Editorial: formalize screenshot waiting algorithm

[jugglinmike]

Alternatively, we might define an algorithm for HTML to explicitly invoke during Update the rendering. There's certainly precedence for this, as HTML already signals three other specifications near the moment we're interested in here in BiDi.

By integrating with requestAnimationFrame, this approach precludes coordination with HTML, but it's not without its downsides, e.g.

• calling out to ECMAScript might be distracting for some readers
• the scheduling is observable to authors' code in that it increments the target's animation frame callback identifier
• authors' rAF callbacks may execute after this algorithm resumes

...and with that, I've all but talked myself out of this patch :P

---

Preview | Diff

[jugglinmike]

I believe the correct way to do this would be with a specific step in https://html.spec.whatwg.org/#update-the-rendering (c.f. mark paint timing) rather than invoking DOM APIs.

Great minds think alike--that's exactly what I was getting at in the description of this pull request! I've added a commit to use that approach, instead. I'm expecting folks will want to workshop the design some more, so I'm going to hold off on submitting anything to HTML. But to help explain this patch, here's how it could be integrated in its current form:

      <li><p>For each <span>fully active</span> <code>Document</code> <var>doc</var> in
      <var>docs</var> run <span>process top layer removals</span> given <var>doc</var>.</p></li>

+     <li><p>For each <span>fully active</span> <code>Document</code> <var>doc</var> in
+     <var>docs</var> run <span>resume suspended renderings</span> given <var>doc</var>'s
+     <span>browsing context</span>. <ref>WEBDRIVERBIDI</ref></p></li>
     </ol>
    </li>

[jugglinmike]

I just realized that BiDi has a formal mechanism for tracking patches to external specs. I've pushed up a commit to include the change I described in my previous comment, and I've resolved merge conflicts from recent changes in the main branch.

What do you think, @jgraham?