The client bundle carries only the active locale — resolved on the
server and handed to the provider with no hydration flash, halving the
dashboard route to under a megabyte parsed (288 KiB gzip). The auth
shell renders the app chrome immediately instead of a spinner while the
session resolves in parallel, the widgets request folds into the
snapshot it duplicated, and insight pills prefetch their sub-pages —
a first visit to a sub-page lands in about ninety milliseconds instead
of nearly a second.
The route-level loading state now renders the same structured tile and
chart silhouettes the page uses, so the first paint no longer flashes a
grid of bare cards before the skeletons arrive. The chart reveal cap
drops to 1.2 seconds, the QR library loads only when an invite is
minted, and list filtering moves onto a shared pill-rail filter bar —
active filters read as removable chips, date ranges live in a popover,
and the measurements and mood lists speak it first.
Bundles the deferred v1.4.47 Mediums + Lows with the v1.4.48-deferred items
into a single release. Headline: server-side suppression of MEDICATION_REMINDER
APNs for iOS clients that own their local reminders (closes GH #206), a
push_attempts table backing /api/admin/notifications/diagnostic, and a sweep
of v1.4.48 QA forward-findings.
Performance (the user-visible win):
- /api/analytics cold-fallback path splits the heavy single-statement
aggregate into two parallel queries. The linearly composable columns
(count / min / max / mean) keep the full-partition scan; the windowed
7 / 30 / 90-day avg / slope / r² columns take a 90-day outer cap so
the planner does an index range scan on (user_id, type, measured_at)
instead of a full-partition sequential scan. Output is bit-identical;
cold slim slice drops from multi-second to sub-second on tenants with
large measurement partitions.
Server-side notifications:
- GET + PATCH /api/auth/me/notification-prefs with deep-merge semantics.
Initial category medication.clientManaged: boolean (default false). Cron
medicationReminderJob skips dose-due APNs when the flag is true,
emitting a medication_reminder.suppressed_client_managed wide-event
annotation per skip.
- push_attempts table backs the diagnostic endpoint. Every APNS /
Web-Push / Telegram / NTFY sender writes a fire-and-forget row.
Daily 03:35 Europe/Berlin cleanup prunes rows older than 90 days.
- /api/admin/notifications/diagnostic — admin-only endpoint surfacing
device APNs token presence (masked), channel config presence, and
recent push attempts. Registered in OpenAPI.
- Per-channel APNs "Send test" button now fires with
eventType: "MEDICATION_REMINDER" so the dispatcher's time-sensitive +
priority-10 branch runs. A successful self-test now exercises the
exact path production reminders take; an unsuccessful one points the
user at iOS Settings → Time Sensitive Notifications rather than
appearing to be a server-side bug.
Withings reliability:
- OAuth callback consumes the nonce atomically (delete-first with
P2025 catch for the replay branch). The previous read-then-delete
pattern allowed a race between two concurrent callbacks.
- ?reason=state redirect differentiated into csrf1 / replay / expired /
cross_user so operators can tell mismatch types apart without DB
shell access.
- connect endpoint rate-limited to 10/60s/user; row-create failures
redirect to a consistent error surface.
Observability hardening:
- redactSensitiveFields(body) denylist helper covers password / token /
secret / apiKey / authorization / csrfState / nonce, recursive,
case-insensitive. Caught: /api/devices invalid_format issues were
echoing apnsToken values verbatim into the audit log.
- sanitiseZodIssues(issues, { stripValuesFromMessage: true }) overload
applied to 14 audit-log sites.
Migrations: 0079 (notification_prefs), 0080 (push_attempts).
Tests: 4815 → 5272 unit + integration. Workout-batch integration suite
that had been red on main since v1.4.45 (three release commits shipped
through it) now green again — updated to match v1.4.42 cross-source
dedup semantics.
Resolves PR #207.
Maintainer-voice sweep across `src/`. ~190 references to the
maintainer by first name in code comments and JSDoc neutralised to
"the maintainer", "the live tenant", "users", or "the user" depending
on context. Three sed passes covering ~50 distinct patterns
(`Marc reported`, `Marc wants`, `Marc kept`, `Marc rolled back`, etc.).
Test fixtures (`displayName: "Marc B."`, `username: "marc"`,
`<HeroStrip userName="Marc">`) kept as opaque test data per the brief.
CHANGELOG normalisation. The v1.4.14 + v1.4.15 entries paired a
German bullet with its English translation by historical design — the
in-app copy is English-default and the user docs are English, so the
mixed-language style was a holdover from the early-marathon notes.
A one-off node script extracted the `_..._` English paragraph from
each bilingual bullet and rewrote the entry as that text. ~44 bullets
reduced; one Docs bullet that lost a leading word in the regex
fixed by hand.
CLAUDE.md → CONTRIBUTING-AI.md rename so the filename is no longer
AI-vendor-specific. `CONTRIBUTING.md` reference updated. `AGENTS.md`
stays in place for multi-agent compatibility (the conventions doc is
shared with Codex / Cursor / etc.).
