Document `WEB_FEATURES.yml` files by jugglinmike · Pull Request #17 · bocoup/wpt

jugglinmike · 2026-04-07T22:55:34Z

A new "test metadata" page initially seemed like a logical place to group this
information alongside the "metadata" that's already documented for CSS tests,
reftests, and testharness.js tests.

Beyond their use of the term "metadata", though, it's difficult to concisely
explain what the various applications have in common. Many of them describe
critical information for interpreting the tests (e.g. rel="match" in
reftests), so what may technically be metadata in terms of HTML probably
shouldn't be presented on the same level as, e.g. the names of suitable code
reviewers.

Co-authored-by: wpt-pr-bot <wpt-pr-bot@users.noreply.github.com>

…platform-tests#59032)

The beforefilter event is fired on customizable comboboxes and filterable selects when the browser is about to filter the options, and can be canceled to allow the site to implement its own filtering instead. Context: openui/open-ui#931 Bug: 453705243 Change-Id: Ia1328a4b82226f2440737a07b5ea1c2306949951 Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/7727021 Reviewed-by: David Baron <dbaron@chromium.org> Reviewed-by: Joey Arhar <jarhar@chromium.org> Commit-Queue: Joey Arhar <jarhar@chromium.org> Cr-Commit-Position: refs/heads/main@{#1611068}

This CL integrates HTMLSubmitButtonBehavior into the form submission pipeline to allow CEs to act as submit buttons. Submission logic is updated to find a submit behavior and treat the associated custom element as the submitter during the event lifecycle. The implementation enables support for formAction, formMethod, formEnctype, formNoValidate, and formTarget defined within the behavior. It also ensures that the behavior's name and value properties are appended to the form data. `IsEffectivelyDisabled` method is added to check for the behavior's and element's states and the ancestor fieldsets. Activation events on the custom element are now delegated to the behavior to trigger the submission process. Tests are added to ElementInternals-submit-behavior.tentative.html to verify behavior. Explainer: https://github.com/MicrosoftEdge/MSEdgeExplainers/blob/main/PlatformProvidedBehaviors/explainer.md Design doc: https://docs.google.com/document/d/1LA1hhzxmi4OmZoGtIdnwvL3g7y48YjXTOoUvFtxFugE/edit?usp=sharing Bug: 486928684 Change-Id: Ifa93f060594e789a45d1b1809023427c4fcabf69 Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/7727461 Reviewed-by: Dan Clark <daniec@microsoft.com> Reviewed-by: Mason Freed <masonf@chromium.org> Commit-Queue: Ana Sollano Kim <ansollan@microsoft.com> Cr-Commit-Position: refs/heads/main@{#1611135}

…platform-tests#59035)

The previous CL (89f1c5e8fda3) moved the tests out of tentative/ but missed moving the shared-worker-memory.js resource and updating the references in the test files. This CL completes the cleanup. Original CL: https://chromium-review.googlesource.com/c/chromium/src/+/7734182 Bug: 400473072 Change-Id: I5d9fab0a3f0a0135384dd94a99d0064de723b44c Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/7734485 Reviewed-by: Minoru Chikamune <chikamune@chromium.org> Commit-Queue: Yoshisato Yanagisawa <yyanagisawa@chromium.org> Cr-Commit-Position: refs/heads/main@{#1611273}

…ces (web-platform-tests#59039)

)

…then appending a text node should not hang or crash Signed-off-by: Euclid Ye <yezhizhenjiakang@gmail.com>

Bumps [taskcluster](https://github.com/taskcluster/taskcluster) from 98.0.0 to 99.0.3. - [Release notes](https://github.com/taskcluster/taskcluster/releases) - [Changelog](https://github.com/taskcluster/taskcluster/blob/main/CHANGELOG.md) - [Commits](taskcluster/taskcluster@v98.0.0...v99.0.3) --- updated-dependencies: - dependency-name: taskcluster dependency-version: 99.0.3 dependency-type: direct:production update-type: version-update:semver-major ... Signed-off-by: dependabot[bot] <support@github.com>

