Files
F-r-ldraledigplannerare/backend/tests/unit/test_db_unit_of_work.py
T
jacobh2 782415122d Separate user intent from planner output via day assertions
The calendar mixed two kinds of truth in one table. `planned_leave_day` was
simultaneously the allocator's output cache, the user's override store
(`is_pinned`, plus ad-hoc rows with no `wish_id`), and the resource ledger that
analytics and the age-4 maths read. The planner therefore had to ask "am I
allowed to touch this row?" on every sync, and got it wrong at the edges.

Every user assertion is now one `day_assertion` row — a concrete `day_type`
replaces a pinned day, a row with no `wish_id` replaces an ad-hoc day, and
`day_type='none'` replaces `planned_day_exclusion`. `planned_leave_day` becomes
purely derived with a single writer, and `PlanningService.pinned_keys()` is the
single answer to "is this day pinned", used by the allocator's loader, by
`sync_plan` and by the read router alike.

On top of that, every calendar band is now a real event: an ad-hoc band opens
the same full editor as a wish-backed one and saving it promotes it to a
`LeaveWish`, so it can be named, repeated yearly, assigned to either or both
parents, and handed to the planner. "Planering" is a `moveable` toggle plus a
from/to month bound plus a separate lock/unlock action on the days themselves,
replacing a three-rung ladder whose "planner decides" rung set a search window
equal to the drawn dates and so could never move anything.

Bugs found and fixed along the way:

- The plan erased its own history. Fixed windows clip to today, so an elapsed
  auto-placed day fell into `stale_keys` and was deleted on the next sync —
  and a background sync runs on every workspace load. Past rows are now frozen
  and counted as spent.
- Migration 010's downgrade silently destroyed every pinned day and exclusion
  (`is_pinned` returned as DEFAULT 0, the exclusion table came back empty,
  exit 0, no warning). It now reconstructs both from `day_assertion`, and 009
  itemises what it cannot recover before dropping the table.
- Deleting a period left its assertions behind, locking those dates forever
  with no UI path to clear them. Wish deletion now clears them by `wish_id`.
- `DayAssertionService` and `LeaveWishService` committed multi-statement writes
  without a transaction; a crash mid-write diverged the tables or half-deleted
  a series. Both now run in a unit of work.
- A sibling year's shortfall was never surfaced: the toast only inspected the
  edited occurrence's outcome, so a fully unfunded later year rendered as an
  empty band with no warning.
- A flexible window whose bound had entirely elapsed was dropped from the
  proposal with no outcome at all. It now reports as `unplaced`.
- Unlocking a period deleted its days instead of returning them to the planner.
- The e2e global setup posted to a route deleted in this change, which would
  have failed the whole suite before its first test.

Gates: backend 938 tests / 91% coverage, frontend 1351 / 97.4 | 92.2 | 94.37 |
97.4. Migrations 009 and 010 audited against a hand-built realistic database:
the upgrade reconciles exactly and a constraint violation fails safe, leaving
the schema at 008.

Not complete. Step 4's toggle, month picker and lock/unlock action are verified
by browser walk and unit tests, but the partial-lock readout was not exercised.
`ideas.md` §7 adds a full plan for the vacation-forfeiture defect (the planner
detects expiring days and then does nothing about them) — not implemented here.
Remaining smaller items are listed in the session's restructure notes.

Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01T4wjgbVL4HAC5Rdaw75VbV
2026-09-21 07:20:14 +02:00

