updating aria-live=off to match the aligned implementations.#2677
updating aria-live=off to match the aligned implementations.#2677spectranaut merged 3 commits intomainfrom
Conversation
✅ Deploy Preview for wai-aria ready!
To edit notification comments on pull requests, go to your Netlify project configuration. |
|
Added Agenda+ to the related issue, just to make sure this doesn't slip by any WG members accidentally. Removing the duplicate keyword here. |
|
Assigning myself to test orca. |
|
Braille tests revealed no unexpected results. @spectranaut Assuming Orca aligns with the others, you can merge since this has two other approvals. |
There was a problem hiding this comment.
From my testing in chrome on linux, Orca ignores updates in timer and marquee, like expected.
However, this does actually need bugs on browsers -- at least chrome exposes live:none and container-live:none on these roles from my local testing, and if we remove it from the AAM, then it shouldn't.
|
@joanmarie -- this change to Core-AAM removes "container-live:off" and "live:off" from the object attributes for But chrome exposes this information on linux for both regardless of whether aria-live is present. I assume, but wanted to make sure, that Orca doesn't need that information (in the case that aria-live is not present) before opening the bug on chrome to remove it? |
|
Just checked. Orca's definition is: @staticmethod
def is_live_region(obj: Atspi.Accessible, _role: Atspi.Role | None = None) -> bool:
"""Returns True if obj is a live region."""
attrs = AXObject.get_attributes_dict(obj)
return "container-live" in attrs and attrs.get("container-live") in ["polite", "assertive"]So removing the attribute is no different than exposing it with a value of "off". Remove away! :) And thanks for checking!! |
|
Ok, I added bugs. Because (1) change to Core-AAM is not technically normative -- we do not say that you "MUST NOT" expose attributes not listed in the AAM tables and (2) exposing container-live:off and live:off (container-live-role:.. is already not exposed) has no downstream effect, I think this is fine to merge. |
SHA: 2d9ac48 Reason: push, by spectranaut Co-authored-by: github-actions[bot] <41898282+github-actions[bot]@users.noreply.github.com>
|
sorry for the naive question, but: does this change now mean that (i.e., quite self-servingly, can i now update the first part of that final note in https://tetralogical.com/blog/2024/05/01/why-are-my-live-regions-not-working/ that rants about "off" being weird? and do we need to make updates to MDN and such?) |
According to w3c/aria#2677 Bug: 477317119 Change-Id: Icab122430cdc141d582810aa2cc6f1f7221a1c9b Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/7529653 Reviewed-by: Joey Arhar <[email protected]> Reviewed-by: David Tseng <[email protected]> Commit-Queue: Marco Esau Garcia Martinez <[email protected]> Cr-Commit-Position: refs/heads/main@{#1578813}
Closes #2144
Closes #2653
Closes #2664
Closes #2674
Updating
aria-live="off"to match the aligned implementations:… Effectively that the following are not live regions, becausearia-live="off"negates all live region functionality.role="marquee"role="timer"<div aria-live="off">(aria-liveis global, and the default isoffon everything, so this is identical to<div>)See the issues for more detail.
marqueeandtimerlisted as live region roles but have intrinsicaria-live="off"#2144aria-live="off"that is less confusing to authors #2664Implementations:
Preview | Diff