Clarification on animation frame task

[Johanna-hub]

Currently in the spec we have:

The following is used but not defined in [[FULLSCREEN]]: animation frame task.

According to the fullscreen issue referenced above the timing is now defined and the issue has been closed.

@foolip Would you please be able to clarify what this means for the definition in the Screen Orientation spec and also where animation frame task appears in the algorithms:

If locking the orientation did not result in a change of orientation, as part of the next animation frame task, resolve pending-promise with undefined and set pending-promise to null.

Whenever the viewport's angle changes, the user agent MUST run the following steps as part of the next animation frame task:

Whenever a document becomes visible per [[!PAGE-VISIBILITY]], in other words after the now visible algorithm is run, the user agent MUST run the following substeps as part of the next animation frame task:

Thank you very much for your help 🙂

[foolip]

The way to integrate with the event loop that we ultimately landed on is https://fullscreen.spec.whatwg.org/#run-the-fullscreen-steps. There is a step in https://html.spec.whatwg.org/multipage/webappapis.html#update-the-rendering that calls out to HTML.

This does mean that HTML needs to know about all the specs that integrate with it, which is arguably backwards layering-wise, but the upside is that the order of things is clearly defined, and cannot really be well defined without a central point of order.

[Johanna-hub]

@foolip Thanks so much

[marcoscaceres]

Closing as duplicate of #219