Keep the keyboard focus where it was when a revealjs presentation is shown again

[jooyoungseo]

Description

Fixes #15079.

Each time the page becomes visible again, for example after switching tabs, the bundled reveal.js moved the focus to <body> if anything else had it (onPageVisibilityChange()). A reader lost their place on a link or in a chart. NVDA and JAWS, which pass the arrow keys on only while the focus is in the deck, went back to browse mode.

reveal.js does this "to ensure we can use keyboard shortcuts". But keys pressed on a link or a button reach it anyway: it ignores keys only in inputs, text areas and editable content. The keys it cannot get are the ones pressed inside a frame. So the new source patch revealjs-0003-focus-on-return.patch moves the focus only out of a frame, and onto the deck (with tabindex="-1") rather than the body. It blurs the frame first, as reveal.js did: Playwright's Firefox, which CI runs, otherwise leaves the focus in the frame. Anywhere else, the focus stays where the reader left it.

The patch follows #15020 and #15034. update-html-dependencies.ts applies it after 0001 and 0002, and the bundles and maps are rebuilt from 5.1.0 with npm ci --ignore-scripts and gulp js, as that step does. Before adding the patch, the same build with 0001 and 0002 alone reproduced all four committed files byte for byte. That needed the sources the committed maps embed with CRLF line endings to have CRLF line endings in the build too. With 0003, only js/reveal.js's source and the mappings change in the maps.

Every reveal.js release from 3.9.2 to 6.0.2 does this, so I'm reporting it to reveal.js as well.

Validation

• revealjs-visibility-focus.spec.ts, with a new fixture, checks what happens after the page is hidden and shown again:

  • a focused link keeps the focus;
  • a focused frame hands it to the deck;
  • the deck keeps it;
  • a focus on the body stays there;
  • after the first two, → still changes slides.

  Playwright does not hide the page when another tab comes to the front, so the test marks the page hidden and then visible and dispatches visibilitychange. The first three fail without the patch.

• All revealjs Playwright specs pass in Chromium, except revealjs-multiplex.spec.ts, which needs the multiplex server that the CI wrapper starts. revealjs-visibility-focus.spec.ts also passes in Firefox 158 over WebDriver BiDi. WebKit and Playwright's own Firefox were not run locally.

• The three patches apply in order to a pristine 5.1.0 tree with git apply --ignore-space-change.

Checklist

I have (if applicable):

• referenced the GitHub issue this PR closes
• updated the appropriate changelog in the PR
• ensured the present test suite passes (revealjs Playwright specs in Chromium only; full suite not run)
• added new tests
• created a separate documentation PR in Quarto's website repo and linked it to this PR (not applicable)

AI-assisted PR

• AI tool used: an AI coding assistant
• Codebase grounding: Local clones of Quarto main and reveal.js 5.1.0, the reproduction in revealjs: switching to another tab and back moves the keyboard focus to the page body #15079, the existing reveal.js patch workflow, and the repository's Playwright tests and contribution guide.
• Human review: I reviewed the changes and test results and approved this submission. The screen reader behaviour described in revealjs: switching to another tab and back moves the keyboard focus to the page body #15079 is from my own testing with NVDA and JAWS.

---

Generated by Claude Code

[Copilot]

Copilot was unable to review this pull request because the user who requested the review has reached their quota limit.

[posit-snyk-bot]

✅ Snyk checks have passed. No issues have been found so far.

Status	Scan Engine	Critical	High	Medium	Low	Total (0)
✅	Open Source Security	0	0	0	0	0 issues
✅	Licenses	0	0	0	0	0 issues

💻 Catch issues earlier using the plugins for VS Code, JetBrains IDEs, Visual Studio, and Eclipse.