All-day: a single-day all-day event displayed across two days. iCal all-day
DTEND is EXCLUSIVE (1-day event = DTSTART:24/DTEND:25) and the server occurrence
carries that exclusive end, but Schedule-X treats all-day end as INCLUSIVE.
hydrateEvents now subtracts one day (clamped to >= start) so a 1-day event shows
on one day and an N-day event spans N days. Write path was already correct
(verified against stored VEVENTs). +regression test.
Color: a member's coral (#E8734A) was mistaken for the shared-family rose
(#F25C7A). Reorder COLOR_PALETTE so warm near-rose hues (amber, coral) are
assigned LAST; early members get cool, clearly-distinct colors (blue/green/teal).
Gate 2 A3 fail: a second member (amelia) got the same color as the first (luc),
both #E8734A. Color was assigned by COUNT(*) % palette; a deleted spike user
shifted the count so two live members landed on the same slot. Replace with
'first palette color not already in use by another user' (fall back to count
round-robin only once the palette is exhausted) — guarantees distinct, stable
colors for up to palette length members. +1 regression test (deletion frees a
slot → next member fills it, no collision).
The legend showed 'Member 972be1a3' because Authelia does not emit
name/preferred_username/email in the ID TOKEN (only at the userinfo endpoint),
and @hono/oidc-auth reads ID-token claims only. The real fix is an Authelia
claims_policy adding those claims to id_token for the familysync client.
App-side robustness so it self-heals once Authelia is fixed (no DB surgery):
- deriveDisplayName now returns null (not a synthetic 'Member <sub>') when no
real claim is present, so we never persist an ugly sub string; the UI degrades
to a generic 'Member'.
- upsertUser now tracks the IdP name authoritatively: a non-null displayName that
differs from the stored value updates the row (blank/stale 'Member …'/email →
real name on next login). A null value never overwrites a good stored name.
Two write-path cache bugs surfaced during Gate 2 live testing:
P1 (delete didn't work / ghost event): syncCalendar only UPSERTED events
present on Fastmail and never removed cache rows for events that disappeared.
A successful CalDAV delete left the row in calendar_events forever, so
GET /api/events kept returning it and the UI showed a ghost that 'wouldn't
delete' (even after refresh). Add a prune step: delete calendar_events rows for
this calendar whose uid is absent from the server response (scoped to cal.id so
it never touches another calendar or the other member's rows — BUG B). Empty
server result prunes the whole calendar's cache.
P2 (edit needed a manual refresh): the outbox worker marked a row 'done' BEFORE
triggerTargetedResync refreshed the cache. The PWA's SyncStateToast invalidates
['events'] the instant sync-status flips to 'done', so it refetched stale cache.
Re-sync first, then mark done — 'done' now guarantees the cache reflects the write.
Tests: +2 prune regressions (present-subset prune, empty-server prune-all).
The displayName claim-preference logic (name → preferred_username → email →
sub fallback) was duplicated verbatim in me.ts and events.ts resolveUserId.
Extract it to auth/user.ts as deriveDisplayName and use it in both call sites,
so the rule has one definition. Update the events.test.ts user.js mock to keep
the real helper (spread importActual) while stubbing only upsertUser.
The original toSQL() regression test hand-built the joined query inside the
test body and asserted the SQL contained a join — tautological: it never
exercised the handler, so removing .innerJoin from events.ts left it green.
Replace it with two tests that issue real PATCH/DELETE requests against the
mocked select-chain (from → innerJoin → where) and assert the handler returns
202 (not 503) AND invokes the innerJoin spy. Verified RED: removing the
edit+delete joins fails both tests; GREEN with the joins present.
BUG 3: GET / had no ownership predicate — it returned all users' events.
Second household member would see other member's private events.
- Resolve currentUserId at top of GET handler (same resolveUserId helper
as write endpoints); return 401 if unauthenticated.
- Add ownership predicate to WHERE: AND (calendars.userId = currentUserId
OR calendars.isShared = true). Combined with and() around the existing
date-window or() block. Mirrors the /writable-calendars idiom (D-03).
BUG 2: Both me.ts and events.ts resolveUserId were passing email (often
absent) as displayName to upsertUser, resulting in blank legend names.
Also, upsertUser returned existing rows unchanged even when displayName
was null and a better value was now available.
- me.ts: derive displayName via name → preferred_username → email →
"Member <sub-prefix>" fallback, checked defensively. Updated JSDoc.
- events.ts resolveUserId: same derivation so write-path upserts don't
re-blank a correctly-set displayName.
