ContextQMD
Back to Plate

Slate v2 HTML Formatting Phase 8 Batch

docs/plans/2026-04-06-slate-v2-html-formatting-phase8.md

53.0.63.1 KB
Original Source

Slate v2 HTML Formatting Phase 8 Batch

Supporting plan. For current queue and roadmap truth, see master-roadmap.md.

Goal

Land the next Phase 8 batch from docs/slate-v2/package-end-state-roadmap.md:

  • widen the current app-owned paste-html surface conservatively
  • keep the behavior on the slate-react runtime seam
  • prove it in browser lanes

Constraints

  • no legacy plugin-override resurrection
  • no fake “full HTML parity” claim
  • no lower-layer churn unless the proof exposes a real gap
  • prefer explicit app-owned inline element shapes over broad runtime mark support unless forced

Likely First Slice

  • support a small explicit formatting set:
    • strong
    • em
    • code
  • preserve current paragraph + link support
  • prove browser paste behavior through the existing paste-html example

Evidence

  • current paste-html surface only deserializes paragraphs and links
  • current EditableBlocks seam already supports app-owned inline rendering via renderElement
  • current EditableBlocks renderSegment does not carry text-node mark data, so mark-style rendering is not the obvious minimal seam
  • legacy ../slate/site/examples/ts/paste-html.tsx proves the broader semantic reference, but the v2 slice should stay narrower and explicit

Progress

  • loaded task, goal workflow, learnings-researcher, tdd, and react
  • checked .tmp/slate-v2 for deeper AGENTS/CLAUDE files: none found
  • read current paste-html example, runtime seam, browser test, and legacy Slate example
  • read clipboard-boundary learning: docs/solutions/logic-errors/2026-04-03-slate-v2-clipboard-boundary-proof-must-split-fragment-semantics-and-dom-transport.md
  • wrote the first red browser test for explicit strong / em / code support in playwright/integration/examples/paste-html.test.ts
  • proved the red through the maintained local browser runner
  • implemented the minimal seam:
    • parse strong / em / code into explicit inline elements
    • keep the behavior app-owned in the example layer
    • let links carry inline descendants instead of flattening to plain text
  • synced the Phase 8 / compatibility docs to the newly proved subset
  • verification green:
    • browser proof on port 3010: bash ./scripts/run-slate-browser-local.sh 3010 /examples/paste-html "yarn build:slate-browser:playwright && yarn exec playwright test playwright/integration/examples/paste-html.test.ts --project=chromium --workers=1"
    • yarn tsc:examples
    • yarn eslint site/examples/ts/components/link-utils.tsx site/examples/ts/components/paste-html-surface.tsx playwright/integration/examples/paste-html.test.ts
    • yarn prettier --check ... on changed slate-v2 files
    • pnpm exec prettier --check ... on changed plate-2 docs
  • caveat:
    • the JS example mirrors in site/examples/js/** follow an existing semicolon-heavy generated style that currently fails targeted repo ESLint even outside this slice, so lint evidence is intentionally scoped to the TS/example/test path that backs the proved behavior