This CL adds a simple Web platform test for Content Security Policy covering the algorithm "Is element nonceable?", cf. https://w3c.github.io/webappsec-csp/#is-element-nonceable. Force-WPT-Export: true Change-Id: Ie245ddb02d79196aaac9c2b151955bf56a15e9e9 Bug: 500099106 Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/7735128 Commit-Queue: Antonio Sartori <antoniosartori@chromium.org> Reviewed-by: Mike West <mkwst@chromium.org> Cr-Commit-Position: refs/heads/main@{#1611380}

Add missing commas after random()'s random sharing options. Tests were falsely passing before since the initial values for shorthand expansion longhand properties were same. Bug: 413385732 Change-Id: I1683e53cbdd6542d7cc39b0552ec204a139ed159 Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/7736428 Commit-Queue: Munira Tursunova <moonira@google.com> Reviewed-by: Rune Lillesveen <futhark@chromium.org> Cr-Commit-Position: refs/heads/main@{#1611402}

Co-authored-by: wpt-pr-bot <wpt-pr-bot@users.noreply.github.com>

Bug: 40940960 Change-Id: Ie735caab77704a645b57f0dd3be4aeafa6021569 Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/7726843 Reviewed-by: Free Debreuil <freedebreuil@google.com> Commit-Queue: Kevin Ellis <kevers@chromium.org> Cr-Commit-Position: refs/heads/main@{#1611434}

Ref: web-platform-dx/web-features#2900

…#58554) * Change PR reviewer to wpt-pr-bot for wasm workflow * Remove reviewer from PR creation in workflow Removed the reviewer parameter from the PR creation command.

)

When border-shape is active, background-clip:border-area clips backgrounds to the border-shape ring rather than the rectangular border contour. This uses a DstIn compositing mask that matches BorderShapePainter::Paint() so the two stay in sync. BorderShapePainter::PaintBorderArea() paints the ring as an opaque mask, sharing the same helper as Paint() to avoid duplication. Border-shape reference rects are resolved through BoxBackgroundPaintContext to keep fragment-aware geometry and fix table-part backgrounds. BackgroundImageGeometry computes and stores border-shape state so image-backed fills cover the full visible border-shape area including miter-join overflow. Bug: 329302543 Change-Id: I60a087df4fc93b400e0cde3c2fee669fcdf44348 Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/7707757 Reviewed-by: Fredrik Söderquist <fs@opera.com> Commit-Queue: Jason Leo <cgqaq@chromium.org> Cr-Commit-Position: refs/heads/main@{#1611443}

Flips the flag for Immediate Mode with passwords and passkeys. This also updates the getClientCapabilities WPT because enabling this feature changes the output in the default configuration. Fixed: 500653050 Change-Id: I51822e7d64e8f14eb89640ebc44aef7196458ea7 Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/7738102 Auto-Submit: Ken Buchanan <kenrb@chromium.org> Commit-Queue: Ken Buchanan <kenrb@chromium.org> Reviewed-by: Nina Satragno <nsatragno@chromium.org> Cr-Commit-Position: refs/heads/main@{#1611659}

ChrisC · 2026-04-08T19:08:16Z

+
+They define one property, `features`, which is a list of rules that relate one
+web-feature to one or more tests in that directory. Each mapping rule includes
+two properties: `name` (whose value is the string identifier of a web-feature)


@jugglinmike I wonder if it's worth calling out that this doesn't map to the web-feature name a little bit more explicitly? You pointed out the confusion the other day... Might also highlight the need to possibly update the web_features schema to avoid this confusion.

Sure. I'm proposing doing this as a footnote because this discrepancy is tricky to describe concisely (e.g. "the property named named [...]"). Hopefully we can obviate this caveat soon!

These changes are gated behind the AppearanceBase flag. See https://drafts.csswg.org/css-forms-1 Bug: 450139531 Change-Id: I1e962a5493ca2ee13184f98d34fa644424943048 Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/7650474 Reviewed-by: Joey Arhar <jarhar@chromium.org> Commit-Queue: Luke <lwarlow@igalia.com> Cr-Commit-Position: refs/heads/main@{#1611676}

ChrisC · 2026-04-08T19:12:44Z

+There is no mechanism for matching specific subdirectories (only for matching
+*all* subdirectories via `**`). To define mappings for files in a given
+subdirectory, write mapping rules in a `WEB_FEATURES.yml` file within that
+subdirectory.


I might have missed it, but I think it's helpful to explicit mention the use of the* pattern and how it's different from **. (And maybe also highlight that the ** pattern is the only value that can go into the files property without being part of a list.)

