BattlePassRepository._curveCache and MissionCatalogRepository._maxLevelCache
were private-static fields populated lazily on first read from whatever
DbContext happened to be in scope. In production "one DbContext lineage
per process" makes that fine. Under parallel test execution each
SVSimTestFactory owns its own SQLite :memory: DB, so the first reader's
DB (often empty, in tests that don't seed BP) poisoned the cache for
concurrent readers from a seeded DB — assertions like "BP level info
must be present after seeding" failed because the process-static cache
returned an empty list populated by the other test's empty DB.
The first patch attempted a `BypassCacheForTests` static flag, which is
exactly the kind of test-only seam that rots the production code: future
caches get the same flag, repos accumulate hidden knobs, and the
underlying invariant ("a cache populated from arbitrary scope serves
arbitrary scope") goes unaddressed.
Instead, move both caches into the DI-registered IMemoryCache.
AddMemoryCache() registers it as singleton-per-service-provider:
production has one provider → one IMemoryCache → identical caching
semantics to before. Each WebApplicationFactory builds its own
provider → its own IMemoryCache → cache is naturally scoped per fixture,
no cross-test bleed possible.
The ResetLevelCurveCache() method and its three call sites
(SVSimTestFactory.SeedGlobalsAsync, BattlePassServiceTests,
LoadControllerTests) are deleted — a fresh factory owns a fresh empty
cache, no manual invalidation needed.
With this and the previous StoryService fixture-instance fix in place,
ParallelScope.All works: 776/776 in 57s wall clock (down from 59s on
Fixtures, 2m13s pre-parallelism).
Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com>
NUnit's default FixtureLifeCycle is SingleInstance — every test in a
class shares one fixture instance, so [SetUp]-initialised fields like
_master / _viewer / _service are reset on every test against the same
object. Under serial execution that's fine; under parallel execution
concurrent SetUps wipe each other's Mock setups and the service code
NREs trying to dereference unconfigured stubs.
Compounding it, NewInMemoryDb was being called with nameof(SetUp) which
is the literal string "SetUp", so every test in the fixture also shared
the same EF InMemory database (the provider keys stores by name).
Two fixes:
- [FixtureLifeCycle(LifeCycle.InstancePerTestCase)] on StoryServiceTests
so each test gets its own instance with its own Mocks.
- Suffix the InMemoryDb name with a Guid so concurrent callers never
share a store.
Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com>
The full-catalog regression test hardcoded "35 active packs as of
2026-05-23" but the controller filters by DateTime.UtcNow against each
pack's commence/complete dates. When two packs (99047, 80047) crossed
their complete_date of 2026-06-01 01:59:59 UTC, the test started
failing with Expected: 35 / But was: 33 — which had been masked all
along by NUnit's trx serializer OOMing on a different test.
The hardcoded count conflated three things that happened to be equal
on the day the test was written: packs in the seed file, packs active
right now, and 35. The test's real intent (per its class docstring) is
"every pack the importer ingests round-trips through /pack/info";
pinning the clock with TimeProvider would solve today's drift but
re-break the moment someone regenerates the seed or retires a pack.
Expected count now derives from the seed file at test time, filtered
by the same predicate the controller uses (PackRepository
.GetActivePacks: IsEnabled && commence <= now <= complete) via the
shared ImporterBase.ParseWireDateTime parser so any date-string quirk
parses identically on both sides. Spot-check on pack 99047 swapped for
"any pack with non-default pack_category" — same schema-fidelity
coverage (non-zero category survives JSON round trip) without pinning
to an id that rotates.
Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com>
The unit-test suite was spending most of its wall clock writing logs.
NUnit captures stdout per test and embeds it in the trx; with HttpLogging
emitting full request/response per controller call, EF Core SQL at
Information level, and ReferenceDataImporter banners running ~500x
(once per factory construction), the trx grew to 3.2 GB and the NUnit
result-XML serializer OOMed in StringBuilder.ToString() — which the
runner reported as one mysteriously failed test, masking a real
date-dependent failure underneath.
Three sources silenced under environment "Testing":
- appsettings.Testing.json drops Default + Microsoft.AspNetCore +
HttpLoggingMiddleware + EntityFrameworkCore to Warning.
- Program.cs skips app.UseHttpLogging() entirely (avoids the
middleware overhead, not just the log emission).
