Re: [w3c/webcomponents] slotchange event should not bubble (#571)

I think this might just be a Canary bug?

Here's the output I'm seeing:
```
Event {isTrusted: true, type: "slotchange", target: slot#slot, currentTarget: slot#slot, eventPhase: 2…} "bubbled" false
Event {isTrusted: true, type: "slotchange", target: slot#slot, currentTarget: slot#innerSlot, eventPhase: 3…} "bubbled" true
Event {isTrusted: true, type: "slotchange", target: slot#innerSlot, currentTarget: slot#innerSlot, eventPhase: 2…} "bubbled" false
```
(eventPhase: 2 is AT_TARGET, 3 is BUBBLING_PHASE)

I don't think the second event should happen in this situation:
- The `a` was assigned to `#slot`, so the first event is certainly meant to occur.
- The '[signal a slot change](https://dom.spec.whatwg.org/#signal-a-slot-change)' algorithm also causes 'signal a slot change' to run on the slot's assigned slot (`#slot`'s assigned slot is `#innerSlot`), so the third event makes sense.
- The second event - with target `#slot`, current target `#innerSlot`, and in the *bubbling* phase - doesn't make sense either way its shown state is interpreted:
    - This can't be the (correct) bubbling phase of the first event, because these slots are in different ShadowRoots and the event is *not* composed.
    - This can't be the (correct) result of 'signal a slot change' propagating 'slotchange' because it does not have target `#innerSlot` and is in the bubbling phase.

-- 
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/571#issuecomment-248385715