Add shadowrootslotassignment attribute

[keithamus]

Implements the shadowrootslotassignment attribute.

Refs WICG/webcomponents#967

• At least two implementers are interested (and none opposed):
  • Mozilla: Interested
  • Chromium: Assuming supportive based on Add shadowrootslotassignment attribute #12267 (comment)
  • WebKit: Supportive position pending: Support shadowrootslotassignment= attribute WebKit/standards-positions#631
• Tests are written and can be reviewed and commented upon at:
  • Tenative PR: Add tests for shadowrootslotassignment web-platform-tests/wpt#58546
  • …
• Implementation bugs are filed:
  • Chromium: https://issues.chromium.org/issues/493315747
  • Gecko: https://bugzilla.mozilla.org/show_bug.cgi?id=2023824
  • WebKit: https://bugs.webkit.org/show_bug.cgi?id=310090
  • Deno (only for timers, structured clone, base64 utils, channel messaging, module resolution, web workers, and web storage): N/A
  • Node.js (only for timers, structured clone, base64 utils, channel messaging, and module resolution): N/A
• Corresponding HTML AAM & ARIA in HTML issues & PRs: N/A
• MDN issue is filed: …
• The top of this comment includes a clear commit message to use.

(See WHATWG Working Mode: Changes for more details.)

---

/indices.html ( diff )
/parsing.html ( diff )
/scripting.html ( diff )

[keithamus]

/cc @mfreed7

[rniwa]

It seems that we also need a mechanism to specify which nodes are assigned in what order.

[keithamus]

It seems that we also need a mechanism to specify which nodes are assigned in what order.

Right. I think that it would be possible to ship the two independently, right? It's possible today to create a manual assignment shadowroot and never assign nodes.

[rniwa]

It seems that we also need a mechanism to specify which nodes are assigned in what order.

Right. I think that it would be possible to ship the two independently, right? It's possible today to create a manual assignment shadowroot and never assign nodes.

Maybe? It seems like we want some kind of plan for it before we this feature though.

[annevk]

You also need to add the IDL attribute for the template element.

[mfreed7]

/cc @mfreed7

This looks great to me! I could finally address this 4 year old Chromium TODO.

Right. I think that it would be possible to ship the two independently, right? It's possible today to create a manual assignment shadowroot and never assign nodes.

+1 - I think this PR provides perfect parity to the imperative attachShadow() call, which just sets the slotAssignment mode but doesn't actually assign nodes. That might be enough, in that it keeps slots from being automatically assigned to slots before the imperative code boots up and assigns them.

[keithamus]

Maybe? It seems like we want some kind of plan for it before we this feature though.

AIUI the design described in WICG/webcomponents#967 adds a new attribute to slot elements. To me this stands to reason that these become opt-in, and therefore the design is additive - which is to say if we go with the proposed solution there, I believe they can ship independently.

So I don't think landing this now would close off the design space for us; landing this will not preclude us from the design in WICG/webcomponents#967, and I suspect it doesn't close off any other potential solutions in a way which would still make sense, but I'd be happy to be corrected on this.