Commit Graph
8 Commits
Author SHA1 Message Date
Marc-André Bombeck 07f1b6ce92 refactor(modules): one record-keyed write for a record's module state
The guardian's modules route and the setup flow's confirm both decide a
record's modules, and each had its own copy of what that means: write the
preference blob, write the delegated cycle column, invalidate the health
score. Two copies of a three-step write drift, and the second one was added by
this branch. They share one function now, keyed on the record, which is the
carrier the managed-profile work needs.

Authentication stays at the routes and stays different on purpose — the
guardian route acts on a managed record, the setup routes refuse outright
under a switch until a guardian's route into a record's setup lands with the
rest of that record's configuration. So does the merge: what the answers imply
and what somebody just ticked are different questions about what may be
overwritten. The actor's own modules route keeps its own write, because that
one is guarded on a concurrency token the record-keyed callers do not carry.
2026-09-10 11:54:19 +02:00
Marc-André Bombeck fb1cb4e3ed feat: gate a switched browser on the record's modules, not the actor's
Turning a module off for a profile you look after wrote a column nothing
on screen was reading. GET /api/auth/me published the ACTOR's module map
while switched, and the navigation, the dashboard tiles and the insights
sections all gate on it — so a guardian who turned Cycle off for the child
kept seeing a Cycle destination inside the child's record, and a guardian
who tracks it saw one there whether or not the child did (#939).

The map and the resolved cycle flag now answer for the record the session
is inside. They are not display preferences: every surface they gate shows
the record's data, so the record is what decides. Locale, theme and layout
stay the actor's. The record is taken from the grant-validated active
entry rather than the session's own selector, so a revoked switch cannot
publish somebody's configuration, and with no switch the payload is what
it was.

Cycle also gets a row on the guardian's Modules surface, which it never
had. It is a delegated module, so its state is the record's own cycle
profile and modulePreferences.cycle is refused — which left a guardian no
way to turn it off for a record whose sex says it should be on. The family
names the real column instead of a blob entry the gate ignores.

Refs #939
2026-09-10 00:02:03 +02:00
Marc-André Bombeck a45dde2039 Retire the guardian toggle over a pref nothing reads
The evidence disclosure stopped honouring showEvidenceByDefault in
v1.4.27 and the owner surface retired the control with it — but the
managed-record coach form still offered the switch, silently governing
nothing. Remove the control, the managed-schema pick, the response
field, and the label in all six locales; the strict schema now refuses
the key outright.
2026-08-14 01:56:25 +02:00
Marc-André Bombeck 6a1d1dd6e8 fix: preserve managed target defaults
- Return effective threshold ranges after managed updates\n- Localize closed Coach and Insights display inventories
2026-08-06 03:52:49 +02:00
Marc-André Bombeck 247ac8494c fix: complete managed record settings
- Expose canonical defaults and reject invalid settings families\n- Localize Guardian controls and refresh dependent caches
2026-08-06 03:40:13 +02:00
Marc-André Bombeck e788675b82 feat: configure managed record settings
- Scope writable settings to the selected managed record\n- Render typed Guardian controls with separated audit attribution
2026-08-06 03:21:02 +02:00
Marc-André Bombeck 03ee01ba56 feat: show managed integration status
- Render a record-keyed status view for managed profiles\n- Refuse shared Settings links before actor controls mount\n- Localize intentional unavailable states
2026-08-06 02:38:54 +02:00
Marc-André Bombeck 009de7cb5f feat: add record keyed settings
- Separate managed record settings from actor identity\n- Limit integration reads to the active managed record
2026-08-06 02:26:44 +02:00