The type difference is described above, but I can certainly call out the distinct behavior of *.

Setting outerHTML on an element whose parent is a DocumentFragment was incorrectly throwing a NoModificationAllowedError. Per the HTML spec[0], the outerHTML setter should only throw when the parent is a Document. When the parent is a DocumentFragment, the spec requires using a temporary body element as the context for fragment parsing. Added web tests for DocumentFragment, Document, and no-parent cases. [0] https://html.spec.whatwg.org/multipage/dynamic-markup-insertion.html#the-outerhtml-property Fixed: 40885158 Change-Id: I79169dfbef86fbf4b6dc4d133fe90c13ca938857 Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/7530588 Reviewed-by: Fernando Fiori <ffiori@microsoft.com> Reviewed-by: Joey Arhar <jarhar@chromium.org> Commit-Queue: Stephanie Zhang <stephanie.zhang@microsoft.com> Cr-Commit-Position: refs/heads/main@{#1611702}

AIClassifier is a new feature that allows the browser to classify text. This new feature is exposed to Blink via a new Mojo interface. I2P: https://groups.google.com/a/chromium.org/g/blink-dev/c/5dQNl-gyjgU/m/6YBJJIiDAQAJ The change includes the following: Blink Mojom: New AIClassifier interface and associated data structures. Chrome Implementation: AIClassifier class in chrome/browser/ai that uses an OnDeviceSession to interact with the classification model. AIManager Integration: Added CanCreateClassifier and CreateClassifier methods to AIManager to allow Blink to request and create AIClassifier instances. Content Implementation: EchoAIClassifier for testing purposes. Metrics: Added kClassifier to the AISessionType enum. Renderer: Updated classifier.cc to call the new AIManager methods. Bug: 483707607 Change-Id: I04676439afb3feed54237a701365b332910a10f6 Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/7688370 Reviewed-by: Nathan Memmott <memmott@chromium.org> Reviewed-by: Jingyun Liu <jingyun@google.com> Commit-Queue: Devin Cabillo <devincabillo@google.com> Cr-Commit-Position: refs/heads/main@{#1611826}

This changes the behavior when selecting a new option in the case that there are nested selectedcontent elements inside the select. Before, the disabled selectedcontent would get cloned, but now it won't. Context: whatwg/html#12263 (comment) Bug: 458113204 Change-Id: I9afd550f21787b7daefb48c5a76712b5899e5b42 Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/7737623 Reviewed-by: David Grogan <dgrogan@chromium.org> Reviewed-by: Joey Arhar <jarhar@chromium.org> Commit-Queue: Joey Arhar <jarhar@chromium.org> Cr-Commit-Position: refs/heads/main@{#1611875}

Co-authored-by: wpt-pr-bot <wpt-pr-bot@users.noreply.github.com>

The test was using a 1ms delay between closing one popup and opening another, which likely caused intermittent failures and timeouts in headless environments like wpt.fyi. This change increases the delay to 1000ms, matching the 'Wait for a second' comment in the code. Bug: 400473072 Change-Id: Ic8bf7cd47bd4ae362d9b9a3bd0c5081afc875772 Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/7742622 Commit-Queue: Minoru Chikamune <chikamune@chromium.org> Auto-Submit: Yoshisato Yanagisawa <yyanagisawa@chromium.org> Reviewed-by: Minoru Chikamune <chikamune@chromium.org> Cr-Commit-Position: refs/heads/main@{#1612597}

This patch changes to set the responsive embedded sizing flag of the document only when `is_sync_parser_` flag is set, which is used by the `Accept-CH`/`Delegate-CH` headers. Bug: 418397278 Change-Id: I766aab06ae07841ed1f0a4d8e02efed1b4eae846 Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/7743780 Reviewed-by: Ian Kilpatrick <ikilpatrick@chromium.org> Reviewed-by: Koji Ishii <kojii@chromium.org> Cr-Commit-Position: refs/heads/main@{#1612619}

CSSWG resolution [1]: RESOLVED: image-animation applies to <video> poster images Since the implementation already provide image-animation behavior, add WPT cases. [1] w3c/csswg-drafts#13641 Fixed: 498558533 Change-Id: I6b9f07db083fc2b5e094ac7a50c5664ba71a5def Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/7738202 Commit-Queue: Seokho Song <seokho@chromium.org> Reviewed-by: Anders Hartvoll Ruud <andruud@chromium.org> Cr-Commit-Position: refs/heads/main@{#1612622}

