserver, which caused performance issues. This patch changed to installing observer only when the `anchor` attribute (IDL or content) is modified. Original change's description: > Revert "[anchor-position] Make `anchor` attribute work for non-popovers" > > This reverts commit 4abd0d04120561b80ad2e6dd398651d7e909463f. > > Reason for revert: Caused a lot of performance regressions. > > Original change's description: > > [anchor-position] Make `anchor` attribute work for non-popovers > > > > This patch reimplements how we mark elements that are implicit anchors. > > Previously it was only marked for popover element's implicit anchors, > > and now it works for all implicit anchors. This is done by a new class > > AnchorElementObserver that observes all possible changes in implicit > > anchor and replaces the old PopoverAnchorObserver. > > > > This patch also reveals two existing issues: > > 1. Element::GetElementAttribute() may return non-null result even if > > the element is out of tree scope. As fixing the bug is out of the > > scope here, this patch just works around it. crbug.com/1425215 was > > filed. > > 2. popover-anchor-idl-property.html has some bugs that made it pass > > previously. This patch fixes those bugs. > > > > > > Fixed: 1417346 > > Change-Id: I798977ab1ed7df0c528b7eb98c64fc04476e6106 > > Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/4338487 > > Commit-Queue: Xiaocheng Hu <[email protected]> > > Reviewed-by: Mason Freed <[email protected]> > > Cr-Commit-Position: refs/heads/main@{#1118933} > > Bug: 1425610, 1425635, 1425650 > Change-Id: I50a650f011e097c48c0efd110cf765d0df5835e4 > Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/4353985 > Bot-Commit: Rubber Stamper <[email protected]> > Reviewed-by: Mason Freed <[email protected]> > Commit-Queue: Xiaocheng Hu <[email protected]> > Cr-Commit-Position: refs/heads/main@{#1119483} Bug: 1425610, 1425635, 1425650 Change-Id: I222e0cc302e16f208314641b153ce1746ca814a2 Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/4355591 Reviewed-by: Mason Freed <[email protected]> Commit-Queue: Xiaocheng Hu <[email protected]> Cr-Commit-Position: refs/heads/main@{#1121203} Co-authored-by: Xiaocheng Hu <[email protected]>
ViewTransitions have a callback that didn't have its parent task ID properly set, as part of TaskAttribution. This CL fixes that, and ensures they are properly tracked. (and hence, that SoftNavigationHeuristics properly work with ViewTransitions) Change-Id: Ia531017fb9e86efde93b2307f2e4a2d8e7c994ce Bug: 1405448 Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/4350628 Commit-Queue: Yoav Weiss <[email protected]> Reviewed-by: Scott Haseley <[email protected]> Reviewed-by: Khushal Sagar <[email protected]> Cr-Commit-Position: refs/heads/main@{#1121132}
Table column merging is only a good idea if there are adjacent columns with nothing "interesting" taking place between them, such as table cells beginning. Otherwise we'd get unfair space distribution in some cases. Mark the first column of a spanned cell as non-mergeable (so that space can be distributed to it). If there are collapsed columns (visibility:collapse), though, mark the first non-collapsed column instead. Bug: 1430449 Change-Id: I0a9a5caf62f0a77ec57d45c9bc4b59552522c52a Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/4401357 Reviewed-by: Ian Kilpatrick <[email protected]> Commit-Queue: Morten Stenshorne <[email protected]> Cr-Commit-Position: refs/heads/main@{#1126636}
This extend that test coverage to offscreen and worker. Change-Id: I82726eef623c88dd95ccacacfacb584622546e1f Bug: 1428652 Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/4386063 Reviewed-by: Aaron Krajeski <[email protected]> Commit-Queue: Jean-Philippe Gravel <[email protected]> Cr-Commit-Position: refs/heads/main@{#1126679}
Fixed: 1429575 Change-Id: I246067c93bbcb9d384e6b5059602b93afc15a3f5 Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/4394967 Reviewed-by: Robert Flack <[email protected]> Commit-Queue: Anders Hartvoll Ruud <[email protected]> Cr-Commit-Position: refs/heads/main@{#1126771}
