1876163 - implement recent HTML spec changes to directionality and shadow DOM

Status: RESOLVED FIXED

(Core :: DOM: Core & HTML, enhancement)

Description

[David Baron :dbaron: (⌚️UTC-5, no longer working on Mozilla)]

There were a bunch of recent HTML spec changes related to directionality (and, in small ways, language) and their inheritance and interaction with Shadow DOM. The most substantive changes were in https://github.com/whatwg/html/pull/9796 but there were also other changes that I've linked in the "See Also" field.

Comment 1

[Vincent Hilla [:vhilla]]

Attachment: Bug 1876163 - Part 1: Refactor and add more text directionality tests. r=#dom-core

Comment 2

[Vincent Hilla [:vhilla]]

Attachment: Bug 1876163 - Part 2: Update static case of compute auto directionality. r=#dom-core

Depends on D208337

Comment 3

[Vincent Hilla [:vhilla]]

Attachment: Bug 1876163 - Part 4: Slot auto directionality for dynamic changes. r=#dom-core

Depends on D208338

Comment 4

[Vincent Hilla [:vhilla]]

Attachment: Bug 1876163 - Part 5: Consider input / textarea children of dir=auto slot. r=#dom-core

Depends on D208339

Comment 5

[Vincent Hilla [:vhilla]]

Attachment: Bug 1876163 - Part 3: Track whether node affects assigned dir=auto slot of ancestor. r=emilio,smaug

Depends on D208338

Comment 6

[Vincent Hilla [:vhilla]]

https://github.com/web-platform-tests/interop/issues/668

Comment 7

[Pulsebot]

Pushed by [e-mail]:
https://hg.mozilla.org/integration/autoland/rev/613c14c84d7a
Part 1: Refactor and add more text directionality tests. r=emilio
https://hg.mozilla.org/integration/autoland/rev/66a9df8d3bd2
Part 2: Update static case of compute auto directionality. r=emilio,smaug
https://hg.mozilla.org/integration/autoland/rev/763a627a0504
Part 3: Track whether node affects assigned dir=auto slot of ancestor. r=emilio,smaug
https://hg.mozilla.org/integration/autoland/rev/166e0ae30792
Part 4: Slot auto directionality for dynamic changes. r=smaug,emilio
https://hg.mozilla.org/integration/autoland/rev/160c9642bf7a
Part 5: Consider input / textarea children of dir=auto slot. r=smaug

Comment 8

[Web Platform Test Sync Bot [:wpt-sync] (Matrix: #interop:mozilla.org)]

Failed to create upstream wpt PR due to merge conflicts. This requires fixup from a wpt sync admin.

Comment 9

[Cosmin Sabou [:CosminS]]

https://hg.mozilla.org/mozilla-central/rev/613c14c84d7a
https://hg.mozilla.org/mozilla-central/rev/66a9df8d3bd2
https://hg.mozilla.org/mozilla-central/rev/763a627a0504
https://hg.mozilla.org/mozilla-central/rev/166e0ae30792
https://hg.mozilla.org/mozilla-central/rev/160c9642bf7a

Comment 10

[Web Platform Test Sync Bot [:wpt-sync] (Matrix: #interop:mozilla.org)]

Created web-platform-tests PR https://github.com/web-platform-tests/wpt/pull/47069 for changes under testing/web-platform/tests

Comment 11

[Web Platform Test Sync Bot [:wpt-sync] (Matrix: #interop:mozilla.org)]

Upstream PR merged by moz-wptsync-bot

Comment 12

[Web Platform Test Sync Bot [:wpt-sync] (Matrix: #interop:mozilla.org)]

Upstream PR merged by moz-wptsync-bot

Comment 13

[David Baron :dbaron: (⌚️UTC-5, no longer working on Mozilla)]

It took longer than I hoped, but I finally had a chance to look at the three tests that were added in this bug that Chromium is failing.

I have fixes for two of them.

But I think this test is testing something that was a mistake in the spec:
https://github.com/web-platform-tests/wpt/blob/b6bca1a1aa8042d6651dc8ac9a0541cd506377b2/html/dom/elements/global-attributes/dir-auto-dynamic-changes.window.js#L381

If I'm understanding this test correctly, it looks to me like it is assuming that a form element being "auto-directionality form associated" influences the dir=auto state on other elements. However, I think being "auto-directionality form associated" was intended to influence what dir=auto does on the element itself, but not have any effect on dir=auto on ancestors.

However, the spec text has special handling for slot elements because what matters is their assigned nodes and not their children -- but I think that traversal was intended to be like a traversal of children, but differ only in that it traverses the assigned nodes rather than the children. So I think that spec text's reference to the auto directionality of the child is wrong, and the spec probably needs to be fixed to make that traversal (step 2 in https://html.spec.whatwg.org/multipage/dom.html#auto-directionality) behave more like traversal of children.

Does that make sense?

Comment 14

[David Baron :dbaron: (⌚️UTC-5, no longer working on Mozilla)]

I created https://github.com/whatwg/html/pull/10561 for a spec update, so probably better to discuss there rather than here.

Comment 15

[Dave Letorey]

Asscoiated pull requests

• Content PR
• Firefox Release PR