1037709 - Investigate if accesskeys should be handled when element is in a shadow dom bound to in-doc shadow host

Status: RESOLVED FIXED

(Core :: DOM: UI Events & Focus Handling, defect, P2)

Description

[Olli Pettay [:smaug][[e-mail]]]

Per HTML spec such accesskeys shouldn't be handled, but perhaps
shadow DOM or HTML spec need to be changed.

Comment 1

[Rich Caloggero]

Please support this

Why should accesskey not work within shadow dom. If the UI is usable (i.e. is keyboard controlable) then the use should not know (or care) whether the elements live in shadow dom or normal dom, thus IMO accesskey should be supported since it is supported in normal dom (and can be very useful in web apps since it may save lots of tabbing). It is also a very simple way of attaching a keyboard shortcut to an element!

Comment 2

[Rich Caloggero]

Attachment: t.html

Comment 3

[Anne (:annevk)]

This is tracked by https://github.com/whatwg/html/issues/4385 standards-wise and per bug 1588156 comment 2 this is problematic for about:addons. The tentative outcome from the standards discussion is that accesskey should work across tree boundaries.

Hsin-Yi, can you find someone to take this?

Comment 4

[Hsin-Yi Tsai (she/her) | away [:hsinyi]]

(In reply to Anne (:annevk) from comment #3)

This is tracked by https://github.com/whatwg/html/issues/4385 standards-wise and per bug 1588156 comment 2 this is problematic for about:addons. The tentative outcome from the standards discussion is that accesskey should work across tree boundaries.

Hsin-Yi, can you find someone to take this?

Sure, though we probably wouldn't get it to Q4 but put it the queue to prioritize.

Comment 5

[Olli Pettay [:smaug][[e-mail]]]

Some relevant code for this https://searchfox.org/mozilla-central/rev/088e2cf29c59d733d57af43903eb0267dbf72e2a/dom/html/nsGenericHTMLElement.cpp#379-380,404-405
https://searchfox.org/mozilla-central/rev/088e2cf29c59d733d57af43903eb0267dbf72e2a/dom/html/nsGenericHTMLElement.cpp#379-380,404-405,532

Might need to tweak some EventStateManager code too, depends on what the spec will look like.

Comment 6

[Edgar Chen [:edgar]]

Attachment: Bug 1037709 - Part 1: Remove nsGenericHTMLElement::RegAccessKey/UnregAccessKey; r=smaug

Comment 7

[Edgar Chen [:edgar]]

Attachment: Bug 1037709 - Part 2: Make element in a shadow DOM also handling accesskey; r=smaug

This is tracked by https://github.com/whatwg/html/issues/4385 standards-wise and
the tentative outcome from the standards discussion is that accesskey should work
across tree boundaries.

Depends on D110425

Comment 8

[Edgar Chen [:edgar]]

https://treeherder.mozilla.org/#/jobs?repo=try&revision=c7cbd3106d51eadd4e1d4e61556e8f17a227df96

Comment 9

[Pulsebot]

Pushed by [e-mail]:
https://hg.mozilla.org/integration/autoland/rev/36d3002f77b6
Part 1: Remove nsGenericHTMLElement::RegAccessKey/UnregAccessKey; r=smaug
https://hg.mozilla.org/integration/autoland/rev/8bfe6426b2de
Part 2: Make element in a shadow DOM also handling accesskey; r=smaug

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/28339 for changes under testing/web-platform/tests

Comment 11

[Bogdan Tara[:bogdan_tara | bogdant]]

https://hg.mozilla.org/mozilla-central/rev/36d3002f77b6
https://hg.mozilla.org/mozilla-central/rev/8bfe6426b2de

Comment 12

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

Upstream PR merged by moz-wptsync-bot