Having another read request on an empty chunk sounds nice, but I guess it's not worth allowing recursion for this edge case. Now the next read request will happen asynchronously by the next OnOutputStreamReady callback, which is similar to what Blink does and the spec recommends. * Blink: https://source.chromium.org/chromium/chromium/src/+/main:third_party/blink/renderer/core/fetch/readable_stream_bytes_consumer.cc;l=179-186;drc=059796845b1738dbf28ea76f0e9ff4b6f8787d3a (queues a microtask to prevent recursion) * Spec: https://streams.spec.whatwg.org/#readablestreamdefaultreader-read-all-bytes (See the note below "chunk steps") Differential Revision: https://phabricator.services.mozilla.com/D173813 bugzilla-url: https://bugzilla.mozilla.org/show_bug.cgi?id=1823429 gecko-commit: 34e9465f50b6cb26b9b73e357922b10cb4528c5b gecko-reviewers: smaug
This reverts commit 5a426f9babceea8ffa93d870690634371a79457b. Reason for revert: Caused crbug.com/1430953 Original change's description: > Nested calc() expressions need to consume all their tokens > > When parsing a nested calc() expression that had a valid first part > (term) that was followed by an invalid operator the parsed part was > returned to the caller - which had already consumed the tokens within > the sub-expression. The caller only checked that an expression node was > returned, which meant that the sub-expression was considered valid. > > Require that all tokens that are part of the sub-expression have been > consumed after parsing has completed to fail the parsing in such a case. > > Fixed: 1430413 > Change-Id: I3072e22e2a7b81296510173266814a18fb6bfc37 > Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/4400956 > Auto-Submit: Fredrik Söderquist <[email protected]> > Reviewed-by: Xiaocheng Hu <[email protected]> > Commit-Queue: Xiaocheng Hu <[email protected]> > Cr-Commit-Position: refs/heads/main@{#1126635} Bug: 1430953,1430413 Change-Id: I866b130615c4328bd4b99716e74e3a045e0f348c No-Presubmit: true No-Tree-Checks: true No-Try: true Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/4404759 Bot-Commit: Rubber Stamper <[email protected]> Auto-Submit: Xiaocheng Hu <[email protected]> Commit-Queue: Rubber Stamper <[email protected]> Cr-Commit-Position: refs/heads/main@{#1126878}
This is implemented by 3rd party storage partitioning: https://github.com/wanderview/quota-storage-partitioning/blob/main/explainer.md#storage-apis ``` Quota: The quota system must manage each partition as a separate bucket for determining how much space is permitted and when it should be cleared. This also includes APIs that give information on quota to the website, such as navigator.storage.estimate(). ``` Bug: 1226469 Change-Id: I848dcb1a0e8d491d69742680dc038a2a29adf633 Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/4396391 Commit-Queue: Arthur Sonzogni <[email protected]> Reviewed-by: Titouan Rigoudy <[email protected]> Cr-Commit-Position: refs/heads/main@{#1127054}
The CSS Font Module Level 5 defines a new value, from-font for font-size-adjust, which allows the browser to automatically decides an aspect value based on the primary font [1]. This change updates existing parsing tests for from-value and adds a new reference test to examine the rendering results as expected. font-size-adjust-013.html is inspired by font-size-adjust-012.html. The tests originated from a WebKit implementation of from-font for font-size-adjust [2]. [1] https://www.w3.org/TR/css-fonts-5/#font-size-adjust-prop [2] WebKit/WebKit#12303
Of the 15 cases here (combinations of text-align and directionality), mozilla-central currently fails 9 and passes only 6. (FTR, a bunch of these tests also currently fail in Chrome, with similar errors to Firefox; only Safari seems to handle them consistently/correctly.) Differential Revision: https://phabricator.services.mozilla.com/D174726 bugzilla-url: https://bugzilla.mozilla.org/show_bug.cgi?id=1712703 gecko-commit: 0c96eb4bdfd1cc63041711cc951350d6c1813836 gecko-reviewers: dholbert
