[Webkit-unassigned] [Bug 233383] AX: aria-owns in trees results duplicate rows, third level rows not being exposed

[bugzilla-daemon at webkit.org]

https://bugs.webkit.org/show_bug.cgi?id=233383

--- Comment #12 from Sepand Parhami <sparhami at apple.com> ---
(In reply to Tyler Wilcock from comment #10)
> FWIW, I think we only need to be concerned about not crashing when the
> author makes the mistake of a circular reference. I tested a circular
> aria-owns reference on `main`-branch WebKit and couldn't get a crash, so we
> just need to make sure that this patch doesn't make us susceptible to it.
> 
> One option would be to create a separate test with an intentional circular
> reference, and to call it passing if it doesn't crash.
> 
> Thanks for digging into this!

I had to rework things a bit, but I'm unsure about the approach I took. I pulled out the logic to get the children considering aria-owns into a separate piece. Would be happy to try another path if it makes more sense.

One thing that I'm unsure about is if/how to support `aria-owns` in general. It seems for that case, the general tree structure (i.e. children, parentObject, nextSibling, etc) may need to be modified to account for aria-owns (or provide separate functions to access those and use those instead). I'm don't think this patch  useful step in that direction, though I would be happy with improving support for aria-owns in trees/lists/tables/grids if the larger work is large in scope.

I added a circular reference to the existing test case, but can make it a separate file if that makes more sense.

-- 
You are receiving this mail because:
You are the assignee for the bug.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.webkit.org/pipermail/webkit-unassigned/attachments/20211203/e257bcdb/attachment.htm>