Re: Extending Class Proposal

On Thu, Jan 31, 2013 at 1:10 PM, Tab Atkins Jr. <jackalmage@gmail.com>wrote:

> On Thu, Jan 31, 2013 at 12:28 PM, Philip Walton <philip@philipwalton.com>
> wrote:
> > I agree with you that in CSS, whenever a selector appears that you've
> > extended, you need to simply pretend those extended selectors are there
> as
> > well. This make sense because you're being declarative and building
> styles.
> >
> > I dont think it works as well in JavaScript because the function is
> actually
> > returning a list of DOM nodes. It's not backwards compatible. If tomorrow
> > Chrome suddenly started including extended elements in the set of
> returned
> > nodes, you'd break a lot of websites. That's why I think a new simple
> > selector would work best.
>
> No, we wouldn't break anyone, *because you can't extend anything yet*.
>

I wasn't accusing you or Chrome of breaking anything, I was simply pointing
out why the current Sass syntax, which you were promoting in a previous
email, wouldn't lend itself well to a querySelectorAll function.

We'll only have a compat problem if we add the ability to extend
> selectors, but don't cover the querySelector case, and it doesn't get
> properly updated in tandem.


Yes, obviously. I was offering a suggestion as to how to cover the
querySelector case. If you didn't like the suggestion or had other ideas
for how to address the issue, I'd love to hear them.