This reverts commit 5f586d44396d70cef3ce480dc49a37d69a17cb1c. Reason for revert: Added test is very flaky: crbug.com/1431180 Original change's description: > Iframe credentialless: Test quota-storage-partitioning > > This is implemented by 3rd party storage partitioning: > https://github.com/wanderview/quota-storage-partitioning/blob/main/explainer.md#storage-apis > ``` > Quota: The quota system must manage each partition as a separate bucket > for determining how much space is permitted and when it should be > cleared. This also includes APIs that give information on quota to the > website, such as navigator.storage.estimate(). > ``` > > Bug: 1226469 > Change-Id: I848dcb1a0e8d491d69742680dc038a2a29adf633 > Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/4396391 > Commit-Queue: Arthur Sonzogni <[email protected]> > Reviewed-by: Titouan Rigoudy <[email protected]> > Cr-Commit-Position: refs/heads/main@{#1127054} Bug: 1226469 Bug: 1431180 Change-Id: Ib0bff9b46d04891ae467bcbe66bf727e1b64d988 No-Presubmit: true No-Tree-Checks: true No-Try: true Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/4403192 Auto-Submit: Andy Paicu <[email protected]> Commit-Queue: Rubber Stamper <[email protected]> Bot-Commit: Rubber Stamper <[email protected]> Cr-Commit-Position: refs/heads/main@{#1127190}
Patch: https://chromium-review.googlesource.com/c/chromium/src/+/4394857 ...introduced a NOTREACHED_NORETURN() to ContainingBlockLogicalWidthForPositioned. Turns out this was reachable! When we compute the static-position from "legacy" within: LayoutNGMixin<Base>::UpdateOutOfFlowBlockLayout However the result of this static-position call shouldn't be needed as this codepath should always require at least one inset set in both axes. Bug: 1431205 Change-Id: I81a463bf9fab8fb2958065bb3e7a51dfea3e82ef Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/4406493 Commit-Queue: Ian Kilpatrick <[email protected]> Reviewed-by: David Grogan <[email protected]> Cr-Commit-Position: refs/heads/main@{#1127321}
…atform-tests#39416) `tools.lint.lint` is unimportable when `git` does not exist in the `PATH`: ``` File "C:\b\s\w\ir\third_party\blink\tools\blinkpy\tool\commands\lint_wpt.py", line 16, in <module> from tools.lint import lint as wptlint File "C:\b\s\w\ir\third_party\wpt_tools\wpt\tools\lint\__init__.py", line 1, in <module> from . import lint # noqa: F401 File "C:\b\s\w\ir\third_party\wpt_tools\wpt\tools\lint\lint.py", line 1135, in <module> subprocess.check_output(["git", "--version"]) ... File "C:\b\s\w\ir\.task_template_vpython_cache\vpython\store\cpython-e6tenrlsesftvg6k08ajgkk7b4\contents\bin\Lib\subprocess.py", line 1311, in _execute_child hp, ht, pid, tid = _winapi.CreateProcess(executable, args, FileNotFoundError: [WinError 2] The system cannot find the file specified ```
Currently, when getComputedStyle() returns values for `left` etc, it needs to re-resolve them (after layout) into a pixel value. This patch provides an NGAnchorEvaluatorImpl for the resolution, so that anchor functions can also be resolved. There are also a few cases that are unsolved and left for future. They are marked with TODOs. Bug: 1423493 Change-Id: I319543230008e61a8363fdebb111a9840adc6e5d Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/4330385 Commit-Queue: Xiaocheng Hu <[email protected]> Reviewed-by: Koji Ishii <[email protected]> Cr-Commit-Position: refs/heads/main@{#1127332}
