ResolveDispatcherScope now admits the Scheduler role, which owns the
viewAllDispatchersOnDashboard permission, so it can load Stats, Workload,
Performance, Regions and Trend instead of being denied.
GetTrendAsync now resolves scope via the shared ResolveDispatcherScope so a
picker DispatcherId selection scopes Trend consistently with the other
endpoints and unauthorized roles fail closed.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Break collection initializer entries onto one line each in
WorkOrderBoardRegions and WorkOrderBoardSearchTests so the changed-file
formatting gate (dotnet format --verify-no-changes) passes.
DeleteUserWithCascadeAsync never removed UserPermissionOverrides rows, and
the FK on UserId is Restrict. Once an override was saved for a member, the
admin hard-delete (DeleteUserAsync -> DeleteUserWithCascadeAsync) failed the
foreign key inside the transaction and the controller surfaced it as a 400,
so the user was never deleted. Add an explicit ExecuteDelete on
UserPermissionOverrides before the user is removed, matching how UserRoles is
already cleared in the same method (no schema change).
SetOverrideAsync was check-then-insert on the composite key with no
DbUpdateException handling, so two concurrent PUTs for the same
(UserId, PermissionKey) let the loser violate PK_UserPermissionOverrides and
return 500. Catch the conflict, detach the pending insert, and converge by
updating the persisted row to the caller's requested state.
The added service test asserts DashboardStatsDTO, which already carried
ScheduledTomorrow, PendingUplifts and AvetaPending, so it does not guard
the regression that was actually shipped: the DashboardController Stats
anonymous response dropping those keys and leaving the tiles empty.
Add DashboardControllerTests.GetStats_SerialisesKpiCountsOnWireObject,
which drives the controller and asserts the three keys are present on the
serialised wire object, so re-dropping any of them fails a test.
Also reword the pending-uplifts comment in DashboardServiceTests: the
Pending-only-vs-ChangesRequested exclusion is the behaviour the code
ships today, an open product call, not a settled review decision.
Locations store State as either a two-letter postal code or a full state
name (the location list filter expands codes to both forms via
UsStateCodes.ExpandStorageValues, and dev fixtures carry State = "indiana").
DashboardRegions.Resolve only matched the two-letter key, so every location
stored as a full name fell into Unmapped/Other and the East/Central/West/
California bars undercounted.
Normalise the stored value through a new UsStateCodes.ToCode, which accepts a
code or a full name and returns the canonical postal code, before the zone
lookup. Extend the region test with full-name storage forms and add an
invariant asserting every US state and its full-name form resolves to a
canonical bucket.
GetKpiCountsAsync computes ScheduledTomorrow, PendingUplifts and
AvetaPending and DashboardStatsDTO carries them, but the Stats
endpoint's anonymous wire object never serialised them, so the tiles
had nothing to read. Add scheduledTomorrow, pendingUplifts and
avetaPending to the response.
Add a DashboardServiceTests case covering GetKpiCountsAsync end to end:
it asserts all three counts and pins the pendingUplifts semantics to
Status "Pending" only, excluding "ChangesRequested" (which is back with
the vendor, not awaiting an approval decision).
ScheduledDate is persisted as a midnight calendar value (board create writes
request.ScheduledDate.Value.Date; board patch writes the parsed .Date into a
datetime2 column with no offset). GetTrendAsync treated it as a UTC instant and
converted to America/New_York, so midnight became 19:00/20:00 the previous day
and every board-scheduled work order fell into the prior bucket: a job scheduled
today read as yesterday and overdue, and the Today bucket showed zero.
Bucket by DateOnly.FromDateTime(scheduled.Date) with no conversion, matching
DashboardMetrics and WorkOrderDerivedFields, so Trend and Stats agree on the
same rows. Rewrite the daily and yearly trend tests to assert calendar-date
classification (the removed conversion had encoded the shift into their
expectations) and add a regression test that a midnight-today work order stays
in the Today bucket and is not overdue.
