Re: [SVGMobile12] Processing model for unknown elements

Hi Chris.

Chris Lilley:
> >   Within an SVG document fragment, unknown elements, including unknown
> >   elements in the SVG or XML Events namespaces as well as known elements
> >   in the SVG or XML Events namespaces occuring in unexpected locations,
> >   and their descendant elements do not participate in SVG rendering and
> >   are not processed beyond their contribution to the construction of the
> >   DOM. The processing model for unknown elements is equivalent to an
> >   <svg:g> element with the 'display' property set to 'none'.
> > 
> > I think it should be made clear what the "processing model" here means.
> > Does it mean that the unknown elements have DOM objects that implement
> > SVGLocatableElement, since that is what 'g' elements have?  Or just that
> > the rendering model is the same as for a "display: none"ed group (which
> > is what I suspect)?
> 
> You are right. We suggest rewording the last sentence to start "The
> rendering for unknown elements is ..."
> 
> Given that, is it clear enough or do we need to add something like "Note
> that elements outside the SVG namespace do not implement any
> SVG-specific DOM methods" but thought that was a bit tautologous and
> would be exactly the sort of sentence that we would later wonder why we
> put it in.

I don’t think it is necessary to mention that.

> And it does mean that elements in the 'known elements occurring in
> unexpected locations' still implement their normal interfaces.

Are the ‘known elements occurring in unexpected locations’ meant to
implement their normal DOM interfaces still?  (I don’t know if you mean
that that is what your suggested tautologous sentence would imply,
correctly or incorrectly, or that that is what is meant to be reflected
in the spec.)  If so, perhaps this text is clearer:

  Within an SVG document fragment, any subtree that is rooted by an
  unknown element (including those in the SVG or XML Events namespaces)
  or a known element that occurs in an unexpected location, is not
  rendered.  The nodes in the subtree are not processed beyond
  including the relevant DOM objects in the document tree.  These DOM
  objects will still implement the DOM interfaces appropriate for the
  element type, however.

> Does that cover your comment?

Yes, rewording the existing sentence to say “The rendering for unknown
elements is…” satisfies my comment, but consider some clearer text such
as that suggested above.

-- 
Cameron McCormack, http://mcc.id.au/
 xmpp:heycam@jabber.org  ▪  ICQ 26955922  ▪  MSN cam@mcc.id.au

Received on Friday, 2 June 2006 06:09:12 UTC