Re: Selectors erratum (was Re: Various comments on selectors' Test)

> On Mon, 7 Feb 2000, Matthew Brealey wrote:
> >> --- "Eric A. Meyer" <emeyer@sr71.lit.cwru.edu> wrote:
> >>
> >> BODY P is a selector, with BODY a simple selector, P a simple
> >> selector, and ' ' a combinator (defined thus in the grammar:
> >> : '+' S* | '>' S* | /* empty */)
> >
> > I think the use of descendant selector in the spec must be
> > considered erroneous - if they are selectors, they are exactly the
> > same as CSS-1 contextual selectors, and therefore the change is
> > pointless, so therefore they must be combinators.
> 
> CSS1's contextual selectors is a generic name for selectors which use
> what in CSS2 became combinators.
No.

<q cite="http://www.w3.org/TR/REC-CSS1#contextual-selectors">Contextual
selectors consist of several simple selectors separated by
whitespace</q>

<q
cite="http://www.w3.org/TR/REC-CSS2/selector.html#descendant-selectors">
A descendant selector is made up of two or more selectors separated by
whitespace. 
</q>

[snip]

> (in particular the sentence reading "Contextual selectors in CSS1 look for
> ancestor relationships, but other relationships (e.g. parent-child)
> may be introduced in later revisions.").

The necessary implication from the quotation above is that these would
not be contextual because contextual selectors are 'simple selectors
separated by whitespace'.