Re: [w3c/webcomponents] Algorithms in 6.3. Sequential Focus Navigation are too complex (#496)

I've put an example in JSbin (and will include in the spec, if it makes sense).
http://jsbin.com/cubiye/edit?html,output

The document has 3 shadow roots, and each shadow root has 2 slots.
If we run the algorithm described in 6.3,

- Step 1: Scope owners are identified: document, shadow roots, slots
    - document, 3 shadow roots (under `#host1`, `#host2`, `#host3`), 6 slots (2 slots in each shadow tree)
- Step 2: All elements in document and shadow trees are identified to which scope it belongs
    - e.g. `<input>` in slotted `<div>` belongs to the `<div>`'s assigned slot. So `<input>`s under `#host1`'s shadow labeled `RA1` and `RA2` will belong to its `<slot name="A">`, etc.
- Step 3: For each scope, navigation order is determined
    - e.g. in `#host1`'s `<slot name="A">` scope, the order is `RA1` to `RA2`, determined by their tabindices. `#host2` is excluded because it has `tabindex="-1"`.
- Step 4: Document-wide navigation order is determined
    - e.g. all the focusable areas in `#host1` will be visited after `<input>` labeled `D4`. All `<input>`s under `#host2` will be skipped because it was excluded in the previous step.

I hope this kind of examples would help understanding the spec.
Any feedback is welcome.

---
You are receiving this because you are subscribed to this thread.
Reply to this email directly or view it on GitHub:
https://github.com/w3c/webcomponents/issues/496#issuecomment-227940954