Re: [SelectorsAPI] Thoughts on querySelectorAll

On Wednesday 2008-04-30 15:24 -0700, John Resig wrote:
> > I presume what you want instead is that element-rooted queries are
> > handled by matching such that each simple selector (CSS2 definition)
> > or sequence of simple selectors (css3-selectors definition, and I
> > *still* object to changing the meaning of existing terms) must match
> > an element that is or is a descendant of the specified element (the
> > root of the query).
> > 
> > Or do you want to eliminate "is or" in my previous paragraph, to
> > exclude the query root?  Or something else?
> 
> I would exclude "is or" - only descendant elements, in this case. For example:

But then you'd need to have an exception to that rule to allow
":scope > div" to work.  If you want to be able to do ":scope >
div", it seems preferable to have the "is or" rather than have some
subset of selectors be special.  (Which ones?  Just ":scope > div"
or also "p:scope > div"?)

-David

-- 
L. David Baron                                 http://dbaron.org/
Mozilla Corporation                       http://www.mozilla.com/

Received on Thursday, 1 May 2008 00:36:19 UTC