242 lines
9.4 KiB
Python

"""Tests for the unit-of-work mechanism (`app.db.connection`) and the
batched `PlannedLeaveRepository` methods it exists to support.
See `scratchpad/audit/backend-perf.md` §3: the app runs on one shared
connection, so a multi-statement write needs (a) mutual exclusion for the
whole unit of work and (b) the per-row `commit()` calls suppressed while one
is in progress. These tests exercise that mechanism directly against an
in-memory connection, plus the `PlanningService.apply_proposed_days`
atomicity it is meant to guarantee end-to-end.
"""
import asyncio
from datetime import date, datetime
from uuid import uuid4
import aiosqlite
import pytest
from app.db.connection import commit, ensure_schema, unit_of_work
from app.entities.leave_wish import LeaveWish
from app.entities.planned_leave import PlannedLeaveDay
from app.repositories_impl.sqlite.leave_wish_repository import SQLiteLeaveWishRepository
from app.repositories_impl.sqlite.planned_leave_repository import SQLitePlannedLeaveRepository
from app.services.planning_service import PlanningService, ProposedDayInput
from tests.conftest import SeededIds
@pytest.fixture
async def uow_connection() -> aiosqlite.Connection:
conn = await aiosqlite.connect(":memory:")
conn.row_factory = aiosqlite.Row
await conn.execute("PRAGMA foreign_keys = ON;")
await ensure_schema(conn)
try:
yield conn
finally:
await conn.close()
async def _workspace_count(conn: aiosqlite.Connection) -> int:
cursor = await conn.execute("SELECT COUNT(*) FROM workspace")
row = await cursor.fetchone()
return int(row[0])
async def _insert_workspace(conn: aiosqlite.Connection, workspace_id: str) -> None:
await conn.execute(
"INSERT INTO workspace (id, name, created_at) VALUES (?, ?, ?)",
(workspace_id, "Test", datetime.now().isoformat()),
)
class TestUnitOfWork:
async def test_commits_on_success(self, uow_connection: aiosqlite.Connection) -> None:
async with unit_of_work(uow_connection):
await _insert_workspace(uow_connection, "ws-commit")
assert await _workspace_count(uow_connection) == 1
async def test_rolls_back_on_exception(self, uow_connection: aiosqlite.Connection) -> None:
with pytest.raises(ValueError):
async with unit_of_work(uow_connection):
await _insert_workspace(uow_connection, "ws-rollback")
raise ValueError("simulated failure")
assert await _workspace_count(uow_connection) == 0
async def test_commit_helper_is_noop_inside_uow(
self, uow_connection: aiosqlite.Connection
) -> None:
"""`commit()` must not durably commit inside a unit of work — otherwise
a later rollback in the same block could only undo writes made after
that no-op `commit()` call, not before it (see backend-perf audit §3,
"Is BEGIN/COMMIT ... safe on the shared connection? No, not alone.").
"""
with pytest.raises(RuntimeError):
async with unit_of_work(uow_connection):
await _insert_workspace(uow_connection, "ws-noop-commit")
await commit(uow_connection) # must be a no-op here
raise RuntimeError("still rolls back everything, including the row above")
assert await _workspace_count(uow_connection) == 0
async def test_concurrent_calls_serialize(self, uow_connection: aiosqlite.Connection) -> None:
"""Two concurrent `unit_of_work()` blocks on the same connection never
interleave — the second blocks at entry until the first fully exits.
"""
events: list[str] = []
async def worker(name: str, hold: float) -> None:
async with unit_of_work(uow_connection):
events.append(f"{name}-start")
await asyncio.sleep(hold)
events.append(f"{name}-end")
await asyncio.gather(worker("a", 0.05), worker("b", 0))
assert events in (
["a-start", "a-end", "b-start", "b-end"],
["b-start", "b-end", "a-start", "a-end"],
)
class TestPlannedLeaveRepositoryBatchMethods:
async def test_create_many_then_delete_many(
self,
planned_leave_repo: SQLitePlannedLeaveRepository,
seeded_ids: SeededIds,
) -> None:
now = datetime.now()
entities = [
PlannedLeaveDay(
id=uuid4(),
workspace_id=seeded_ids.workspace_id,