When a grid-lanes container with row-axis tracks has an initially indefinite block size that resolves to a definite value, and `NeedsAdditionalLayoutPass()` returns `false` (no full re-run needed), non-default align-content values still need their row geometry re-finalized with the resolved block size. This mirrors the same logic in `GridLayoutAlgorithm::ComputeGridGeometry()`. This is a follow-up CL for CL:7713389. Bug: 343257585 Change-Id: I5c6d2da887a656497aab6a5fe6801d6fd147a881 Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/7737713 Commit-Queue: Yanling Wang <yanlingwang@microsoft.com> Reviewed-by: Alison Maher <almaher@microsoft.com> Cr-Commit-Position: refs/heads/main@{#1612637}

If the `Unencoded-Digest` header contains an unknown algorithm, we can't enforce its assertions on the content we download. This enables forward compatibility with new algorithms, but also opens a hole in signature checks, insofar as the signature over an unenforceable digest is equally unenforceable. Here, we follow the spec change in [1] to require enforcability. [1]: WICG/signature-based-sri@c6300af Bug: 498080391 Change-Id: I7422e261831cd1af0da9fb97c603350733cff262 Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/7734936 Reviewed-by: Antonio Sartori <antoniosartori@chromium.org> Commit-Queue: Mike West <mkwst@chromium.org> Cr-Commit-Position: refs/heads/main@{#1612754}

Resolved in [0] and [1] to change random syntax to following: <random()> = random( <random-cache-key>? , <calc-sum>, <calc-sum>, <calc-sum>? ) <random-cache-key> = auto | <random-name> | fixed <number [0,1]> <random-name> = <dashed-ident> [ element-scoped || [ property-scoped | property-index-scoped ]]? Previous syntax: <random()> = random( <random-value-sharing>? , <calc-sum>, <calc-sum>, <calc-sum>? ) <random-value-sharing> = [ [ auto | <dashed-ident> ] || element-shared ] | fixed <number [0,1]> Spec has not been fully updated yet. [0] w3c/csswg-drafts#13132 (comment) [1] w3c/csswg-drafts#13601 (comment) Bug: 413385732 Change-Id: I30f355ef6241b3d0c8e259bc1044ba2f57ce2b97 Fixed: 498620677, 489743837 Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/7735052 Commit-Queue: Munira Tursunova <moonira@google.com> Reviewed-by: Rune Lillesveen <futhark@chromium.org> Cr-Commit-Position: refs/heads/main@{#1612756}

…ating frame. This check was originally introduced in crrev.com/c/6763697 to only allow LNA permission prompts to trigger within a single tab/context. However, crbug.com/491509051 details a valid scenario where the initiator of a subframe navigation is not in the frame tree of the subframe being navigated. Specifically, this happens when: * top window (origin1) iframes origin2 * iframe of origin2 opens a new window, also to origin2 * new window of origin2 does `opener.location.href = <LNA-url>` With this fix, this scenario ends up checking the LNA permission on the initiator, which may pop up the permission prompt on the initiator window. This was chosen as it is the initiator which is triggering the LNA navigation. One quirk here: if instead of the opened window initiating the navigation, if instead it posted a message to the iframe that triggered the iframe to navigate itself, then the permission check would be on the iframe. This might lead to slightly different behaviour depending on permission policy settings, but is not a permissions bypass as the LNA permission is still required from an initiator in either case. Bug: 491509051 Change-Id: Iac1b4ded1459a61a93b9a6fdaf8278db529bf50b Cq-Do-Not-Cancel-Tryjobs: true Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/7681373 Reviewed-by: Camille Lamy <clamy@chromium.org> Commit-Queue: Hubert Chao <hchao@chromium.org> Reviewed-by: Chris Thompson <cthomp@chromium.org> Cr-Commit-Position: refs/heads/main@{#1612807}

@font-face

Add support for request modifiers for @font-face src and external SVG resources, as well as corresponding tests. Bug: 435625756 Change-Id: Ic02b348e95302f8c62bfe230f59a357509ce7789 Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/7748606 Reviewed-by: Fredrik Söderquist <fs@opera.com> Commit-Queue: jj <jj@chromium.org> Cr-Commit-Position: refs/heads/main@{#1612810}

