P2.1: Point test_presence.py ISO-week scan at core.presence (the real
model since task 510c1f78); the integration's presence.py is now a
re-export shim, so scanning it was vacuous.
P2.2: Add direct core-builder unit tests to test_events.py — exact
payload key sets for completed/uncompleted/day-complete events,
was_on_time threading (True/False/None), event ordering (completed
fires before day-complete on the bus), and the four day-complete
outcomes (zero-quest day, non-today instance, partial day,
fully-cleared today). Add a shim timezone test that sets
hass.config.time_zone to a zone whose local date differs from UTC and
asserts the day-complete rule uses that local date.
P3.1: Split build_quest_completed_events into
build_quest_completed_event (single event, no day-complete state) and
build_child_day_complete_event (separate, returns the day-complete
event or None). The shim fires nestquest_quest_completed BEFORE it
reads the HA-local today and evaluates the day-complete rule,
restoring the pre-extraction ordering so a failure in the day-complete
DB reads or the ZoneInfo lookup cannot suppress the completed event.
Payloads and the day-complete rule are identical.
P3.2: test_startup_db.py glob -> rglob so core/ is scanned by the
'never unlinks or truncates DB files' guardrail (non-recursive glob
skipped the extracted core package).
P3.3: RELEASE_NOTES.md notes the children and admin_allowlist loggers
moved to .core.children / .core.admin_allowlist (parent logger still
catches them).
P3.4: Refresh the stale test_core_isolation.py docstring to cover the
whole core package (storage-layer task d0b691d8 + engines/business-
layer task 510c1f78).
uv run pytest -q = 1507 passed (baseline 1497 + 10 new).
Move presence.py, materialize.py, completion.py, events.py, children.py,
quest_definitions.py, admin_allowlist.py and service_policy.py into
custom_components/nestquest/core/, leaving re-export shims at the
original paths so the integration keeps working unchanged.
core/events.py is refactored to PAYLOAD CONSTRUCTION ONLY: it builds
and returns (event_type, payload) tuples and never touches hass.bus or
hass.config. The HA-local 'today' the day-complete rule needs is now a
parameter (read from hass.config.time_zone by the integration shim).
The events.py shim retains fire_quest_completed/fire_quest_uncompleted,
wiring the returned events to hass.bus.async_fire so the four event
names and payload shapes stay identical (nestquest_quest_completed with
was_on_time, nestquest_quest_uncompleted, nestquest_child_day_complete
with quests_due/quests_completed, occurred_at strict UTC ISO-8601).
Update two test logger-name references (children now logs as
...core.children) and the materialize _now_stamp monkeypatch target
(now patches the core module the walk reads globals from).
1497 passed (baseline). No homeassistant import under core/, no
top-level core import under the integration, no repo-root core/.
The completed event's was_on_time travels on the CompletionResult
(decided with the write) instead of re-reading latest state a racing
reversal could replace; unload clears coordinator listeners before
the database close.
complete/uncomplete return a CompletionResult str subclass carrying
appended (decided with the write under the lock), so duplicate
concurrent calls fire one transition event; day-complete only
evaluates for an instance due today.
quest_completed and quest_uncompleted fire on actual transitions with
documented payloads; child_day_complete fires when a completion clears
the child's day; quest_missed is defined as Feature 11's contract.
Payloads documented in the README for automation authors.
overseer
orchestrator-agent