@scope

[andruud]

Request for Mozilla Position on an Emerging Web Specification

• Specification Title: CSS Cascading and Inheritance Level 6 (parts about @scope).
• Specification or proposal URL: https://drafts.csswg.org/css-cascade-6/#scope-atrule
• Caniuse.com URL (optional): N/A
• Bugzilla URL (optional): N/A
• Mozillians who can provide input (optional): @emilio

Other information

At this time I'm requesting a position for @scope only, not including:

• Scoped Descendant Combinator, e.g. .a >> .b.
• Selector Scoping Notation, e.g. (.foo / .bar) .a { ... }.

[emilio]

Over-all seems reasonable, though I'm a bit confused/concerned about the :scope rules. The spec says:

Selectors are scoped to the given scope, with the :scope element being the scoping root.

I think the :scope definition is circular there (because the scoping root is defined to be :scope), but ignoring that... I think that means that you need to compute the scope for a given selector before you start matching it, right? Which means you need to do a bunch of extra work for selectors that don't end up matching, which is a bit unfortunate... Of course you can do the bloom filter rejection earlier, presumably, but still..

[andruud]

Yeah, :scope is the most annoying part here.

I think that means that you need to compute the scope for a given selector before you start matching it, right?

I was exploring the possibility of dynamically figuring out the scope element when :scope is encountered in the selector, so at least for e.g. :scope > .foo we'd check .foo first before doing horrible things to check for :scope. But that has some challenges as well ...

[zcorpan]

Closing as dup of #472