[11/16] refactor(painter): remove blockLookup — painter reads only resolved data#2828
Conversation
tupizz
changed the title
tupizz
force-pushed
the
tadeu/sd-2563-10-provider-resolved-items
branch
from
84f6ae2 to
d7e10ba
Compare
tupizz
force-pushed
the
tadeu/sd-2563-11-delete-blocklookup
branch
from
ead112c to
2d2ce0d
Compare
tupizz
force-pushed
the
tadeu/sd-2563-11-delete-blocklookup
branch
from
2d2ce0d to
11ca970
Compare
Codecov Report✅ All modified and coverable lines are covered by tests. 📢 Thoughts on this report? Let us know! |
|
🎉 This PR is included in superdoc v1.28.0-next.12 The release is available on GitHub release |
|
🎉 This PR is included in superdoc-cli v0.8.0-next.12 The release is available on GitHub release |
|
🎉 This PR is included in superdoc-sdk v1.6.0-next.48 |
|
🎉 This PR is included in superdoc v1.29.0 The release is available on GitHub release |
|
🎉 This PR is included in @superdoc-dev/mcp v0.3.0-next.1 The release is available on GitHub release |
|
🎉 This PR is included in superdoc-sdk v1.8.0-next.1 |
|
🎉 This PR is included in esign v2.6.0 The release is available on GitHub release |
|
🎉 This PR is included in template-builder v1.7.0 The release is available on GitHub release |
|
🎉 This PR is included in superdoc-cli v0.8.0 The release is available on GitHub release |
|
🎉 This PR is included in superdoc-sdk v1.8.0 |
|
🎉 This PR is included in @superdoc-dev/mcp v0.3.0 The release is available on GitHub release |
3 participants
What this PR does
Removes the
blockLookup: Map<string, BlockLookupEntry>fromDomPainter. After the previous refactors (#2810–#2827), every piece of data the painter reads at paint time is available on the resolved layout. The lookup map and the bridge fields that fed it (blocks,measures,headerBlocks,headerMeasures,footerBlocks,footerMeasuresonDomPainterInput) are now dead — this PR deletes them.Finishes the promise of SD-2563: painter is now a "dumb" rendering stage that consumes fully-resolved
ResolvedLayoutand produces DOM. No upstream lookups, no fallback paths, no mixed-data bridges.Key changes
DomPainterInputshape shrinks to{ resolvedLayout, sourceLayout }— body/header/footer block-measure pairs removed.DomPainter.#updateBlockLookup,#normalizeOptionalBlockMeasurePair,#resolveBlockAndMeasure,#resolveBlock,#buildBlockLookupall deleted.isHeaderWordArtWatermarknow takes the drawing block from itsResolvedDrawingIteminstead of looking it up byblockId.hasFragmentGeometryChanged(added to main by fix: rebuild fragment on geometry change #2842) preserved as a standalone helper — it was semantically independent ofblockLookup.createDomPainterlegacy layout-only input path no longer synthesizes header/footer lookups.PresentationEditor.#paintLayoutstops gathering header/footer blocks+measures into the paint input.BlockLookup—group-analysisreads resolved paragraph borders directly.index.test.ts: "rejects resolved-layout-only paint input…", "uses resolved block versions…", "uses resolved continuesFromPrev…";PresentationEditor.test.ts: "passes body blocks and measures to the painter…") — each asserted fields that no longer exist onDomPainterInput.Diff shape
Verification
Ran against commit on the rebase worktree before push:
pnpm type-check(tsc -b across all packages)pnpm build(full workspace)pnpm --filter @superdoc/painter-dom buildpnpm --filter @superdoc/painter-dom testpnpm --filter @superdoc/pm-adapter testpnpm --filter super-editor test -- --run presentation-editorRebase history
ERR_PNPM_OUTDATED_LOCKFILEon the stacked base and renderer drift from main's#2842(hasFragmentGeometryChanged). Resolved by resetting to the new [10/16] refactor(painter): deliver resolved items to decoration provider callback #2827 tip and cherry-picking the single PR 11 commit.main. Branch still carried the pre-squash PR 10 cherry-pick (d7e10ba7f), which conflicted with the squashed PR 10 in main (9c6ccb0bc) — same content, different SHAs. Fixed by resetting to freshorigin/mainand cherry-picking only the PR 11 commit. Main's new commits#2922(tracked changes in headers/footers) and#2796(image rendering) auto-merged cleanly against PR 11.PR Stack (SD-2563: Dumb Painter Refactor)
Status legend: ✅ merged to
main· 🔁 closed when merged into stack parent, not inmainyet (need to re-open stacked on top of this once it lands) · 👈 this PRAfter this lands
The downstream PRs (#2829, #2845, #2853, #2906, #2909) need to be reopened as a new stack on top of
mainto finish SD-2563. None of them currently live inmain— they only merged into their respective stack parents, which never flowed through.