AiResponse extends with dueDate: string|null. zod regex
^\d{4}-\d{2}-\d{2}$, follow-up roundtrip check coerces invalid
dates (2026-13-99 etc.) to null. PROMPT_VERSION → 2: prompt now
takes todayKst arg, asks model to extract due_date as ISO or null.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Regex + KST math, returns ISO YYYY-MM-DD or null. 14 high-confidence
rules (literal date, N월 N일, MM/DD, N일/주/개월 뒤, 모레/내일/글피/오늘,
다음/이번 주 X요일, 다음 달). Ambiguous tokens (월말, 주말, 퇴근 전,
시각) return iso=null + confidence='medium' so caller (AiWorker)
can defer to AI. 26+ golden fixtures.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Three-state outcome per note: inserted (new id), skipped (id+rawText
match), forked (id match but rawText differs → new uuidv7 to preserve
raw_text invariant from slice §1.1). Media files copied into
MediaStore convention <profileDir>/media/{noteId}/{n}.{ext} with
new media DB rows.
NoteRepository.importNote handles full provenance: ai_status='done',
ai_provider, ai_generated_at, edited flags, intent fields, tags
with source preserved.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
ExportService composes pure exportFormat layer + reads from
NoteRepository.listAll (new, asc-ordered) + MediaStore.absolutePath
(new helper). Writes notes/{date-id8-slug.md}, media/{id8__n.ext},
index.jsonl, manifest.json, README.md to user-picked dir.
6 unit tests against tmp dirs + :memory: DB.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Skips when marker matches today's KST date. Marker written after
successful snapshot, before rotation. lastSnapshotAt() exposed for UI.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Code review I1: wrap snapshot's backup+rename in try/catch, unlink
orphan tmp file on failure so the next run does not encounter a
confusing 'existing file is not a database' error from sqlite.
Code review I2: JSDoc note that snapshot() is not safe for concurrent
calls — callers should serialize via runDaily()'s marker.
New unit test injects a fake db whose backup() rejects after a partial
write, asserts no .tmp / .sqlite remains.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Code reviewer minor nitpicks:
- Add test for inkling-2026-02-30.sqlite (locks roundtrip-validation contract)
- Add test for weekly window inclusive at oldest boundary
- Precompute today=startOfDayUtc(now) once outside the loop, pass to helpers
No behavior change; tests added cover existing semantics.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
- Vitest needs the node-ABI prebuilt sqlite binary; Electron 41 needs ABI 145.
Add `rebuild:node` / `rebuild:electron` scripts and wire them as
pre-hooks for `test`, `test:e2e`, `start`, `dev`, `test:integration`.
- Smoke test was racing on `firstWindow()`, which non-deterministically
returned the quickcapture window. Strip ELECTRON_RUN_AS_NODE from the
launch env, wait for both windows to register, then pick the inbox by
title and assert the heading via `getByRole`.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Task 32 of the slice plan. playwright.config.ts (testDir
tests/e2e, single worker, 60s default timeout, list reporter)
and a smoke spec that launches the built main bundle, waits
for the inbox renderer, and asserts the v0.2 empty-state copy
'첫 기억을 구출해보세요.' is rendered.
Build verified end-to-end (`npm run build` exits 0, produces
out/main/index.js + out/preload/index.mjs + the inbox /
quickcapture renderer chunks). Playwright's chromium bundle
installed.
Known issue (deferred to a follow-up debug session): on this
specific Windows machine, electron@41.3.0's main-process
module hook is not injecting the real `electron` API into the
launched bundle. `require('electron')` from the entry script
returns the on-disk path string (the package's index.js
default), and `process.electronBinding` / `process.type` are
undefined inside the spawned process. This reproduces with a
minimal package.json + main.js outside this repo — i.e. it's
not a slice bug. The smoke test currently fails with
'Process failed to launch!' at the
`app.whenReady().then(...)` line because `app` is undefined.
When the host issue is sorted, no test changes should be
required: build, config, and spec are correct. Likely fixes:
clean-reinstall electron + node-modules, try electron@latest,
or invoke through `npx electron` rather than the binary
directly. Captured in project_inkling_status memory under
'open issues'.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Task 20 of the slice plan. Thin wrapper over the repo's intent
methods that adds two preconditions: setIntent rejects empty
trimmed text, both methods throw "note not found" when the
note id doesn't exist. Repo-level COALESCE on
intent_prompted_at preserves the first-prompt invariant
(spec §3.3); IntentService's job is just input guarding so
the IPC handler stays a one-liner.
Verification: `npx vitest run tests/unit/IntentService.test.ts`
4 passed.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Task 15 of the slice plan. Strategy §4.1 immediate-reward
toast. celebrate(noteId) deterministically picks one of the
4 reward copies via SHA-256(noteId)[0] % 4, then forwards to
the injected send() callback (which Task 30 wires to a real
electron Notification). Skips silently when isSupported() is
false (denied OS permission), and swallows send() errors so
that capture path never fails because of a notification quirk.
Verification: `npx vitest run tests/unit/NotificationService.test.ts`
3 passed.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Task 14 of the slice plan. Orchestrates a Quick Capture submit:
trims/validates input, creates the note row + pending_jobs row
in one repo.create transaction, persists each pasted PNG via
MediaStore (relPath stored in media table), then awaits the
injected enqueue() (AiWorker.enqueue at runtime) and fires
celebrate() (NotificationService.celebrate). deleteNote drops
the db row (cascading note_tags / media / pending_jobs) and
removes the on-disk media directory afterwards.