This patch invalidates the scrollbar painting when the used value of pointer events changed. This is necessary since this can affect the opaqueness value stored on the compositor (which is updated via paint). Without the patch, the value on the compositor can be stale which allows the user to interact with the scrollbar after the inert is toggled on. R=flackr@chromium.org Bug: 497896149 Change-Id: I9beb7e34b676d0379b6d5609a1c03fe8ac535566 Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/7738301 Reviewed-by: Robert Flack <flackr@chromium.org> Commit-Queue: Vladimir Levin <vmpstr@chromium.org> Cr-Commit-Position: refs/heads/main@{#1612828}

Refactor the test so that each failure is surfaced individually. Reduce boilerplate to make it easier to test additional cases. Bug: 499073687 Change-Id: Ibcb7173d764082164ee317853c8a11adbf47685f Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/7741968 Reviewed-by: Divyansh Mangal <dmangal@microsoft.com> Commit-Queue: Divyansh Mangal <dmangal@microsoft.com> Auto-Submit: Fredrik Söderquist <fs@opera.com> Cr-Commit-Position: refs/heads/main@{#1612956}

Dialogs only restore focus to the previously focused element on close when an element inside the dialog is focused. When there is no focusable element inside of the dialog and the dialog element itself becomes focused, we should restore focus to the previously focused element when the dialog is closed. This change matches the spec, which uses an inclusive descendant check on the dialog element. Fixed: 497884156 Change-Id: I48e0e617a82a3ebe784884bd60f7d81e2776b71d Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/7735579 Commit-Queue: Joey Arhar <jarhar@chromium.org> Reviewed-by: David Baron <dbaron@chromium.org> Reviewed-by: Joey Arhar <jarhar@chromium.org> Cr-Commit-Position: refs/heads/main@{#1613005}

Before this patch, customizable combobox would automatically set the first option in the datalist to :active-option every time that the datalist is opened. However, this doesn't match the behavior of any comboboxes I looked at. With this patch, the user must arrow down after opening the datalist in order for any option to start matching :active-option. This behavior is not applied to filterable select because the "Labels" picker on github issues always sets the first option as the active one when the filtering input is focused. Bug: 453705243 Change-Id: I6680a98c95144043bbea67aa923eedf7d54a9d84 Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/7728808 Reviewed-by: David Grogan <dgrogan@chromium.org> Reviewed-by: Joey Arhar <jarhar@chromium.org> Commit-Queue: Joey Arhar <jarhar@chromium.org> Cr-Commit-Position: refs/heads/main@{#1613020}

When restoring form control state, the DOM could be modified to add or remove more listed elements to the form if a select element is being restored which has a selectedcontent element. Fixed: 499384399 Change-Id: I18f69c30ae25396c53625f7a3172626b79de3ae3 Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/7732030 Reviewed-by: Joey Arhar <jarhar@chromium.org> Commit-Queue: Joey Arhar <jarhar@chromium.org> Reviewed-by: Dominic Farolino <dom@chromium.org> Cr-Commit-Position: refs/heads/main@{#1613032}

This change adds "onstream" to the list of recognized HTML attributes and maps it to the "stream" event type. Also added WPTs to verify that both the onstream IDL property and the onstream HTML attribute function correctly, along with addEventListener("stream"). Bug: 498583077 Change-Id: I752da8de0de0d08c625ae89c738c8f3be9b4c39a Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/7717754 Reviewed-by: Joey Arhar <jarhar@chromium.org> Commit-Queue: Ravjit Uppal <ravjit@chromium.org> Auto-Submit: Ravjit Uppal <ravjit@chromium.org> Cr-Commit-Position: refs/heads/main@{#1613047}

Updates to match recent spec changes These changes are gated behind the AppearanceBase flag. See https://drafts.csswg.org/css-forms-1 Bug: 450139531 Change-Id: I09140a81190ec0697adb077229356dcd2aeb57aa Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/7747467 Reviewed-by: Joey Arhar <jarhar@chromium.org> Commit-Queue: Luke <lwarlow@igalia.com> Cr-Commit-Position: refs/heads/main@{#1613068}

