Add a nullable ClassId axis to PackDrawCardWeightEntry. Null means the
row applies to any draw (the existing behavior for non-RS packs); 1..8
restricts the row to a specific class draw on a RotationStarterCardPack.
Slot rates are class-invariant (verified against per-klan captures), so
PackDrawSlotRateEntry is unchanged.
PackOpenService.Draw now takes an optional classId parameter. With a
class supplied, the card pool is filtered to (ClassId == classId ||
ClassId == null); without one, class-tagged rows are excluded so a
mistaken non-RS call against a mixed table can't leak class-specific
cards.
PackController.Open lifts the 501 guard on RotationStarterCardPack and
plumbs class_id through. Class_id is now required (and validated 1..8)
for RS packs and rejected as 400 BadRequest on non-RS packs. The skin
overload (target_card_id) remains 501 — out of scope for this work.
Seed JSON regenerated end-to-end: 32 RS packs now carry 30,160 class-
tagged weight rows; 247 non-RS packs keep their existing ~87k null-class
rows. Slot rate fix from the upstream extractor change replaces the
~99.99% conditional-rate sums on 97xxx/93xxx packs with the page's
authoritative 1.5/6/25/67.5 distribution.
Migration AddPackDrawCardWeightClassId. End-to-end bootstrap verified
against fresh Postgres; pack 93025 reads back exactly 122/119/120/120/
120/121/118/119 weights for classes 1-8.
Tests: 3 new PackOpenServiceTests (class filter happy path, disjoint
pools, defensive null-class isolation); existing class_id-on-non-RS
controller test updated 501 → 400 BadRequest; one stub-importer test
corrected to match extractor-authored gacha_detail.
Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com>
Implement all five guild_chat attachment endpoints:
- add_deck: server looks up deck by (viewerId, format, deckNo) and snapshots
the current state into a DECK-type chat message DeckPayload column.
- delete_deck: author OR leader/sub-leader may clear; returns refreshed
deck_log. Extended IGuildChatMessageRepository.ClearDeckAsync with
leaderOverride flag; added GetDeckMessagesAsync + GetByMessageIdAsync.
- add_replay: stores {"battle_id": N} in ReplayPayload; full shape is
TODO(post-merge) pending replay-subsystem capture.
- replay_detail: returns the stored ReplayPayload as JsonElement passthrough.
- deck_log: returns all active DECK messages grouped by API-side Format string
key ("1"=Rotation, "2"=Unlimited, "3"=PreRotation always present;
"4"=Crossover, "5"=MyRotation only when non-empty).
Adds DeckLogDataDto (typed) for deck_log / delete_deck responses.
Replay response uses JsonElement passthrough (TODO(post-merge)).
8 new integration tests — all green. Full suite 1530/1530 pass.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Implements PostTextOrStampAsync: validates membership (NotInGuild),
validates stamp id against UsableStampList (PermissionDenied), rejects
type other than 0 or 1, allocates next per-guild message_id, appends row.
Wires /guild_chat/post controller action mapping failures to result_code=2.
Response is GuildChatPostResponse (empty body per spec). 7 integration tests.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
- ChangeRoleAsync: Leader-only role management with SubLeader cap check.
Atomic leader transfer loads both member rows + Guild.LeaderViewerId in
one _db.SaveChangesAsync call (no split saves). Emits ChangeLeader (6)
or ChangeSubLeader (7) chat events. Same-role is no-op / returns Ok.
- /guild/change_role controller: calls ChangeRoleAsync, returns full updated
member list (GuildChangeRoleResponse.Members[]).
- /guild/friend_list controller: calls IFriendService.GetFriendsAsync, then
annotates each friend with is_join_guild from Viewer.GuildId column.
- IGuildRepository.UpdateLeaderViewerIdAsync: added for completeness even
though the atomic transfer path bypasses it via direct EF mutation.
- Tests: 8 GuildServiceChangeRoleTests (role changes, cap, atomic transfer,
no-op, permission); 3 GuildChangeRoleFriendListFlowTests (HTTP roundtrip,
is_join_guild); 2 GuildWireShape tests (literal JSON shape); 1508 pass.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
- LeaveAsync: regular/subleader removes self; leader blocked if members remain;
sole-member leader auto-routes to BreakupAsync.
- RemoveAsync: leader-only kick (SubLeader has no remove authority per
GuildManager.cs decompile); self-remove and cross-guild target rejected.
- GuildController: wired /guild/leave (BaseRequest) and /guild/remove
(GuildRemoveRequest.remove_viewer_id field).
- SpyGuildChatService: extended Emissions with EmissionsWithBody tuple to
allow body-content assertions in leave/remove tests.
- 8 new integration tests; all 1493 pass.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
GrantIfDueAsync advances streak (1..15 cycle), grants the day's reward via
inventory tx, returns wire-shape DailyLoginBonus. IsDue helper for MyPage.
