- ColosseumBattleFinishResponseDto gains get_class_experience,
class_experience, class_level. Client-safe: same BattleFinishResponsProcessing
handles all three keys as ints (decomp verified).
- ArenaColosseumBattleController.FinishInternal grants win/loss XP via
IBattleXpService with BattleXpMode.Colosseum for both /colosseum_battle/finish
and /colosseum_rank_battle/finish URLs.
- Existing win test extended: XP wire assertions + persistence check.
Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com>
Implements Task 9 from docs/superpowers/plans/2026-06-13-decomp-only-followups.md
— the pre-purchase commit endpoint the client fires from StarterClassSelectDialog
before opening a RotationStarterCardPack. Without this the per-class draw path
from yesterday's work is unreachable: the dialog blocks on the AJAX call and
never advances to /pack/open.
Wire surface:
- POST /pack/set_rotation_starter_class { pack_id, class_id } → EmptyData.
Validates 1..8, rejects unknown packs (404), non-RS packs (400), and second
commits (400 — one-shot per pack per spec).
- PackConfigDto carries selected_class_id (nullable, omitted via global
WhenWritingNull policy when unset). Placement verified against decomp at
PackInfoTask.cs:86 — it's on the parent PackConfig, NOT the child gacha as
the original plan text had it.
- /pack/open cross-checks request.class_id against the persisted choice for
RS packs; rejects 400 on missing-commit or class-mismatch so a tampered
request can't swap pools after the choice is locked.
Storage:
- ViewerPackStarterClass owned entity; (ViewerId, PackId) unique index via the
pattern from project_owned_collection_unique_index memory.
- Migration PackStarterClass — composite PK + FK cascade verified against
fresh Postgres bootstrap.
Tests (9 new in PackControllerSetRotationStarterClassTests):
- Round-trip set → /pack/info shows selected_class_id.
- Field absent before any commit.
- Invalid class (0/9/-1/100) → 400.
- Already-chosen rejection.
- Non-RS pack / unknown pack rejection.
- /pack/open rejects no-prior-commit AND class-mismatch; succeeds when class
matches the persisted choice.
Full suite: 1552 passed (1543 + 9).
Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com>
C1: replay_detail — flatten stored payload to data level. ChatReplayDetailTask.Parse() calls
new ReplayDetailInfo(data) which unguardedly accesses data[battleId], data[seed], data[vid1],
etc. Wrapping under replay_info key crashes the client. Controller now returns Ok(JsonElement)
directly so stored battle fields are at data root. Wire-shape test added.
C2: Guild.LeaderViewerId long to long?; add HasOne<Viewer> FK with OnDelete=SetNull; migration
AddGuildLeaderViewerIdFk; all consumers null-guarded with ?? 0L.
C3: BreakupAsync — wrap 6 destructive ops in IDbContextTransaction with InMemory fallback.
C4: CommitJoinAsync — wrap member-add + viewer-guildId-set + invite/request cleanup in
IDbContextTransaction with InMemory fallback. Chat event emitted after commit.
I1: GuildController — remove SVSimDbContext field; inject IViewerRepository + IGuildMemberRepository.
Add IGuildMemberRepository.GetViewerIdsInAGuildAsync (batch guild-membership check).
All _db.Viewers / _db.GuildMembers queries replaced with repo calls.
I2: GuildService — extract 3 _db.Viewers queries: GetEquippedEmblemIdAsync (CreateAsync),
LoadGuildProfileBatchAsync (ListOutgoingInvitesAsync + ListPendingJoinRequestsForMyGuildAsync).
Add GuildMemberProfile record with IsOfficialMarkDisplayed. GetEmblemListAsync for EmblemList.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Remove StringifiedIntConverter from GuildChatMessagesResponse.WaitInterval so the
wire emits a raw JSON number (3) matching prod, not a quoted string ("3").
Update GuildChatPollTests (2 assertions) and GuildCaptureReplayTests (1 assertion +
comment) to expect JsonValueKind.Number. Fix last_message_id → start_message_id typo
in GuildChatMessages_Member_ReturnsProdShape request body.
Co-Authored-By: Claude Sonnet 4.6 <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>
GuildFriendListTask.Parse() reads base.ResponseData[data][i] directly —
data must be a bare JSON array, not {friends:[...]}. Delete the
GuildFriendListResponse wrapper DTO; promote GuildInviteCandidateDto to its
own file. Controller now returns List<GuildInviteCandidateDto> so the
middleware wraps it as data:[…]. Update wire-shape and integration tests to
assert array-at-root.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
GuildUpdateTask.Parse() reads data[guild] as GuildDetailInfo directly; the previous
GuildDetailSubTree wrapper (data[guild][detail]) caused json[guild_id].ToInt() to
crash on the client. Split GuildUpdateResponse (flat) from GuildUpdateEmblemResponse
(nested, correct for GuildEmblemUpdateTask.Parse). Also surfaces guild_name from the
request through UpdateGuildRequest.Name -> service -> repo so the client rename is honoured.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Pre-materialize ViewerEventCounters before the mission loop so in-flight
Adds are visible to subsequent iterations sharing the same (EventKey, Period);
add optional ImportMission.Slot field so callers can supply weekly slot identity
(2 or 3) rather than relying on the LotType-derived default of 1.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
- numeric fields ship as quoted strings (campaign_id, img, effect_id,
reward_type, reward_detail_id, reward_number)
- DailyLoginBonus.Campaign is a List<>, not a single object
- LoginBonusCampaign.Img replaces Image (string, asset-key)
- add IsOneDayMultiRewards (SpecialData-only on client, byte-faithful)
The captured shape ({"1":[], "3":[], "4":[]}) is the empty-period
wire echo; the client parser (LoadDetail.cs:553) only reads normal/total/
campaign. Table, entity, seed, and importer were never wired to a code path.
