Re: Proposal: :column pseudo-class

On 7/5/05, Lachlan Hunt <lachlan.hunt@lachy.id.au> wrote:
> Orion Adrian wrote:
> > On 7/5/05, Lachlan Hunt <lachlan.hunt@lachy.id.au> wrote:
> >
> >>Orion Adrian wrote:
> >>
> >>>How about this? Prevent reassignment of display in in pseudo-element
> >>>or pseudoclass.
> >>
> >>What problem, exactly, would that solve?
> >
> >
> > That when CSS is making it's pass to determine what selectors apply
> > you can't have pseudo-class selectors based on other properties
> 
> No it doesn't.  The reason you can't select an element based on
> properties, which has already been given by Hixie [1] in this thread, is
> because:
>    "The cascade happens before layout. You don't know an element's
>     display type at the time you are doing selector matching."

But you could know an element's display type before laying out a
particular element as long as you know its parents and previous
siblings display types, something that would be trivial to add to
algorithm. Since the algorithm specifies that these will already have
been rendered or at least processed, they will be known.

> > because those properties could change in the the selector that
> > includes or depends on the pseudoclass.
> 
> Circular references is *not* the core of the problem, it's just an issue
> that would result *if* it were possible, which it's not.

I wasn't talking about that.

Received on Tuesday, 5 July 2005 13:58:34 UTC