Also fixes GameConfigurationJsonbTests section-count (15→16) which was broken
since Task 3 added the LoginBonus [ConfigSection] but the assertion wasn't updated.
Closes the family. arena-colosseum 10/16 → 15/16 zero stubs (the 16th —
finish_load — is dead per project_dead_battle_endpoints).
* 2-Pick draft lift onto ArenaColosseumController:
- get_candidate_classes samples from ArenaTwoPickConfig.AllowedClassIds
and persists the slate onto the run.
- class_choose accepts class_id XOR chaos_id; both populate run.ClassId,
chaos branch stores ChaosId for replay.
- get_candidate_cards is the idempotent draft-resume snapshot.
- card_choose appends both cards from the picked pair, advances turn 1..15.
- Pool override: ArenaTwoPickCardPoolService gets a non-breaking
GeneratePickSetsForTurn(..., poolCardSetIds) overload; Colosseum routes
pass ColosseumSeasonConfig.PoolCardSetIds (falls back to challenge →
rotation when empty).
* Curated-deck schema: ColosseumHofDeck / ColosseumWindFallDeck /
ColosseumAvatarDeck — three identical tables (separate per per-pool
operational lifecycle), unique on DeckNo. Migration AddColosseumCuratedDecks.
* IColosseumCuratedDeck interface + a generic ColosseumCuratedDeckImporterBase<T>
with three concrete subclasses (HOF / WindFall / Avatar), registered in
Bootstrap. Seed files ship empty.
* 6 curated endpoints on ArenaColosseumController: get_{hof|windfall|avatar}_deck_list
return BARE-ARRAY data per spec; register_{hof|windfall|avatar}_deck share
one generic RegisterCuratedAsync<T> dispatcher. Cross-pool register
rejected via per-pool lookup. Curated register has no is_published flag
(constructed-only) — clears the run's flag for state consistency.
* Tests: 5 draft HTTP tests + 11 curated-deck HTTP tests (4 parameterized
3 ways across HOF/WindFall/Avatar + a cross-pool isolation test + an
is_published clear test). Existing TwoPick service tests updated for the
new pool overload. Full suite: 1347/1347.
Phase 3 ship gate met. Branch ready for merge.
Closes 4 more arena-colosseum spec shapes (6/16 → 10/16): the dual-prefix
battle URLs (colosseum_battle/* and colosseum_rank_battle/*) for both
do_matching and per-match finish, plus the bracket-end /finish + /retire
pair on /arena_colosseum.
* ColosseumProgressionService: pure-logic decisions for win-threshold
advancement, 3008 promotion trigger, and per-round reward bundles.
Extends ColosseumRoundsConfig with FinishRewards/RetireRewards per
round + a top-level ChampionRewards list.
* MatchContextBuilder.BuildForColosseumAsync: lifts the registered deck
via IDeckRepository (NOT viewer-graph) per project_ef_nav_include_pitfall.
* ArenaColosseumBattleController: dual route prefixes off one controller
via explicit absolute [HttpPost("/...")] routes — does NOT extend
SVSimController's [Route("[controller]")]. Same pattern as FreeBattle.
URL-vs-IsRankMatching mismatch returns 400 with a phase-mismatch error;
no-deck triggers the standard 3001 matchmaking-illegal state. 3008
flips run.IsRankMatching exactly once via the progression service.
* /arena_colosseum/finish + /retire share FinishResponse (rewards +
reward_list + colosseum_status). Champion finish appends ChampionRewards
and surfaces colosseum_status.is_champion = true. Both delete the run
row after granting via IInventoryTransaction.GrantAsync.
* DI: ModePolicy entries for "colosseum_battle" and "colosseum_rank_battle".
* Tests: 8 progression-service unit tests (pure logic), 5 battle-controller
HTTP tests, 5 bracket-terminate HTTP tests, 2 full-bracket E2E walks
(champion path + retire mid-round). Full suite: 1331/1331.
is_skip_gacha_effect, use_challenge_two_pick_premium_card, and
challenge_two_pick_sleeve_id all persist on ViewerInfo and round-trip
through /load/index.
The two challenge fields move from the global ChallengeConfig section to
ViewerInfo — they're viewer preferences, not server-wide policy. Premium
card defaults to 0; sleeve falls back to DefaultLoadoutConfig.SleeveId
(3000011) when unset. MatchContextBuilder's TK2 sleeve read switches to
the same viewer-scoped path.
Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com>
Pure deterministic monthly period generator for the four ranking
families. Anchor dates derived from prod capture (2026-06-09): id=1 is
each family's launch month in JST; id=N is anchor + N-1 months. Used
by /ranking/get_viewable_ranking_period_list to render the period
picker and by per-family leaderboard endpoints to echo the requested
period back.
Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com>
Idempotent on (ViewerId, BattleId); evicts oldest CreateTime row when
at cap. No-op when ctx is null (server-restart safety).
Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com>
Bridges the start-time -> finish-time gap. /finish carries neither
battle_id nor opponent identity; this store holds both for the finish
handler to compose a ViewerBattleHistory row.
Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com>
SendApplyAsync, ApproveApplyAsync, RejectApplyAsync, CancelApplyAsync,
RejectFriendAsync all implemented. 11 new tests appended; all 23 friend
tests pass, full suite 1182/1182 green.
GetFriendsAsync, GetReceiveAppliesAsync, GetSendAppliesAsync, GetPlayedTogetherAsync,
SearchAsync all implemented. LoadViewerProjectionAsync materialises the full Viewer
entity (with Include/ThenInclude for SelectedEmblem/Degree) then projects in-memory —
avoids the EF Core limitation where Include is silently ignored under Select.
FriendService + IPlayedTogetherWriter registered as Scoped in Program.cs.
12 read tests, all green; full suite 1171/1171 still passing.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Centralizes which UserGoodsType values IInventoryTransaction.GrantAsync
can handle in the gift-inbox flow; both GiftController (post-fix) and
CampaignController will consult this instead of duplicating the set.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Adds Commit_writes_history_row_with_Unknown_source_when_caller_omits_Source
to InventoryHistoryTests — asserts that BeginAsync without a configure callback
writes AcquireType=0 / Message="Unknown", per spec §10.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Adds PruneAcquireHistoryAsync to InventoryTransaction.CommitAsync; runs
inside the open DB transaction after history rows are flushed, keeping at
most 300 rows per viewer (oldest discarded). Adds a covering test that
verifies the cap and per-viewer isolation.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
- Drop IsWalletCurrency (duplicate of IsCurrency); use IsCurrency in WriteAcquireHistory.
- Add comment on first SaveChangesAsync in CommitAsync explaining the two-phase flush.
- Guard WriteAcquireHistory loop with grant.Num == 0 check so synthetic DebitItem post-state ops do not produce history rows.
- Add InventoryHistoryTests.Commit_writes_no_history_row_for_item_debit to lock in the fix.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
CommitAsync now calls WriteAcquireHistory() between the two SaveChanges
calls: iterates _ops, skips SpendOps, writes one ViewerAcquireHistoryEntry
per GrantOp. Cascade rows get GrantSource.CardCosmeticCascade; wallet
currencies zero RewardDetailId; all rows in a single commit share one
DateTime.UtcNow timestamp. Closes _source plumbing from Task 5.
5 new tests added (46 total inventory, 0 failures).
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Adds the ViewerAcquireHistoryEntry model (8 fields: Id, ViewerId,
RewardType, RewardDetailId, RewardCount, AcquireType, Message,
AcquireTime), registers DbSet<ViewerAcquireHistoryEntry> on
SVSimDbContext, configures model (PK, FK cascade to Viewer, MaxLength
64 on Message, composite index on ViewerId/AcquireTime/Id), and adds a
DbSet round-trip integration test.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Adds a `GrantSource Source { get; set; }` property (defaults to
`GrantSource.Unknown`) to `InventoryLoadConfig`. Plumbing-only — no
behavior change; callers that don't set `Source` get Unknown rows,
greppable via `acquire_type=0` in dev.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Introduces GrantSource (17 values, DB-persisted) and GrantSourceMessages.For()
for the item-acquire-history feature. Values 1/2 mirror prod captures;
coverage test verifies every enum value has a non-empty message.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Singleton keyed by ShortUdid; lock on per-viewer set to avoid
cross-viewer contention. Process lifetime — restart re-fires.
Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com>
Behavior-preserving; full solution builds, 1013 tests green.
ClassId is the one genuinely-closed set of the three flagged stringly fields, so it
becomes a CardClass enum (1..8). Wire stays "1".."8": producer casts
(CardClass)run.ClassId, ServerBattleFrames renders via CardClassWire.ToWireValue().
RankBattleController's AI-start path drops a fragile int.TryParse(...)?:-1 for (int)cast.
CharaId (free-form leader/skin id, e.g. "5000123") and CountryCode (open-ended account
data) stay string with proper XML docs; CountryCodes.Korea/Japan name the captured values.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Behavior-preserving; 271 BattleNode/Matching/Services tests green, full solution builds.
"BattleType" meant two things: the Sessions.BattleType enum (Pvp/Bot) and an int
"mode id" field. Renamed the int field on MatchContext AND the BattleStartBody wire
DTO to BattleModeId (wire key stays "battleType" via JsonPropertyName), so BattleType
now means only the enum project-wide.