This change replaces touch with mouse wheel as the method of scrolling in overscroll-deltas.html, scrollend-event-fired-to-window.html, and scrollend-event-fired-to-document.html This allows these tests to be run on mac and also addresses an issue affecting wpt.fyi[1] runs where webdriver is unhappy about the touchpointer being out of bounds. The increased timeouts addresses a flakiness I observed with rr when I switched to using mouse which could possibly be related to the 500ms timeout that mouse wheel scrollends are based on. This patch adds some Math.round calls to scrollend-event-for-user-scroll.html because it appears the web driver version wpt.fyi uses does not accept non int values for test_driver.Actions().pointerMove(). This patch also marks the wpt overscroll-deltas.html flaky. We should revert this when crbug.com/1430921 is resolved. [1]https://wpt.fyi/results/dom/events/scrolling?label=master&label=experimental&aligned&q=scrollend Bug: 1429306,1430215,1430865,1430947 Change-Id: Icaa96202d2dd1adfc90c426c517fb9512affdc01 Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/4400771 Commit-Queue: David Awogbemila <[email protected]> Reviewed-by: Steve Kobes <[email protected]> Reviewed-by: Mustaq Ahmed <[email protected]> Cr-Commit-Position: refs/heads/main@{#1127417}
Change-Id: I8962ac3dce1d42764f01aa0244df8570ce554530 Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/4406145 Reviewed-by: Caleb Raitto <[email protected]> Commit-Queue: Caleb Raitto <[email protected]> Auto-Submit: Mike Taylor <[email protected]> Cr-Commit-Position: refs/heads/main@{#1127603}
This CL is similar to https://crrev.com/c/4397961, and removes all instances of nested calls to promise_test since it's not clear that nesting is supported by the test harness. Bug: 1427180 Change-Id: I0bc422cb0775c4c29f4969a6e52e7dd333050e07 Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/4407946 Auto-Submit: Chris Fredrickson <[email protected]> Reviewed-by: Shuran Huang <[email protected]> Commit-Queue: Shuran Huang <[email protected]> Cr-Commit-Position: refs/heads/main@{#1127615}
See step 3.2 of this algorithm in the spec: https://fs.spec.whatwg.org/#dom-storagemanager-getdirectory No Chromium bug for this, but I filed a bug against Firefox since they currently return a handle whose name is "root" https://bugzilla.mozilla.org/show_bug.cgi?id=1826886 Change-Id: Ie160ac6571cd287629893239e1df1d97c35b5d40 Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/4408007 Auto-Submit: Austin Sullivan <[email protected]> Reviewed-by: Daseul Lee <[email protected]> Commit-Queue: Daseul Lee <[email protected]> Cr-Commit-Position: refs/heads/main@{#1127619}
This CL adds step-by-step comments to both the internal and external Web-Locks WPTs. Note: some tests share the same resource files, for clarity, I have provided the file that the step listed comes from. Ultimately, I feel it is better to have more comments rather than duplicating the code so that each test may own its own copy. Fixed: 1420979 Change-Id: Iadf3eae2cc68759ed24be0819d2f3aef822fced2 Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/4355293 Reviewed-by: Andrew Williams <[email protected]> Commit-Queue: Kyra Seevers <[email protected]> Cr-Commit-Position: refs/heads/main@{#1127663}
Asserts computedrole for ARIA Core-AAM role synonyms - image/img - none/presentation - list/directory Related Issues: - web-platform-tests/interop-accessibility#10 - w3c/core-aam#166 Supporting Spec Change: - w3c/core-aam#167 Implementation Changes: - WebKit: https://webkit.org/b/224617 - Chromium: https://bugs.chromium.org/p/chromium/issues/detail?id=1428763 Reviewed By: - spectranaut: Valerie Young (Igalia)
HTMLFieldSetElement::DisabledAttributeChanged() traverses its subtree, and recurses on every form control descendant. This causes repeated traversal on subtrees, and results in exponential overall running time. The re-entrancy also causes a DCHECK hit when we blur the focused element in a nested traversal. This patch optimizes it to linear time by adding a non-reentrant override of ListedElement::AncestorDisabledStateWasChanged() on HTMLFieldSetelement. Fixed: 1427047 Change-Id: Ia3ee4195e71e05724af2bfb9abda6f1e94683ed9 Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/4406688 Commit-Queue: Xiaocheng Hu <[email protected]> Reviewed-by: Joey Arhar <[email protected]> Auto-Submit: Xiaocheng Hu <[email protected]> Commit-Queue: Joey Arhar <[email protected]> Cr-Commit-Position: refs/heads/main@{#1127824}
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
discarded PR due to git merge churn