1452640 - Implement functional :host(..) selector.

Status: RESOLVED FIXED

(Core :: CSS Parsing and Computation, enhancement)

Comment 1

[Emilio Cobos Álvarez (:emilio)]

Attachment: Bug 1452640: Implement the functional :host(..) selector.

We could invalidate in a slightly more fine-grained way, but I don't think it's
worth the churn vs. keeping the special-cases minimal.

Review commit: https://reviewboard.mozilla.org/r/235088/diff/#index_header
See other reviews: https://reviewboard.mozilla.org/r/235088/

Comment 2

[Xidorn Quan [:xidorn] UTC+11]

Comment on attachment 8966386 [details]
Bug 1452640: Implement the functional :host(..) selector.

https://reviewboard.mozilla.org/r/235088/#review240806

::: servo/components/selectors/parser.rs:824
(Diff revision 1)
> +    /// NOTE(emilio): This should support a list of selectors, but as of this
> +    /// writing no other browser does, and that allows them to put ::slotted()
> +    /// in the rule hash, so we do that too.

I suppose you mean "to put **`::host()`**"?

::: servo/components/selectors/parser.rs:1137
(Diff revision 1)
>              Scope => dest.write_str(":scope"),
> -            Host => dest.write_str(":host"),
> +            Host(ref selector) => {
> +                dest.write_str(":host")?;
> +                if let Some(ref selector) = *selector {
> +                    dest.write_char('(')?;
> +					selector.to_css(dest)?;

The indention looks weird.

::: servo/components/style/invalidation/element/invalidator.rs:538
(Diff revision 1)
> -        // element state / attribute invalidations (it's needed for XBL though,
> -        // due to the weird way the anon content there works (it doesn't block
> -        // combinators)).
> +        // correctly on :host(..) changes. We could not do this instead and add
> +        // a third kind of invalidation list instead, that walks shadow root
> +        // children, but it's not clear it's worth it.

I find this sentence pretty hard to follow...

Maybe something like:
> Instead of doing this, we could add a third kind of invalidation list that walks shadow root children, but it's not clear it's worth it.

Comment 3

[Pulsebot]

Pushed by [e-mail]:
https://hg.mozilla.org/integration/autoland/rev/77d5f1b8d04f
Update test expectations for :host(..). r=me
https://hg.mozilla.org/integration/autoland/rev/00ac14e01121
Test descendant invalidation for :host(..). r=xidorn

Comment 4

[Raul Gurzau (:RaulG)]

https://hg.mozilla.org/mozilla-central/rev/77d5f1b8d04f
https://hg.mozilla.org/mozilla-central/rev/00ac14e01121

Comment 5

[Web Platform Test Sync Bot (Matrix: #interop:mozilla.org)]

Created web-platform-tests PR https://github.com/w3c/web-platform-tests/pull/10408 for changes under testing/web-platform/tests

Comment 6

[Web Platform Test Sync Bot (Matrix: #interop:mozilla.org)]

Upstream PR merged