feat: yamazumi reporting with mode-specific KPIs and activity breakdown#19
Merged
Conversation
…ng CI Update pnpm.overrides to pin patched versions of flatted (3.4.2), h3 (1.15.9), and fast-xml-parser (5.5.8) to resolve 3 high-severity CVEs in eslint/astro/azure-sdk dependency chains. Co-Authored-By: ruflo <ruv@ruv.net>
Update pnpm.overrides for devalue from 5.6.3 to 5.6.4 to fix prototype pollution in devalue.parse and devalue.unflatten. Co-Authored-By: ruflo <ruv@ruv.net>
Implement ADR-038: calculate short-term Cp and Cpk per subgroup, plot dual series on I-Chart with separate control limits, and show capability distribution on Boxplot. The gap between Cp and Cpk visualizes centering loss — key for supplier capability stability. Core: subgroupCapability.ts (grouping, calculation, control limits) Hooks: useCapabilityIChartData, useCapabilityBoxplotData Charts: IChartBase dual series support (secondary data/stats/legend) UI: CapabilityMetricToggle, SubgroupConfigPopover State: standardIChartMetric + subgroupConfig with full persistence Wiring: both PWA and Azure dashboards with toggle + config Co-Authored-By: ruflo <ruv@ruv.net>
…dology Add capability stability awareness to CoScout so it can reason about per-subgroup Cp/Cpk data, centering loss, and out-of-control subgroups. Includes new glossary concept, AIContext field, system prompt instructions, app-level wiring, unit tests, and documentation updates across 5 docs. Co-Authored-By: ruflo <ruv@ruv.net>
Design spec for adapting the 6-section report structure for yamazumi mode with mode-specific titles, KPI grid, finding-driven Step 2 across all modes, activity breakdown with lean tooltips, and VA Ratio learning loop. Co-Authored-By: ruflo <ruv@ruv.net>
Design spec for extending subgroup capability with time-interval grouping (minute/hour/day/week) using existing time column detection infrastructure. Includes new use case for capability stability over time. Co-Authored-By: ruflo <ruv@ruv.net>
Co-Authored-By: ruflo <ruv@ruv.net>
Co-Authored-By: ruflo <ruv@ruv.net>
…port Co-Authored-By: ruflo <ruv@ruv.net>
…dings Co-Authored-By: ruflo <ruv@ruv.net>
Third subgrouping method alongside column/fixed-size: group by time intervals (5/15/30 min, hourly, daily, weekly) using detected timestamp columns. Smart UI filtering hides minute/hour options for date-only data. Co-Authored-By: ruflo <ruv@ruv.net>
- Step 1: yamazumi shows YamazumiKPIGrid + chart; standard unchanged - Step 2: finding-driven for all modes; yamazumi shows activity breakdown - Step 6: learning loop uses VA Ratio metric for yamazumi mode - Pass analysisMode to useReportSections for mode-specific titles Co-Authored-By: ruflo <ruv@ruv.net>
…i docs Co-Authored-By: ruflo <ruv@ruv.net>
|
The latest updates on your projects. Learn more about Vercel for GitHub.
|
9 tasks
jukka-matti
added a commit
that referenced
this pull request
May 14, 2026
- investigations.md "8f followups" entry: STATUS → RESOLVED 2026-05-14, PR #166 squash-merged as cd93691. LOW #16 (Canvas/index.tsx 1122-line refactor) + LOW #19 (brand ProcessHubId) explicitly carried forward. - investigations.md gains two new entries from the --chrome walk: (1) pre-existing setState-in-render warning from AppMain across canvas transitions (App.tsx untouched by #166), (2) designer-lens canvas journey-clarity observations (9 UX items: Frame→Canvas rename, desktop LODSwitcher parity, Lock-canvas mode-toggle relabeling, L1 capability empty-state copy, L3 CTA visual weight, etc.). - decision-log.md "8f canvas viewport SHIPPED" entry: Amendment 2026-05-14 followups complete block prepended; original 2026-05-13 amendment retained as history. Co-Authored-By: ruflo <ruv@ruv.net>
jukka-matti
added a commit
that referenced
this pull request
May 14, 2026
…emoval
Define `ProcessHubId = string & { readonly __brand: 'ProcessHubId' }` in
`packages/core/src/processHub.ts` (the only definition point). Add
`asProcessHubId()` (throws on empty) and `isProcessHubId()` predicate;
update `DEFAULT_PROCESS_HUB_ID` and `normalizeProcessHubId` to return
`ProcessHubId`. Re-export from stores for back-compat consumption.
Replace `hubId ?? '__wall-canvas-unbound__'` sentinel in WallCanvas with
`hubId ?? null`; update `useCanvasViewportInput` to accept `ProcessHubId | null`
with a null short-circuit (no store writes when hub is absent).