* Add WPTs for navigator_web_install [1] preconditions/bad input * Migrate existing bad input browsertests to unittests * Adjust timing of service binding, so bad input is unit testable [1] explainer - https://github.com/MicrosoftEdge/MSEdgeExplainers/blob/main/WebInstall/explainer.md Low-Coverage-Reason: COVERAGE_UNDERREPORTED covered by web platform tests Bug: 498644992 Change-Id: Icbc7c97c55db1d8ab00b649126008db8578d0aa0 Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/7722810 Reviewed-by: Lu Huang <luhua@microsoft.com> Commit-Queue: Lia Hiscock <liahiscock@microsoft.com> Reviewed-by: Daniel Murphy <dmurph@chromium.org> Reviewed-by: David Baron <dbaron@chromium.org> Cr-Commit-Position: refs/heads/main@{#1613146}

According to the [comment], "immutable" means: (a) it can only be set by a parser-created <meta> (b) it's set and frozen by the first such <meta> the parser sees, and (c) if the parser closes the <head> without seeing the meta This patch implements the (c). The (a) and (b) are already implemented. Technically speaking, (c) means that "sees `<body>` or tags that imply `<body>`", because the ["after head" insertion mode] pushes `<meta>` elements after `<head>` but before `<body>` into `<head>`. This patch has a test for this case too. [comment]: w3c/csswg-drafts#1771 (comment) ["after head" insertion mode]: https://html.spec.whatwg.org/multipage/parsing.html#the-after-head-insertion-mode Bug: 418397278 Change-Id: Ic01b94eb3e502562cdd479435b6d0f77b30bb229 Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/7747506 Reviewed-by: Kent Tamura <tkent@chromium.org> Reviewed-by: Koji Ishii <kojii@chromium.org> Reviewed-by: Ian Kilpatrick <ikilpatrick@chromium.org> Auto-Submit: Koji Ishii <kojii@chromium.org> Commit-Queue: Kent Tamura <tkent@chromium.org> Cr-Commit-Position: refs/heads/main@{#1613459}

In https://crrev.com/c/6802120, to fix a repaint issue, line-clamp ellipses added as part of line-breaking (with the `CSSLineClampLineBreakingEllipsis` runtime flag) were associated with a relevant `LayoutObject`. This involved filtering the `InlineItem`s before the clamp point with the `CanUseItemForNeedsPaint` function added to `logical_line_builder.cc`. Notably, this function did not filter out open tags, meaning that the ellipsis could be associated with non-atomic inline box layout objects, which is not something that could ever happen for the ellipses generated by `LineTruncator`. As it turns out, having an inline box start immediately before a block-in-inline causes the line-breaking ellipsis to be associated with the inline box. If that inline box happens to have an outline, then `PhysicalBoxFragment::AddOutlineRectsForInlineBox` will try to find all fragment items associated with the box, including the ellipsis, leading to wrong assumptions since the ellipsis isn't a wrong item. A solution could be to explicitly filter out inline boxes with an outline, but since `LineTruncator` doesn't associate any ellipses with any non-atomic inline items, we chose to do the case for line-breaking ellipses. As a related note, this fix would also have solved bug 489494018, which was fixed in https://crrev.com/c/7649894 by allowing associating ellipses with line break items. Since both changes are separate, and both bring the behavior closer to `LineTruncator`'s, it makes sense to not revert that one. Bug: 40336192, 499496973 Change-Id: Iacb8cb5298e199198aba0545fdcd3cd64c5f4da6 Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/7726309 Reviewed-by: Koji Ishii <kojii@chromium.org> Commit-Queue: Andreu Botella <abotella@igalia.com> Cr-Commit-Position: refs/heads/main@{#1613505}

Expanding a 0-D input to a 0-D output causes an access violation in a certain ORT EP. Work around this by using an identity node instead, since the operation is a no-op. Bug: 500394293 Change-Id: Ic5ede18f6760e644bf92a8da250fc71af2e2cdcb Cq-Include-Trybots: luci.chromium.try:win11-blink-rel, mac14.arm64-blink-rel, mac14-blink-rel, mac15.arm64-blink-rel, linux-blink-rel Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/7738338 Reviewed-by: Hu, Ningxin <ningxin.hu@intel.com> Commit-Queue: Zou, Shiyi <shiyi.zou@intel.com> Cr-Commit-Position: refs/heads/main@{#1613534}