Wave-4 state wrap. STATE.md now records C1–C5 + D + D2 as complete;
v1421-backlog.md grew a "Closed in v1.4.22" header listing the 20
items the marathon landed; `.planning/phase-W4-report.md` documents
the per-commit decisions, test deltas, and the 23 items deferred to
v1.4.23 with reasoning.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Marc reported a stray "mini scroll-button" on the right side of the
admin feedback inbox tab strip (Open / Acknowledged / Resolved /
Archived). The four short tabs do not horizontally overflow the
viewport, so the painted bar surprised him every time he opened the
inbox.
Root cause is a CSS overflow coupling: setting only `overflow-x: auto`
silently forces `overflow-y` to `auto` too, because a single axis
cannot be `visible` while the other clips. The strip is locked at
`h-9` (36 CSS px) but its badge children (`py-0.5` + text-xs digits)
ride 1-2 px taller than the strip on some glyph stacks, so the
implicit `overflow-y: auto` paints a tiny vertical scrollbar on the
right edge.
Fix is a single class on `tabsListVariants`: `overflow-y-hidden`
(plus `overflow-y-visible` in the vertical-orientation override so
vertical tab columns are not affected). Horizontal swipe and
keyboard-arrow scrolling still work — only the implicit y-axis bar
is suppressed.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
* docs(ui): add UI guidelines as single source of truth + shadcn MCP server
- `docs/ui-guidelines.md` lays out Dracula tokens, spacing, typography,
component picks (button/input/dialog/sheet decision tree), layout
rules (settings split target, dashboard one-row), accessibility
baseline (WCAG 2.1 AA + axe-core in CI), the autofill / honeypot
pattern for measurement forms, and a reusable "Test button" UX for
every integration. New components in v1.4 reference this doc.
- `.mcp.json` wires the official `shadcn` MCP server so the design
system can be browsed and components installed without leaving the
editor.
- Adds `@vitest/coverage-v8` to dev dependencies — `pnpm test
--coverage` now reports real per-file coverage (the v1.4 cycle uses
this to lift critical-path coverage).
Co-Authored-By: Marc-André Bombeck <noreply@anthropic.com>
* fix(settings/ai): connection-test honours the unsaved provider selection
Changing the AI provider in /settings, then clicking "Verbindung testen"
without saving first, used to silently run the test against the *saved*
provider instead — so the result felt unrelated to what the user had on
screen. This often surfaced as a confusing OK / failure depending on
whichever provider had previously been persisted.
The test endpoint now accepts an optional JSON body that mirrors the
form fields (provider, model, baseUrl, anthropicKey, localKey). When
present, the server resolves a transient provider from those values —
falling back to the persisted, encrypted credentials per field, so a
user with a stored Anthropic key can swap models in the dropdown
without re-typing the key. Plaintext keys never persist; the existing
SSRF guard for the Local provider's base URL is reused, the existing
rate limit (5/min) and the V3 error-leak shielding stay in place.
Empty body keeps the previous behaviour for any other caller.
Tests cover the new override path (provider forwarded, unknown
provider rejected with 422, empty body uses persisted config, config
errors surface the resolver's status). Whitespace in
`src/app/settings/page.tsx` is reformatted by Prettier as a
side-effect; logic touches only the AI section.
Co-Authored-By: Marc-André Bombeck <noreply@anthropic.com>
* feat(ui): add Skeleton and EmptyState primitives
Two pieces every page in v1.4 needs and the codebase didn't have:
- `<Skeleton>` — a presentation `<div>` with the standard pulse
animation, `aria-hidden`, and `motion-reduce:animate-none`. Used to
shape the page while data loads instead of dropping a spinner. The
visible layout no longer jumps when the first measurement arrives.
- `<EmptyState>` — the shape every list, tile, and chart panel
reaches for when it has no data: optional icon, title, description,
primary action; `card` (default) and `plain` variants;
`default`/`compact` density. Lives inside an `aria-live="polite"`
region so screen readers announce the message and its action.
Tests cover screen-reader semantics, prettier-merged class names,
density/variant switches.
These two land first because the v1.4 pages — the split settings
sections, the new dashboard hero, the redesigned admin status grid —
all assume the primitives exist. The patterns are documented in
`docs/ui-guidelines.md` §4.1–4.2.
Co-Authored-By: Marc-André Bombeck <noreply@anthropic.com>
* fix(ui): block password-manager autofill on health-data inputs by default
Browsers and password managers were happily autofilling the user's
account credentials into measurement, medication, mood, and AI-token
fields whenever those forms didn't explicitly set `autoComplete`. The
codebase had zero `autoComplete` attributes across the data-entry
surfaces, so this happened often enough to be both a UX papercut
(account password landing in a "Notiz" field) and a real privacy issue
(secrets persisting through CSV/JSON export).
The base `<Input>` primitive now:
- Defaults `autoComplete` to `"off"` when the caller doesn't supply a
value.
- Adds `data-lpignore="true"` and `data-1p-ignore="true"` whenever
autoComplete resolves to `"off"`, so LastPass and 1Password skip
the field too.
- Keeps the password manager active for any form that already passes a
semantic `autoComplete` (`"username"`, `"email"`,
`"current-password"`, `"new-password"`) — the existing login,
register, and AI-key inputs continue to work unchanged.
Tests cover the three meaningful states (default off, explicit
semantic value, login-flow `username`).
Co-Authored-By: Marc-André Bombeck <noreply@anthropic.com>
---------
Co-authored-by: Marc-André Bombeck <noreply@anthropic.com>