Replaced by config-section catalog + service in the follow-up commits.
Three fixups the Phase 1-3 work missed (would bite the first time an admin
flipped on a real event):
* IsDisplayTips + TipsId added to ColosseumSeasonConfig. Drives
Wizard/ColosseumEntryInfoTask.cs:129 (NeedsFirstTips = is_display_tips == 1).
The wire DTO already had the properties; the season config didn't carry
them, so admins had no way to gate the first-time tutorial tip popup.
* colosseum_id is now emitted from SeasonId. Wire field was always-absent
before (DTO had the property but builder never set it); client fell back
to 0 and couldn't disambiguate cups.
* ColosseumLobbyInfoBuilder.Build now takes ColosseumRoundsConfig + nowUtc
and derives the wire is_round_period / now_round / next_round / start_time
/ end_time from the rounds schedule against the current clock. Previous
hardcode (is_round_period: true, now_round: 1, event-window dates) would
have shown "Round 1 ends 2026-07-31" instead of the actual round's end.
Per the client parser at ColosseumEntryInfoTask.cs:111-126:
- inside a round → emit that round's window + is_round_period=true
- between rounds → emit next round's start as start_time + next_round
- pre-event → next_round=1, start_time=event start
- post-event → anchor to last round so the client doesn't NRE
- no rounds configured → fall back to event-level window
Why these were missed: I implemented ColosseumSeasonConfig from the plan's
explicit field list rather than cross-checking against the captured
colosseum.json wire shape; I added ColosseumId to the wire DTO without
wiring the builder side; and no test exercised a mid-event scenario with
realistic dates — Phase 2's E2E hand-edits run.RoundId via
PromoteRunToRoundAsync instead of letting the controller derive it.
* 10 new builder unit tests cover off-season / mid-round / between-rounds /
pre-event / post-event / empty-rounds-fallback + the new field surfacing.
Full suite: 1355/1355 (was 1345 + 10).
/mypage/index data.colosseum_info and /arena_colosseum/{top,get_fee_info}
now both project through ColosseumLobbyInfoBuilder.Build(ColosseumSeasonConfig).
One source of truth for the home-tab gating and the lobby reads — admins
flip the season on by writing to GameConfigs (per the runbook), no second
table to keep in sync.
* New ColosseumLobbyInfoBuilder — static projection from ColosseumSeasonConfig
to the wire ColosseumLobbyInfo. Replaces ArenaColosseumController's private
BuildColosseumInfo and MyPageController's BuildColosseumInfo(ColosseumConfig).
* MyPageController reads ColosseumSeasonConfig via IGameConfigService; the
IGlobalsRepository.GetCurrentColosseum() call goes away.
* MyPageIndexResponse.ColosseumInfo retyped to the new ColosseumLobbyInfo.
* ColosseumSalesPeriodInfo relocated into the ArenaColosseum namespace
alongside the rest of the family DTOs.
* Drops: Colosseums DB table (migration DropColosseumsTable), ColosseumConfig
entity, the captured-shape Models.Dtos.ColosseumInfo DTO, ColosseumSeed,
MyPageGlobalsImporter.ImportColosseumAsync + seed file colosseum.json,
IGlobalsRepository.GetCurrentColosseum + impl, the two related tests.
Net: home-screen Grand Prix tab now hidden by default because the season
ships with IsColosseumPeriod=false. Full suite: 1345/1345 (was 1347 — the
2 dropped tests covered the now-dead surface).
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.
Strict yyyy-MM-dd parse (matches /load/index round-trip format) — rejects
malformed input. data_headers.servertime is emitted by the standard
envelope, which is what the client reads into BirthDayUpdateServerTime.
Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com>
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>
Translation middleware now extracts viewer_id/steam_id/steam_session_ticket
from the decrypted msgpack dict into HttpContext.Items before the typed
DTO deserialize. The Steam handler reads from there instead of re-parsing
Request.Body — so authed action DTOs no longer need to inherit BaseRequest
to keep the auth fields alive through the msgpack→DTO→JSON pivot.
Retires the recurring footgun documented in
docs/superpowers/specs/2026-06-02-baseRequest-auth-footgun-improvement.md
(2026-05-25 basic-puzzle, 2026-05-28 deck-code, 2026-06-02 Phase 3 Bot,
2026-06-10 profile/index + item_acquire_history/info + user_mypage/update).
Pinned by AuthDecouplingTests — posts an encrypted msgpack body to
/profile/index (DTO does not inherit BaseRequest) through the real
translation middleware + auth handler and asserts 200. Adds an
EncryptedMsgpackHelper + useRealAuthHandler factory flag, reusable for
future wire-shape tests.
ProfileIndexRequest, ItemAcquireHistoryInfoRequest, and
UserMyPageUpdateRequest revert to the naked shape — the per-DTO
workarounds become vestigial under the new architecture.
Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com>
All numeric fields on ReplayInfoItemDto ship as string — matches prod
capture (frame 96 of traffic_prod_misc_clicking.ndjson). api-spec doc
will be updated separately.
Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com>