25-file sweep across packages/hooks/ui/apps: all `string`-typed hub slots
at store boundaries cast/narrowed to `ProcessHubId`; test fixtures use
`const h = (id: string) => id as ProcessHubId` helper. Both app lifecycle
hooks (PWA + Azure) use `normalizeProcessHubId(hubId)` → `boundHubId` so
the raw `string | null | undefined` param never reaches store calls.
Closes LOW #19 from the 8f retrospective (docs/investigations.md updated).
Co-Authored-By: ruflo <ruv@ruv.net>
6 tasks
jukka-matti
added a commit
that referenced
this pull request
May 14, 2026
… Canvas decomposition + tsc hygiene (#168) * fix(pwa): eliminate setState-in-render warning by replacing bare usePanelsStore() with individual field selectors The bare `const store = usePanelsStore()` whole-store subscription in `useAppPanels` violated the CLAUDE.md/ADR-078 "Never bare useStore()" rule. In React 19 Strict Mode + Zustand 5 (useSyncExternalStore), the whole-store snapshot causes tearing detection to fire on every panelsStore update (panel-state transitions co-incident with LOD switches, frame-tab activation, Lock canvas toggle), producing 8+ "Cannot update a component (AppMain) while rendering" warnings per interaction. Fix: rewrite useAppPanels.ts to use 24 individual usePanelsStore(s => s.field) selectors — one per state field and action. useEffect dependency arrays cleaned accordingly (no more `store` object in deps). App.tsx return interface unchanged. Regression test added in apps/pwa/src/__tests__/App.test.tsx: mounts App, exercises showFrame() panel transition, asserts zero setState-in-render warnings. All 40 PWA test files green; pwa build green. Co-Authored-By: ruflo <ruv@ruv.net> * docs(investigations): add commit hash + correct rule attribution Spec-review followup: add commit hash 6c5bc1a to the setState-in-render resolution paragraph, and correct rule attribution from "CLAUDE.md / ADR-078" to "packages/stores/CLAUDE.md:18 (cites ADR-041)" — ADR-078 covers the 6-stores-across-3-layers model, not selector discipline. Co-Authored-By: ruflo <ruv@ruv.net> * refactor(pwa): extract openDataTableAtRow compound action + drop dead _vi import Code-review followup: per feedback_fix_absorbed_violations_at_seam, lift the two bare usePanelsStore.setState() calls in useAppPanels.openDataTableAtRow into a single store-action openDataTableAtRow(index, isDesktop). Drops the last bare-setState seam violation in the hook. Also removes a dead `vi as _vi` import from App.test.tsx. Co-Authored-By: ruflo <ruv@ruv.net> * feat(cleanup): brand ProcessHubId opaque type + WallCanvas sentinel removal Define `ProcessHubId = string & { readonly __brand: 'ProcessHubId' }` in `packages/core/src/processHub.ts` (the only definition point). Add `asProcessHubId()` (throws on empty) and `isProcessHubId()` predicate; update `DEFAULT_PROCESS_HUB_ID` and `normalizeProcessHubId` to return `ProcessHubId`. Re-export from stores for back-compat consumption. Replace `hubId ?? '__wall-canvas-unbound__'` sentinel in WallCanvas with `hubId ?? null`; update `useCanvasViewportInput` to accept `ProcessHubId | null` with a null short-circuit (no store writes when hub is absent). 25-file sweep across packages/hooks/ui/apps: all `string`-typed hub slots at store boundaries cast/narrowed to `ProcessHubId`; test fixtures use `const h = (id: string) => id as ProcessHubId` helper. Both app lifecycle hooks (PWA + Azure) use `normalizeProcessHubId(hubId)` → `boundHubId` so the raw `string | null | undefined` param never reaches store calls. Closes LOW #19 from the 8f retrospective (docs/investigations.md updated). Co-Authored-By: ruflo <ruv@ruv.net> * fix(cleanup): move h() cast helper after imports in mapwall test Prettier placed the const between two import statements during the pre-commit format pass. Move it after all imports to keep ES module import grouping clean. Co-Authored-By: ruflo <ruv@ruv.net> * chore(stores): sweep ProcessHubId imports to @variscout/core Removes the back-compat re-export shim at canvasViewportStore.ts:16 per feedback_no_backcompat_clean_architecture; 7 consumers now import the branded type directly from @variscout/core/processHub. Barrel index.ts updated to re-export ProcessHubId from @variscout/core/processHub directly (not via the store file) so existing code that happens to pull from the barrel continues to compile. Co-Authored-By: ruflo <ruv@ruv.net> * chore(cleanup): finish ProcessHubId import sweep to @variscout/core Sweep the remaining ~15 consumers missed by the earlier pass; barrel shim removal exposed the gaps via TS error. All ProcessHubId imports now go directly to @variscout/core/processHub. Co-Authored-By: ruflo <ruv@ruv.net> * test(core, hooks): cover asProcessHubId/isProcessHubId + useCanvasViewportInput null short-circuit Two Important test gaps from PR2 code review: factory throw-on-empty (asProcessHubId, isProcessHubId) and the new hubId:null no-op path in useCanvasViewportInput. Both verify load-bearing contracts that were shipped without exercise. Co-Authored-By: ruflo <ruv@ruv.net> * refactor(canvas): extract useCanvasHypothesisDrawing hook Moves all drawing-mode pointer/keyboard event handlers and the reset effect out of Canvas/index.tsx into a dedicated hook. The hook returns stable handler refs plus endpointLabel and parseEndpointElement helpers — 12 vitest tests cover all branches. Co-Authored-By: ruflo <ruv@ruv.net> * refactor(canvas): extract useCanvasHypothesisArrows hook Moves arrowSegments state, cardElements ref, and both useLayoutEffects (ResizeObserver setup + DOM measurement) out of Canvas/index.tsx into a dedicated hook. Measurement is stable equality-guarded to avoid needless re-renders. 8 vitest tests cover empty, ResizeObserver attach/detach, and remeasure-on-viewport-change paths. Co-Authored-By: ruflo <ruv@ruv.net> * refactor(canvas): extract CanvasLevelRouter sub-component Moves the lens-validity gate and all l1/l2/l3 content composition out of Canvas/index.tsx into an internal sub-component. Canvas drops from 1135 to 848 lines. CanvasAuthoringMode and CanvasL3Archetype are defined in CanvasLevelRouter and re-exported from Canvas to avoid a circular import. 10 vitest tests cover the routing, lens-invalid empty state, archetype switching, and mode-toggle visibility. Barrel updates and CanvasWorkspace.test.tsx mock additions for the two new hooks land in this commit. Co-Authored-By: ruflo <ruv@ruv.net> * chore(types): add vite-env.d.ts + trivial tsc fixes in core/hooks (pre-existing) Category A — add minimal ImportMeta.env / ImportMeta.glob augmentation .d.ts files to packages/core/src/ and packages/hooks/src/ so that import.meta.env.DEV and import.meta.glob(...) resolve without requiring vite as a direct dep. Overloaded signatures distinguish eager (→ T) from lazy (→ () => Promise<T>) so calls into registerLocaleLoaders (which expects LocaleLoaderMap) type-check. Category C — two trivial fixes: - packages/hooks/src/__tests__/setup.ts: cast (window as unknown as Record...) to satisfy TS2352 (neither type sufficiently overlaps). - packages/hooks/src/__tests__/findingSourceLensCapture.test.ts: add _filters param to the vi.fn mock that takes one argument; was inferred as 0-arg. - packages/hooks/src/__tests__/setup.ts: /// <reference types="vitest/globals" /> so beforeEach is recognised by standalone tsc (globals:true is vitest-only). All pre-existing; unrelated to PR1/PR2/PR3 logic. Co-Authored-By: ruflo <ruv@ruv.net> * chore(test): cast DataRow fixtures + log deferred tsc errors (pre-existing) Category B — cast Record<string,unknown>[] test fixtures to DataRow[] at the constant definition site so useParetoChartData and useYamazumiChartData calls type-check without changing test behaviour: - packages/hooks/src/__tests__/timeLensWiring.test.ts: add DataRow import + cast PARETO_100 at buildParetoRows(100) assignment. - packages/hooks/src/__tests__/useYamazumiChartData.test.ts: add DataRow import + change EMPTY_DATA type annotation from Record<string,unknown>[] to DataRow[]. docs/investigations.md — new entry "Pre-existing tsc errors deferred from PR #168" listing the 3 remaining items that are out of scope for a trivial-cast pass: d3 type deps (useCanvasViewportInput.ts:2-4), tuple-mock fetch typing (useHubCommentStream.test.ts:274-277), and beforeEach globals in responsesApi.test.ts:862. All pre-existing; unrelated to PR1/PR2/PR3 logic. Co-Authored-By: ruflo <ruv@ruv.net> * chore(test): import beforeEach in responsesApi.test.ts One-line addendum to the pre-existing-tsc-error cleanup commit: the responsesApi.test.ts file uses beforeEach at line 862 but the vitest import on line 1 omits it. Same category as the other ImportMeta / test-globals fixes in afd8c8b. Co-Authored-By: ruflo <ruv@ruv.net> * docs(investigations): close item #3 + log item #4 (core fixture-shape mismatches) The beforeEach import fix in e73fca6 closes the responsesApi.test.ts item from the deferred-tsc-errors entry. Adds item #4 capturing the 9 newly-surfaced SustainmentRecord + ProcessMap fixture-shape mismatches that became visible once responsesApi.test.ts was unblocked. Co-Authored-By: ruflo <ruv@ruv.net> --------- Co-authored-by: ruflo <ruv@ruv.net>
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.
Summary
Test Plan