- ReferenceDataImporter takes optional TextWriters; the test factory
passes TextWriter.Null. Per-importer helpers become instance methods
so they can use the injected writer.
Result on a fresh run with ParallelScope.Fixtures already in place:
- Test duration: 1m46s -> 59s
- Wall clock: 2m23s -> 1m00s
- trx size: 3.2 GB -> 1.7 MB
The previously-masked date-dependent failure (PackControllerFullCatalog
.Info_returns_full_35_pack_catalog_from_production_seed asserting 35
active packs as of 2026-05-23 against a live clock) is now visible and
can be addressed separately.
Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com>
NUnit's default ParallelScope is Self (serial). With ~736 tests each
constructing its own SVSimTestFactory (full ASP.NET host + SQLite :memory:
+ ReferenceDataImporter seeding 7270 rows from CSVs), the suite was
running ~2m13s serial. ParallelScope.Fixtures drops it to ~1m46s — a
~20% wall-clock reduction with zero new failures.
Stayed at Fixtures rather than All because ParallelScope.All exposes
the process-static BattlePassRepository._curveCache (and likely other
similar caches) to races inside heavy-globals fixtures (LoadController,
PackControllerFullCatalog, StoryService — all consistent failures
under All, flaky 3-7 fails across runs). Within-fixture parallelism
is blocked on cleaning those up first.
Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com>
Boots SVSimTestFactory (in-memory SQLite + reference-data CSV import),
mints a battle via IMatchingBridge, opens a raw Socket.IO v2 client
against the in-process TestServer, drives InitNetwork → Loaded → Swap,
and asserts the right scripted frames come back in order.
Verifies the full transport stack end-to-end: EIO3+SIO2 framing,
encryptForNode codec, MsgPayloadCodec roundtrip, InboundTracker
pubSeq dedup + ack echo, OutboundSequencer playSeq assignment, and
ScriptedLifecycle's Path-A frame builders.
Note: RawSocketIoTestClient.DisposeAsync skips the graceful CloseAsync
handshake — TestServer's in-process WebSocket implementation can hang
on it. Abrupt Dispose is fine: the server's ReceiveAsync throws
WebSocketException, BattleSession.RunAsync returns, and the handler
completes.
Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com>
Cast GetHashCode() result to long before Math.Abs to prevent OverflowException
on the ~1-in-4B case where GetHashCode returns int.MinValue. Adds a regression
test pinning the 12-digit decimal format end-to-end.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
- Wrap HandleMsgEventAsync / HandleAliveEventAsync bodies in try/catch(Exception)
logging at Error, eliminating async-void unobserved-exception crash risk (Issue 1).
- Replace deterministic seq-based key generator with RandomNumberGenerator.GetInt32
so each EncodeAndSendAsync call uses a fresh random key (Issue 2).
- Add `when Phase == …` guards to InitNetwork / Loaded / Swap cases in
ComputeResponses; add default arm that logs+drops out-of-order URIs (Issue 3).
- Widen SendSioAckAsync arg from int to long; drop (int) cast at call site;
boundary cast to int is now checked() for defensive overflow detection (Issue 4).
- Update RunAsync doc comment (was stale Task-13 placeholder) (Issue 5).
- Add Kill and out-of-order-Swap-before-Loaded tests (Issue 6).
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
ToJson now throws ArgumentException when a Body key collides with a reserved
envelope field (uri/viewerId/uuid/bid/try/cat/pubSeq/playSeq); FromJson reuses
the same shared ReservedEnvelopeKeys HashSet. ReceiveNodeResultCode expanded
from 9 to 31 codes to mirror the full enums.md catalog. Two regression tests
added for the collision guard and PascalCase uri serialization.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
- Wrap all JsonDocument.Parse calls in using blocks and Clone() each
retained JsonElement to eliminate UAF hazard after GC.
- Use JsonSerializer.Serialize with UnsafeRelaxedJsonEscaping so event
names with " or \ produce \" / \ rather than " / plain \;
avoids malformed JSON on Encode().
- Guard the [ ] block in Encode() behind EventName-or-args check so
Connect/Disconnect packets round-trip as bare "0"/"1" not "0[]".