Wait for Ahem before taking reftest snapshots in css-ruby and css-writing-modes text-combine-upright tests. This avoids fallback-font first layout differences and makes snapshots deterministic. Remove temporary Linux Failure Pass expectations now that these tests are stable again. Bug: 495770240,431038526 Change-Id: I29b4d8f38c9c2e58e54b4e8e2377cc0663daff50 Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/7705114 Reviewed-by: Yoav Weiss (@Shopify) <yoavweiss@chromium.org> Auto-Submit: Helmut Januschka <helmut@januschka.com> Commit-Queue: Helmut Januschka <helmut@januschka.com> Cr-Commit-Position: refs/heads/main@{#1613609}

…sition Bug: 496807859 Change-Id: Id954d9463407ef4f858005026692398fdd91b296 Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/7703993 Reviewed-by: Fredrik Söderquist <fs@opera.com> Reviewed-by: Koji Ishii <kojii@chromium.org> Commit-Queue: Fredrik Söderquist <fs@opera.com> Auto-Submit: Nishaanth Shriram M S <nishaanthshriramms@gmail.com> Cr-Commit-Position: refs/heads/main@{#1613611}

See bug for details. Bug: 502073069 Change-Id: I39b9328f87899b9f302c3d3fe36aaa7d5dc90010 Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/7749166 Reviewed-by: Noam Rosenthal <nrosenthal@google.com> Commit-Queue: Yoav Weiss (@Shopify) <yoavweiss@chromium.org> Cr-Commit-Position: refs/heads/main@{#1613631}

When the kDataUrlWorkerOpaqueOrigin feature is enabled, data: URL workers use a correct opaque origin for their StorageKey. This eliminates the mismatch between the Blob URL's opaque origin and the StorageKey's origin, which means the Blob URL system no longer needs the ALLOW_OPAQUE_ORIGIN_STORAGE_KEY_MISMATCH workaround. This CL conditionally passes DEFAULT instead of the workaround to the Blob URL registry when the feature is enabled. This proves the workaround is no longer needed and allows it to be removed entirely once the feature is enabled by default, without modifying the underlying Blob URL validation logic unconditionally at this stage. Bug: 40051700 Change-Id: Ie4d2187ad42051aed587da67b01ba7439ba0ba1a Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/7734490 Auto-Submit: Yoshisato Yanagisawa <yyanagisawa@chromium.org> Commit-Queue: Andrew Williams <awillia@chromium.org> Reviewed-by: Andrew Williams <awillia@chromium.org> Cr-Commit-Position: refs/heads/main@{#1612600}

CSP web platform tests for connect-src inside workers started timing out after the change web-platform-tests#58737 which adapted the expectations for reported blocked-uris for websocket connections. This change adapts the tests so that they don't timeout but instead fail with output, so that we don't lose test coverage, until we figure out what to do wrt https://crbug.com/501115606. Fixed: 501216031 Bug: 501115606 Change-Id: Iee84d30b92710b7e8a532f7a998cbf27a9e94190 Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/7748325 Reviewed-by: Mike West <mkwst@chromium.org> Commit-Queue: Antonio Sartori <antoniosartori@chromium.org> Cr-Commit-Position: refs/heads/main@{#1612752}

Have ContainerNode::GetLayoutBoxForScrolling() return the delegated box when it has a ScrollableArea, rather than requiring it to be an actual scroll container. Add GetLayoutBoxIfScrollContainer() for internal callers that need scroll container semantics, and update those call sites accordingly. Also make PaintLayerScrollableArea min/max scroll offsets return zero for non-scroll-container boxes so DOM scroll APIs and PLSA scroll geometry stay in sync. Add WPT coverage for overflow: clip DOM scrolling behavior. Change-Id: I710332bc6f7d225b21643cc91f9cbfbc5b9221c0 Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/7737441 Auto-Submit: Free Debreuil <freedebreuil@google.com> Reviewed-by: Robert Flack <flackr@chromium.org> Commit-Queue: Robert Flack <flackr@chromium.org> Commit-Queue: Free Debreuil <freedebreuil@google.com> Cr-Commit-Position: refs/heads/main@{#1612947}