child_id=seeded_ids.child_id,
parent_personal_data_uuid=seeded_ids.personal_data_uuid,
date=date(2026, 3, day),
day_type="sjukpenning",
created_at=now,
updated_at=now,
)
for day in (1, 2, 3)
]
await planned_leave_repo.create_many(entities)
rows = await planned_leave_repo.list(workspace_id=seeded_ids.workspace_id)
assert len(rows) == 3
deleted = await planned_leave_repo.delete_many([entities[0].id, entities[1].id])
assert deleted == 2
remaining = await planned_leave_repo.list(workspace_id=seeded_ids.workspace_id)
assert [r.id for r in remaining] == [entities[2].id]
async def test_create_many_empty_list_is_noop(
self,
planned_leave_repo: SQLitePlannedLeaveRepository,
seeded_ids: SeededIds,
) -> None:
await planned_leave_repo.create_many([])
assert await planned_leave_repo.list(workspace_id=seeded_ids.workspace_id) == []
async def test_delete_many_is_idempotent(
self, planned_leave_repo: SQLitePlannedLeaveRepository
) -> None:
"""Unlike `delete()`, a missing id is silently ignored rather than raising."""
deleted = await planned_leave_repo.delete_many([uuid4(), uuid4()])
assert deleted == 0
async def test_update_metadata_many(
self,
planned_leave_repo: SQLitePlannedLeaveRepository,
leave_wish_repo: SQLiteLeaveWishRepository,
seeded_ids: SeededIds,
) -> None:
now = datetime.now()
entity = PlannedLeaveDay(
id=uuid4(),
workspace_id=seeded_ids.workspace_id,
child_id=seeded_ids.child_id,
parent_personal_data_uuid=seeded_ids.personal_data_uuid,
date=date(2026, 4, 1),
day_type="vacation",
created_at=now,
updated_at=now,
)
await planned_leave_repo.create(entity)
wish = await leave_wish_repo.create(
LeaveWish(
id=uuid4(),
workspace_id=seeded_ids.workspace_id,
child_id=seeded_ids.child_id,
name="Test wish",
window_start_month=4,
window_start_day=1,
window_end_month=4,
window_end_day=30,
target_weeks=1,
first_year=2026,
last_year=None,
priority=1,
parent_mode="either",
parent_personal_data_uuid=None,
window_mode="fixed",
created_at=now,
updated_at=now,
)
)
await planned_leave_repo.update_metadata_many([(entity.id, wish.id)])
refreshed = await planned_leave_repo.get_by_id(entity.id)
assert refreshed is not None
assert refreshed.wish_id == wish.id
# day_type/is_dubbeldag are untouched by the metadata-only update.
assert refreshed.day_type == "vacation"
class TestApplyProposedDaysAtomicity:
async def test_mid_batch_failure_leaves_nothing_persisted(
self,
planning_service_with_uow: PlanningService,
planned_leave_repo: SQLitePlannedLeaveRepository,
seeded_ids: SeededIds,
monkeypatch: pytest.MonkeyPatch,
) -> None:
"""A batch that replaces an existing row (delete + create, both inside
stage 4) must not leave the delete applied if the create half fails —
the whole `async with self._uow():` block must roll back together.
Uses `planning_service_with_uow` (a real `SQLiteUnitOfWork`), not the
default `planning_service` fixture, which wires a no-op unit of work
and would (correctly, given a no-op) leave the delete committed.
"""
existing = await planning_service_with_uow.add_planned_day(
workspace_id=seeded_ids.workspace_id,
child_id=seeded_ids.child_id,
parent_personal_data_uuid=seeded_ids.personal_data_uuid,
date=date(2026, 5, 1),
day_type="sjukpenning",
)
async def boom(_items: list[PlannedLeaveDay]) -> None:
raise RuntimeError("simulated mid-batch failure")
monkeypatch.setattr(planned_leave_repo, "create_many", boom)
days = [
ProposedDayInput(
date=date(2026, 5, 1),
parent_personal_data_uuid=seeded_ids.personal_data_uuid,
child_id=seeded_ids.child_id,
day_type="lagsta", # differs from the existing row -> collision -> to_replace
)
]
with pytest.raises(RuntimeError):
await planning_service_with_uow.apply_proposed_days(
workspace_id=seeded_ids.workspace_id, days=days, replace=True
)
remaining = await planned_leave_repo.list(workspace_id=seeded_ids.workspace_id)
assert len(remaining) == 1
assert remaining[0].id == existing.id
assert remaining[0].day_type == "sjukpenning"