- user.ts: when existing row has null displayName and caller supplies one,
issue an UPDATE so already-existing blank rows are corrected on next login.
Authelia-side emission of name/preferred_username is an operator concern
(claim mappings / userinfo scope in authelia config) — out of scope here.
The code now reads whatever claims are present and falls back sensibly.
BUG 1: PATCH /:uid/edit and DELETE /:uid selected calendars.url/userId
from .from(calendarEvents) with no join, causing Drizzle to throw at
toSQL() time → 503. Added .innerJoin(calendars, ...) to both lookups,
mirroring the working GET / join idiom.
- Updated PATCH + DELETE beforeEach mocks to route through innerJoin→where
- Updated CR-01 PATCH test mock similarly
- Added regression: edit/delete lookups join calendars describe block with
toSQL() assertions using vi.importActual (real drizzle, no DB needed)
- All 21 tests pass; typecheck clean
BUG A — timed events written 4h off: EventForm sent a naive local wall-clock
string with no offset; the UTC API container parsed it via new Date() as UTC, so
09:00 America/Toronto serialized to DTSTART:...090000Z. Fix: new
apps/pwa/src/lib/eventDateTime.ts serializes timed events to an unambiguous UTC
instant in the browser (where the operator's zone is known); all-day stays a DATE
string. No backend change.
BUG B — created events attached to the wrong user's calendar + duplicate calendar
rows per poll: calendars had no unique key on url, and poller/sync matched
calendars by url alone — so under the shared single Fastmail account (D-16) one
member's collection resolved to the other member's row. Fix: composite
unique(user_id, url); scope poller lookup + sync select to (userId, url); hand
migration 0001 (dedup + add key), applied to the live DB.
Regression tests fail against the buggy url-only predicate. API 98/98, PWA 140/140,
tsc clean both packages.
Root-level PWA files (manifest.webmanifest, sw.js, registerSW.js, workbox-*.js,
icon-*.png, apple-touch-icon.png) were falling through to the index.html
catch-all and returning HTML — breaking the manifest (syntax error) and
preventing the service worker from ever registering. serveStatic('/*') serves
any existing file and calls next() for SPA routes, so index.html stays the
fallback. Registered after /health, /api/*, /callback so those still win.
- git rm apps/api/scripts/seed-credential.mjs (operator-only, run out-of-band;
the credential is already seeded in the running DB)
- gitignore .playwright-cli/, gate2-*.png, and the seed script path
- Register app.get('/api/login', redirect to '/') in protected-routes block
- Route placed after OIDC guard so unauthenticated nav triggers auth flow
- Add login.test.ts covering bypass and OIDC-passthrough redirect paths
- Dockerfile: build apps/pwa into the production image's ./public so the API
serves the PWA on a single port (:3000) for the Pangolin/newt tunnel
- docker-compose.yml: set NODE_ENV=production (mount OIDC unconditionally) and
constrain OIDC_SCOPES=openid profile email offline_access (Authelia rejected
the empty-default's full scopes_supported with invalid_scope)
- apps/api/scripts/seed-credential.mjs: operator tool to seed member_credentials
(encrypted Fastmail app password) out-of-band — fills the documented gap
- In update dispatch, SELECT etag FROM calendar_events WHERE uid = row.uid before PUT
- Use fresh etag as If-Match instead of stale enqueue-time row.etag when available
- Fall back to row.etag when calendarEvents has no matching row
- D-08 conflict detection intact: genuine external changes update calendarEvents.etag
differently from any pending row, so they still 412 correctly
- WR-02 fresh: update PUT must use calendarEvents.etag not stale enqueue-time etag
(fails RED: capturedEtag === 'old-etag', not 'new-etag')
- WR-02 fallback: when calendarEvents has no row, fall back to row.etag (passes in RED)
- Add mockWhereCalEvents to mock infrastructure to isolate calendarEvents selects
- Switch all beforeEach to vi.resetAllMocks() to prevent mockImplementationOnce bleed
- outboxWorker: remove empty-credential fallback; let loadClientForUser throw on error (CR-03)
- outboxWorker: fix backoff index from nextAttemptCount to row.attemptCount so first retry waits 15s not 60s (WR-01)
- events.ts: replace bare crypto.randomUUID() with import { randomUUID } from 'node:crypto' on all three handlers (WR-08)
- Add mockDecryptPassword to vi.hoisted() so tests can control loadClientForUser behavior
