Migrate record-update-link onto the record-mutation lifecycle driver#58
Merged
Conversation
record-update-link already had the record-mutation family's exact execute shape — prepare -> seedReady -> trace-wrapped bulk update -> folded post-op verify, restore-or-delete cleanup, phases [prepare, <metric>, seedReady] — identical to record-update. It now rides record-mutation-lifecycle instead of hand-rolling that protocol. It is the first member whose fixture spans more than one table (a host + linked foreign pair). The driver already treats the fixture opaquely, so this required no driver code change — only relaxing the scope note's "single seeded table" assumption. assertSeedReady stays defined (seed and execute both assert the seed link permutation), so seed and execute route through one shared spec, like record-update/create. Mapping details: - runMeasuredOperation resolves foreign ids (unmeasured), then runs the trace-wrapped updateRecords + routing assertion + sample/full-scan verify, bundled into one primary measurement. - cleanup re-resolves foreignIdByTitle (cleanup is unmeasured) to restore the seed permutation, or drops both tables if restore fails. - buildResult already matched the driver arg shape; the spec delegates to it unchanged (dropping the unused windowId). Verification (local, execute, v1+v2): - pnpm check green. - baseline A vs B (record-update-link): 2/2 pass, zero new diff masks. - G1 baseline vs candidate: 10/10 pass — record-update-link + record-update/ record-create/record-reorder/selection-clear regression (shared driver scope-note touched). - negative: phases[0].name / details.operation / routing.routeMatched / fullScan.scannedRecords / seed.ready.expectedTitle perturbations fail; masked metric value still passes. - candidate artifacts: all 10 result=pass; record-update-link routeMatched=true, actualXTeableV2 matches engine, fullScan.scannedRecords==rowCount(1000), updatedRecords==1000, bulkUpdate1kLinkCellsMs far below max. Tracker: 19/35 runner kinds, 23/55 cases. Co-Authored-By: Claude <noreply@anthropic.com>
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
What
Migrate the record-update-link runner kind (1 case:
record-update/1k-link-cells-bulk-update) onto the existingrecord-mutation-lifecycle.tsdriver, reusing it (5th member afterrecord-update / record-create / record-reorder / selection-clear).
record-update-link already had the family's exact execute shape —
prepare → seedReady → trace-wrapped bulk update → folded post-op verify,restore-or-delete cleanup, phases
[prepare, <metric>, seedReady]— identicalto record-update. The legacy file just hand-rolled that protocol.
It is the first member whose fixture spans two tables (a host + linked
foreign pair). The driver already treats
TFixtureopaquely, so this neededno driver code change — only relaxing the scope note's "single seeded
table" assumption.
assertSeedReadystays defined (seed and execute bothassert the seed link permutation), so both modes route through one shared spec.
Mapping:
runMeasuredOperationresolves foreign ids (unmeasured setup) → trace-wrappedupdateRecords+ routing assertion + sample/full-scan verify, bundled intoone primary measurement.
cleanupre-resolvesforeignIdByTitle(cleanup is unmeasured) to restorethe seed permutation, or drops both tables if restore fails.
buildResultalready matched the driver arg shape; the spec delegates to itunchanged (dropping the unused
windowId, no record window).Verification (local, execute, v1+v2)
pnpm checkgreen.record-update / record-create / record-reorder / selection-clear regression
(shared driver scope-note touched).
phases[0].name,details.operation,routing.routeMatched,fullScan.scannedRecords,seed.ready.expectedTitleeach fails the diff; a masked metric-value change still passes.
result=pass; record-update-linkrouteMatched=true,actualXTeableV2matches engine,fullScan.scannedRecords == rowCount (1000),updatedRecords == 1000,bulkUpdate1kLinkCellsMs(2793ms v1 / 531ms v2) far belowmax90000.Protected surface
No changes to
cases/**,registry.ts,framework/types.tsconfig interfaces,framework/artifacts.ts,scripts/diff-artifacts.mjs, or../teable-ee. Trackerupdated to 19/35 runner kinds · 23/55 cases.
🤖 Generated with Claude Code