`Element::IsFocusableStyle` has special code that looks for a canvas ancestor, and this needs to be updated to cross frame boundaries. This regressed when https://crrev.com/1597048 changed `IsCanvasOrInCanvasSubtree` to be true in iframes, but the code in `Element::IsFocusableStyle` was not updated to find the ancestor canvas element through iframes. Fixed: 492245060 Change-Id: Ieb2f294760fb726ffeb56d1382366e08b9dcd8e7 Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/7750651 Auto-Submit: Philip Rogers <pdr@chromium.org> Reviewed-by: Stephen Chenney <schenney@chromium.org> Commit-Queue: Stephen Chenney <schenney@chromium.org> Cr-Commit-Position: refs/heads/main@{#1613437}

jugglinmike requested a review from ChrisC April 7, 2026 22:55

github-actions Bot and others added 23 commits April 7, 2026 20:10

Sync interfaces/ with @webref/idl 3.75.1 (web-platform-tests#59018)

59b51b3

Co-authored-by: wpt-pr-bot <wpt-pr-bot@users.noreply.github.com>

WebKit export of https://bugs.webkit.org/show_bug.cgi?id=311620 (web-…

38e10fe

…platform-tests#59032)

WebKit export of https://bugs.webkit.org/show_bug.cgi?id=258125 (web-…

ee909cc

…platform-tests#59035)

Tests for ruby-overhang: spaces (web-platform-tests#59038)

ac6cede

Broader coverage of opening/closing punctuation for ruby-overhang:spa…

74c4bdd

…ces (web-platform-tests#59039)

[css-ruby] Test ruby-overhang:spaces with NBSP (web-platform-tests#59040

fb2d803

)

Build detached deeply nested shadow tree, attaching to the document, …

669afb3

…then appending a text node should not hang or crash Signed-off-by: Euclid Ye <yezhizhenjiakang@gmail.com>

Digital Credentials: check insecure context (web-platform-tests#58629)

a48b2d1

Sync interfaces/ with @webref/idl 3.75.2 (web-platform-tests#59043)

67b3da6

Co-authored-by: wpt-pr-bot <wpt-pr-bot@users.noreply.github.com>

Update web-feature mappings for recent rename (web-platform-tests#59027)

7f775df

Ref: web-platform-dx/web-features#2900

Let wpt-pr-bot assign reviewers for wasm workflow (web-platform-tests…

91a495a

…#58554) * Change PR reviewer to wpt-pr-bot for wasm workflow * Remove reviewer from PR creation in workflow Removed the reviewer parameter from the PR creation command.

Rename metadata file to match local convention (web-platform-tests#58927

af8ed4a

)

maps "shape-function" web feature

bae435b

maps "rect-xywx" web feature

32b9a69

ChrisC reviewed Apr 8, 2026

View reviewed changes

stephanieyzhang and others added 3 commits April 8, 2026 13:04

github-actions Bot and others added 28 commits April 13, 2026 11:27

Sync interfaces/ with @webref/idl 3.75.2 (web-platform-tests#59142)

0f6fc94

Co-authored-by: wpt-pr-bot <wpt-pr-bot@users.noreply.github.com>

Document WEB_FEATURES.yml files

3ed1221

jugglinmike force-pushed the docs-web-features branch from ffd6a11 to 3ed1221 Compare April 14, 2026 00:13

jugglinmike closed this Apr 14, 2026

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Document `WEB_FEATURES.yml` files#17

Document `WEB_FEATURES.yml` files#17
jugglinmike wants to merge 99 commits intotmpfrom
docs-web-features

jugglinmike commented Apr 7, 2026

Uh oh!

ChrisC Apr 8, 2026

Uh oh!

jugglinmike Apr 9, 2026

Uh oh!

ChrisC Apr 8, 2026

Uh oh!

jugglinmike Apr 9, 2026

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

20 participants

Conversation

jugglinmike commented Apr 7, 2026

Uh oh!

ChrisC Apr 8, 2026

Choose a reason for hiding this comment

Uh oh!

jugglinmike Apr 9, 2026

Choose a reason for hiding this comment

Uh oh!

ChrisC Apr 8, 2026

Choose a reason for hiding this comment

Uh oh!

jugglinmike Apr 9, 2026

Choose a reason for hiding this comment

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

20 participants