- Add vi.mock for broker/crypto.js to enable CR-03 scenario
- Introduce wireMockChain() helper that differentiates credential vs outbox db selects
- CR-03 RED: credential-load failure must leave row pending, not call createFastmailClient('')
- WR-01 RED: first transient retry must use BACKOFF_SECONDS[0]=15s not BACKOFF_SECONDS[1]=60s
- Update FAKE_CRED_ROW so loadClientForUser can return a real credential-shaped row
- vevent.test.ts: D-13 form-parsed contract block — timed and all-day cases
(all-day DTEND+1 fails: emits 20260610 not 20260611)
- outboxWorker.test.ts: worker integration — create/update must pass BEGIN:VCALENDAR
to CalDAV write functions (fails: raw JSON passes through today)
- worker: unparseable payload must mark row failed (fails: marks done today)
- Update makeRow default payload to form JSON shape the worker should parse
- Import upsertUser from auth/user.js
- resolveUserId now async: dev-bypass path unchanged; OIDC path calls getAuth
then upsertUser(iss, sub, email) to resolve DB user id
- All 5 handlers (create, edit, delete, sync-status, writable-calendars) updated
to await resolveUserId and 401 only when it returns null
- Remove all inline 'For now return 401' stubs and redundant getAuth calls
- grep confirms 0 'For now return 401' stubs remain; upsertUser imported+called
- POST /create with valid OIDC session (devBypassInjectUser.active=false, getAuth
returns valid iss/sub) must return 202 not 401
- POST /create with no session (getAuth=null) must return 401
- Refactor getAuth/devBypass mocks to use vi.hoisted configurable flags for
per-test OIDC path isolation
- Mock upsertUser from auth/user.js so OIDC resolution can be verified
- Replace summary→title, dtstart→start, dtend→end in eventFieldsSchema
- Server now accepts exact CreateEventPayload shape the PWA sends
- Update existing write tests to use new canonical field names
- No internal rename map; one canonical name set end-to-end
- grep confirms no summary/dtstart/dtend in eventFieldsSchema
- POST /create with {title,start,end,allDay,recurrence} asserts 202 (fails: server requires summary/dtstart/dtend)
- PATCH /:uid/edit with same shape asserts 202 (fails: same schema mismatch CR-01)
- POST /create: validates with zod, checks calendar ownership (D-03/T-03-06), enqueues pending outbox row, returns 202 with uid
- PATCH /:uid/edit: looks up event, checks ownership, enqueues update row; uses db.transaction for edit-as-move calendar pair (D-04)
- DELETE /:uid: looks up event, checks ownership, enqueues delete row with server-side etag (T-03-10)
- GET /sync-status: returns outbox status scoped to currentUser only (T-03-07/D-09)
- GET /writable-calendars: returns own personal + shared calendars, never other member's personal (D-03/T-03-11)
- Auth via dev-bypass (c.get('user')) + getAuth(c) fallback; 401 if neither
- No tsdav import — broker boundary enforced (D-12)
- All 69 events tests GREEN; tsc --noEmit clean
- Add write endpoint tests: POST /create, PATCH /:uid/edit, DELETE /:uid
- Add GET /sync-status tests (D-09 outbox polling)
- Add GET /writable-calendars tests (D-03 writable set, access control)
- Wire db.insert and db.transaction into the vi.mock for db/client.js
- Mock devAuthBypass to inject dev user in write-endpoint tests
- All 9 new tests are RED (routes not yet registered)
ICAL.Time.fromJSDate(window, false) interpreted the UTC-midnight window bounds in the
server's local TZ (America/New_York in dev), shifting the window by the server offset and
dropping evening occurrences near a day window's end (e.g. June 11 17:45-04:00 = 21:45Z was
excluded from the June-11 day view). Use UTC so the window is deterministic and correct.
Backend:
- expand.ts: add ownerName: string | null to CalendarOccurrence
interface and expandOccurrences() signature; thread it onto every
emitted occurrence.
- events.ts: SELECT users.displayName as ownerName in the join; pass
it to expandOccurrences().
Frontend:
- client.ts: add ownerName: string | null to CalendarOccurrence.
- EventDetailPopover.tsx: render isShared ? 'Family' :
(ownerName ?? calendarName) in the footer instead of calendarName.
Tests:
- expand.test.ts: pass ownerName to all expandOccurrences() calls;
assert ownerName is carried onto occurrences in the DST test.
- events.test.ts: add ownerName to mock rows; assert ownerName present
on occurrences; add ownerName assertion to timed-recurring test.