New Bridge/BattleModes.cs (TakeTwo = 11) replaces every 11 literal — both prod
MatchContextBuilder sites and the test fixtures/assertions. The arbitrary-passthrough
42 and bot 0 stay literal.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
BuildForRankBattleAsync projected deck.Cards.Select(c => c.Card.Id),
discarding Count. DeckCard is count-based (one row per unique card +
a Count), so a 3-copy card shipped to the node as a single in-battle
card -- matched decks showed 1 of each card instead of the real count.
Expand each row by its Count so SelfDeckCardIds carries one entry per
physical card. TwoPick path is unaffected (flat per-pick list).
Add a regression test seeding 3+2+1 copies (failed Expected 6/was 3).
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
ISteamServer contract forbids null tickets (prod impl and sole caller both assume non-null),
so the dev bypass no longer needs the ?. / ?? 0 defensive form. Also adds a class-level XML
doc summary to DevAlwaysValidSteamServerTests matching the style of other fixtures in the suite.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Replace bare `int RewardType` on 12 catalog/reward entities and GrantedReward
with the existing UserGoodsType enum. Verified against the decompiled client:
every wire reward_type decodes through the single Wizard.UserGoods.Type enum, so
one enum is correct across all endpoint families (item_type is a separate
Item.Type axis, left untouched). EF stores the enum as the same int column, so
there is no migration.
- Importers cast seed int -> UserGoodsType at the ingest boundary.
- New GrantedReward.ToRewardList() extension replaces 8 copy-pasted
GrantedReward -> RewardListEntry projections.
- Fix 3 .ToString() sites that would otherwise emit enum names ("Crystal")
instead of the int wire value ("2").
- Wire DTOs keep int; the enum is widened to int at the wire boundary only.
Build green; 962/962 tests pass.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Live-smoke bug 2026-06-02: queued Bloodcraft (deck #5), wire showed
classId=2 (Swordcraft) for self_info on the /ai_unlimited_rank_battle/start
response — client rendered the wrong leader.
Two layers of the same bug:
1. MatchContextBuilder.BuildForRankBattleAsync hardcoded deckNo=1 instead
of taking it from the do_matching request — verified against
data_dumps/captures/traffic.ndjson L17 where deck_no=5 was on the wire.
Signature changes to (viewerId, format, deckNo); DoMatchingInternal
passes req.DeckNo.
2. AiStartInternal rebuilt MatchContext from scratch — but the /ai_*/start
request body is BaseRequest only, no deck_no on the wire. The fix uses
the MatchContext the bridge already stored at do_matching resolution time
(in the Bot PendingBattle), so deck/cosmetic data is consistent end-to-end.
New IBattleSessionStore.TryFindPendingForViewer(viewerId) finds the
viewer's pending battle for lookup. The store entry persists across
ai_start (idempotent reads are fine — the WS handler removes on connect).
No-pending sentinel: ai_id=-1 surfaces the "no AI assigned" error in the
client.
Tests: 936 → 939 passing.
- MatchContextBuilderTests.BuildForRankBattle_uses_the_caller_supplied_deck_number
seeds deck #1 (class 1) and deck #5 (class 6) and asserts the deckNo
argument picks the right one.
- RankBattleControllerTests.AiStart_self_info_class_matches_queued_deck_number
is the end-to-end regression: register Bot battle with deck #5, hit
/ai_unlimited_rank_battle/start, assert self_info.classId == 6.
- RankBattleControllerTests.AiStart_without_pending_battle_returns_neg1_sentinel
locks the defensive ai_id=-1 path.
- Existing AiStart_* tests bypass do_matching, so adapted to call a new
RegisterBotBattleAsync helper that mirrors what InProcessPairUp does on
AI-fallback resolution.
SeedDeckAsync gains an optional classId so test cases can differentiate
decks by class (was always picking Classes.First()).
Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com>
Sibling to BuildForTwoPickAsync. Routes through IDeckRepository.GetDeck
to pull the viewer's deck #1 for the requested format (avoiding the
viewer-graph nav-ref auto-load pitfall — DeckCard.Card silently ships
card_id=0 via the default include path). Throws if the viewer has no
deck for the format. Cosmetics fall back to DefaultLoadoutConfig
defaults when unequipped, same shape as TK2.
Used by RankBattleController in a later task to build self-context for
/ai_<fmt>_rank_battle/start and to pair-up under /<fmt>_rank_battle/do_matching.
Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com>
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>
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>
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>
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>
Last post-state per currency wins; non-currency grants collapse to final
count per (type, id). Deltas are verbatim queued, no cascade. SaveChanges
+ DB tx commit happen atomically inside Commit; failure leaves rollback
to DisposeAsync. CS0649 warning on _committed is now resolved.
Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com>
Card grants produce a post-state-total entry and run the CardCosmeticReward
cascade (foil twin → id-1 lookup). Cascade additions are skipped when the
viewer already owns the cosmetic; missing-master-row failures are logged
and dropped without failing the parent grant.
Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com>