Re: [Syntax Level 3]

On Tuesday 2012-08-28 08:41 -0700, Tab Atkins Jr. wrote:
> On Tue, Aug 28, 2012 at 4:34 AM, Brian Kardell <bkardell@gmail.com> wrote:
> > !/*really?*/important;
> >
> > Can't reps from google, ms, etc see if they can verify a single instance of
> > this in the wild?  That is just too weird, I don't generally put things
> > beyond people, but I just honestly can't imagine someone doing that for
> > real...
> 
> Pending an investigation by me, I've just made the spec disallow that,
> and require the DELIM(!) and IDENT(important) tokens to be immediately
> adjacent.  Otherwise I have to add a stack that remembers what was
> seen, so I can push them all into the property if I don't eventually
> see the IDENT(important) that I expect. :/
> 
> This is noted as a difference from CSS2.1 in the list at the end of
> the parser, so I'll remember.

I don't see why it's worth making this change.

Does your parsing spec have a framework so inflexible that it can't
describe the current CSS rules?  Why?  And if so, how can it
possibly describe the rest of CSS?

-David

-- 
𝄞   L. David Baron                         http://dbaron.org/   𝄂
𝄢   Mozilla                           http://www.mozilla.org/   𝄂