- EventDetailPopover.test.tsx: add ownerName to fixtures; split
"calendar name in footer" into three targeted tests covering
personal-with-owner, shared→Family, and null-owner fallback.
- Old filter: hasRrule=1 AND dtstartUtc < windowEnd
All-day recurring masters have dtstartUtc=NULL so the comparison evaluates
to NULL/false — 11 such rows in live cache were never returned
- New filter: hasRrule=1 AND (dtstartUtc < windowEnd OR dtstartDate < end)
The OR covers all-day masters whose only date column is dtstartDate (DATE)
- expandOccurrences already does precise per-occurrence window checks, so
over-selecting a master on the DATE path is safe
- Extend events.test.ts: assert timed recurring master (dtstart 2024) returns
occurrences in 2026 window; assert all-day recurring master (dtstartDate 2024,
dtstartUtc NULL) returns its 2026-06-15 occurrence
- Use ICAL.Event.isRecurring() (parity with expand.ts) to detect RRULE/RDATE
- Add hasRrule to .values() INSERT and .onDuplicateKeyUpdate() SET so the flag
is set on first sync and self-heals on every subsequent re-sync
- Without this fix every event had has_rrule=0 (column default), causing the
events route recurring-master pre-filter to return zero recurring occurrences
- Add sync.test.ts cases: hasRrule=true for timed+all-day recurring VEVENTs,
hasRrule=false for non-recurring, and hasRrule in onDuplicateKeyUpdate.set
- Replace dtend ?? dtstart with event.endDate which handles DURATION-only VEVENTs
- Add positive-duration guard (PT30M / P1D) to both non-recurring and recurring branches
- Add single-duration.ics fixture and regression test asserting end > start for DURATION-only events
Schedule-X rejects ids containing ':' '[' ']' (the old ${uid}::${iso} form) — mint ev-<uid>-<epochMs> instead. Add an ErrorBoundary so a render throw shows the error instead of a blank page.
- Assert timed start/end strings include '[America/New_York]' bracket (not offset-only)
- Assert DST boundary offsets: -05:00[America/New_York] pre-transition, -04:00[America/New_York] post
- Add cross-contract regression test: feeds expandOccurrences output directly into
Temporal.ZonedDateTime.from() to prove the expand→hydrate contract holds end-to-end
- Import 'temporal-polyfill/global' at top of test file for the Temporal global
- Rename describe block from 'RED stubs (Wave 0)' to reflect GREEN state
Temporal.ZonedDateTime.from() rejects offset-only ISO strings such as
'2026-06-18T08:00:00-04:00'; it requires an IANA bracket, e.g.
'2026-06-18T08:00:00-04:00[America/New_York]'. serializeTime() was
emitting offset-only for named zones and bare 'Z' for UTC — both
unparseable by the frontend, blanking the calendar view.
Changes:
- Named IANA zone: emit '...±HH:MM[tzid]' using t.zone.tzid
- UTC zone: strip trailing 'Z' from toString(), emit '+00:00[UTC]'
- Floating zone (no registered VTIMEZONE): fall back to '+00:00[UTC]'
- Update CalendarOccurrence docstrings to reflect the IANA-annotated contract
- Add temporal-polyfill@0.3.2 as dev dep in api for cross-contract test
- Asserts GET /api/me returns 200 with DEV_USER (id=1, color=#4A90D9)
when DEV_AUTH_BYPASS=true and NODE_ENV!=production
- Asserts oidcAuthMiddleware is NOT wired when bypass is active
- Asserts oidcAuthMiddleware IS wired when bypass is absent
- Asserts 401 from getAuth(null) fallback path with no OIDC session
- index.ts: compute devBypassActive at startup; skip app.use(oidcAuthMiddleware)
entirely when active so the OIDC guard never runs in local dev
- routes/me.ts: read c.get('user') first; return dev identity directly when
devAuthBypass injected it, bypassing getAuth() and the DB upsert
- auth/devBypass.ts: add ContextVariableMap augmentation for 'user' key;
correct stale comment that claimed getAuth/401 path was still active
- zValidator enforces YYYY-MM-DD regex on start/end (T-02b-01)
- 90-day window cap prevents DoS (T-02b-02)
- innerJoin calendarEvents→calendars→users for color + isShared + ownerUserId
- SQL pre-filter includes hasRrule=true rows regardless of dtstartUtc range
- expandOccurrences() called per row; shared calendar uses #F25C7A rose color
- events.test.ts: added @hono/oidc-auth mock; 4/4 assertions green
Lucas Berger