Re: [webcomponents]: Re-imagining shadow root as Element

On Tue, Mar 19, 2013 at 12:05 AM, Roland Steiner
<rolandsteiner@google.com> wrote:
> AFAICT this has several ramifications:
>
> .) Being an element, you now have a render box for each <shadowroot>, in
> addition to the render box for the host element.

That's not a requirement. We can define that the <shadowroot> element
doesn't show up in the flattened tree.

However allowing the shadowroot to appear in the flattened tree might
have advantages.

This is something that has been debated before. Also note that the
CSSWG has discussed the possibility of introducing "display:content"
which might be useful here.

> .) Being an element, can a <shadowroot> can itself have a <shadowroot>?.
> Contrast:
>
>     <div>
>         <shadowroot>...</shadowroot>
>         <shadowroot>...</shadowroot>
>
>     vs:
>
>     <div>
>         <shadowroot>
>             <shadowroot>
>
> .) Similar to the above: can <content> and <shadow> have <shadowroot>
> children?
>
> .) Behavior of a DocumentFragment being inserted into the document also
> needs twizzling if it contains a <shadowroot>.

All good points.

I'm definitely concerned that allowing the <shadowroot> to appear in
the markup causes more confusion than it's worth. I'm also not sure
what problems that it solves over simply adding real children to the
DOM.

/ Jonas

Received on Tuesday, 19 March 2013 20:58:10 UTC