Verification: `npx vitest run tests/unit/CaptureService.test.ts`
4 passed.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Task 13 of the slice plan. Drives the pending → done/failed
transitions:
- enqueue() pushes a Job and kicks the loop; loadFromDb()
rehydrates pending_jobs at startup so app restart resumes
in-flight work.
- drain() exposes a Promise for tests + graceful shutdown.
- Concurrency 1: a single async loop awaits each provider call
before the next, matching spec §2.2.
- 3-attempt backoff (default [0, 30s, 120s]; tests inject [0,0,0]).
Each failure logs ai.retry, increments pending_jobs.attempts,
and on the final attempt calls markAiFailed and emits onUpdate.
- emit() pushes the freshly-hydrated note to onUpdate (used by
Task 30 to fan out IPC note:updated events).
Verification: `npx vitest run tests/unit/AiWorker.test.ts`
4 passed. Suite total 41 / 41.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Task 12 of the slice plan. Implements the slice's only provider:
- generate(): POST {endpoint}/api/generate with Korean-first
prompt; AbortController-driven 120s timeout; parses
body.response as JSON and runs it through parseAiResponse.
- healthCheck(): GET /api/tags; returns ok when configured model
is in the listing, otherwise reports the missing-model reason.
- Constructor takes opts.endpoint / opts.model so Task 30 main
entry can inject INKLING_OLLAMA_ENDPOINT for LAN dogfood;
defaults are http://localhost:11434 and gemma4:e4b.
Tests: 6 unit cases via undici MockAgent (parse, non-JSON,
timeout abort, healthCheck ok / missing / connection error).
Integration test gated by INKLING_INTEGRATION=1 hits real
Ollama for Korean / English-stack / mixed input cases with a
180s per-test budget — also reads INKLING_OLLAMA_ENDPOINT so
LAN dogfood can be exercised end-to-end.
Plan deviation: undici@8 types treat MockAgent's `.reply()`
callback as sync-return-only, so the 500ms-delayed reply used
in the timeout test is cast `as never` to bypass the overload
mismatch. Behavior is correct at runtime; the cast is local to
the test.
Verification: `npx vitest run tests/unit/LocalOllamaProvider.test.ts`
6 passed.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Task 10 of the slice plan. parseAiResponse runs the model JSON
through a zod RawResponseSchema, then enforces slice rules:
- title MUST contain Korean characters; throw otherwise.
- summary normalized to exactly 3 lines (pad with empty when too
few; collapse trailing lines into line 3 when too many).
- tags filtered to /^[a-z0-9]+(-[a-z0-9]+)*$/ kebab-case and
capped at 3.
buildPrompt assembles the user-facing prompt with explicit
Korean-output / kebab-tag / no-fence rules (PROMPT_VERSION=1).
Verification: `npx vitest run tests/unit/ai-schema.test.ts`
7 passed.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Task 9 of the slice plan. Computes the data behind the Inbox
ContinuityBadge and recovery toast (Strategy §5):
- KST Mon-Sun grouping via toKstDateKey/kstMondayOf helpers
(UTC + 9h shift, then bucket to local Monday).
- weekCount + weekTarget=7 + consecutiveCompleteWeeks (walk
backward from current week, or previous if current incomplete,
counting only weeks with >=7 notes).
- showRecoveryToast=true when the latest note lands on today
(KST) and the prior note is at least 7 days earlier — this is
the "흐름을 다시 이어갑니다" trigger.
- now() injected for deterministic tests.
Verification: `npx vitest run tests/unit/ContinuityService.test.ts`
6 passed. All 24 unit tests across migrations / NoteRepository /
MediaStore / ContinuityService green.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Task 8 of the slice plan. Saves clipboard PNG/JPEG bytes under
{profileDir}/media/{noteId}/{uuid}.{ext} with mkdir -p semantics,
returns relPath/mime/bytes for the repository row, supports
deleteNoteDirectory (used by note delete + GC) and listNoteDirs
(used by media GC to find orphans).
Verification: `npx vitest run tests/unit/MediaStore.test.ts`
4 passed.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Task 7 of the slice plan. Implements the full repository surface
backing every IPC inbox/capture path: create (UUID v7 + atomic
notes + pending_jobs insert), insertMedia, findById/list,
updateAiResult (CASE WHEN guard against title/summary
overwrite when *_edited_by_user flips), markAiFailed (truncates
ai_error to 500 chars + clears pending job), updateUserAiFields
(sets edited flags as a side effect, replaces user-source tags),
setIntent + dismissIntent (intent_prompted_at uses COALESCE so
the first stamp wins), delete, getPendingCount,
getAllPendingJobs, incrementJobAttempt, and a private hydrate
that joins notes with note_tags + media.
Plan deviation: list/list-with-cursor query gets a secondary
"id DESC" tiebreaker. Two notes created in the same millisecond
shared created_at and reordered nondeterministically; UUID v7
sorts monotonically with creation order, so id DESC restores
"newest first" within ties.
Verification: `npx vitest run tests/unit/NoteRepository.test.ts`
12 passed.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