- Add three regression tests: Connect no-bracket, Event round-trip,
special-char event name escaping.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Replace inaccurate GenerateKey docstring (it claimed to port Cryptographer.generateKeyString
directly but the input shape differs: server uses one hex digit per call, client uses
Random.Next(0,65535) per call). New doc is honest about the difference and explains why
it's safe. Add EncryptForNode_FixedVector_ProducesStableOutput: a pinned AES-CBC vector
that catches encoding/IV/padding regressions that would slip past the roundtrip test.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
CommitAsync's inner SaveChangesAsync already flushes the AddClaim
rows + progress.IsPremium mutation alongside the inventory grants
(same scoped DbContext). The trailing _db.SaveChangesAsync was a
no-op in BuyPremium and only meaningful in AddPoints when no level
crossed (no tx opened) — restructured to an else branch.
Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com>
Serializes result.RewardList with snake_case+WhenWritingNull options and
asserts the three entries come out in expected first-touch order:
Crystal post-state (500), Card post-state count (3), Sleeve cascade (1).
Also verifies snake_case key names are actually emitted.
Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com>
Removed RewardGrantService, CurrencySpendService, ICurrencySpendService,
ViewerEntitlements, IViewerEntitlements, CardAcquisitionService,
ICardAcquisitionService, CardGrantResult and their tests
(RewardGrantServiceTests, CurrencySpendServiceTests,
CardAcquisitionServiceTests, ViewerEntitlementsTests). Removed four DI
registrations from Program.cs. No caller references any deleted type;
GrantedReward and EffectiveCosmetics were pre-moved to InventoryGrantTypes.cs
in the prior commit. Build clean, 712/712 tests pass.
Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com>
Both types stay in namespace SVSim.Database.Services so existing using directives
in controllers, services, and tests resolve without change. Their definitions are
extracted to SVSim.Database/Services/Inventory/InventoryGrantTypes.cs; the empty
husks in RewardGrantService.cs and IViewerEntitlements.cs will be deleted in the
next commit.
Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com>
Change signature from (Viewer, packId, cardId) to (IInventoryTransaction, packId, cardId).
Drop RewardGrantService from GachaPointService ctor. PackController.ExchangeGachaPoint opens
tx with GachaPointBalances/Received extra includes, passes tx, commits on success.
Update GachaPointServiceTests to use inv.BeginAsync + tx pattern.
Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com>
Replace RewardGrantService + ICurrencySpendService + IViewerEntitlements with
IInventoryService. tx.IsFreeplay replaces FakeEntitlements.IsFreeplay; debit
helpers take IInventoryTransaction. ComputePostStateRewardList deleted (replaced
by result.RewardList from CommitAsync). Update 5 test files to new 8-arg ctor.
Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com>
Replace RewardGrantService + ICurrencySpendService with IInventoryService tx.
CommitAsync's currency-collision rule replaces the manual Crystal RemoveAll+re-append
scrub in BuyPremiumAsync. AddPointsAsync uses result.Deltas for NewlyClaimed to
preserve per-track visibility (two Rupy grants stay two entries).
Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com>
Replace RewardGrantService with IInventoryService tx. Per-reward GrantAsync
calls inside try/catch preserve the NotSupportedException skip; CommitAsync
returns result.RewardList (post-state totals) and accumulated delta list feeds
story_reward_list. Update StoryServiceTests to inject IInventoryService.
Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com>
Replace RewardGrantService + HttpContext.RequestServices viewer load with
IInventoryService tx. Single BeginAsync/GrantAsync/CommitAsync wraps all
mission rewards on the win path.
Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com>
Replace RewardGrantService with IInventoryService tx. EnsureCurrentAsync
still runs before BeginAsync to avoid EF concurrent-context conflicts;
tx.Viewer replaces the manually loaded viewer graph.
Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com>
Replace RewardGrantService with IInventoryService tx. GrantAsync returns
post-state totals directly, eliminating the manual ResolvePostStateRewardNum
helper. MissionData loaded via extra include on BeginAsync.
Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com>
RedEther debit now goes through tx.TrySpendAsync (freeplay-aware);
Card grants route through tx.GrantAsync (cosmetic cascade for first-time
owners). Validation phase unchanged. DestructCards left on direct-viewer
path (structural mismatch: validation on one viewer, mutation on same
instance — clean tx port deferred to follow-up).
Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com>
EffectiveBalance/OwnsCard/OwnsCosmetic on the tx are freeplay-aware
against the live viewer. EffectiveOwnedCardsAsync/EffectiveCosmeticsAsync
on the service mirror today's ViewerEntitlements projections (used by
/load/index).
Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com>
