feat: make proposal responses durable

multiplayer-todo.md

@@ -1195,7 +1195,7 @@ the local/CI/community transport, not a silent production fallback.
 | 8.15 `[D:7.8,8.3]` | **IN PROGRESS.** Pure Go probe validation treats Steam location as opaque, requires nonce/freshness/region and server-computed RTT, and implements discrepancy quarantine/release; authenticated HTTP now accepts only opaque location/nonce input through a server-owned probe provider | `server/domain/probes.go`, adversarial fixtures and `server/api/service.go` cover stale/wrong/forged evidence, the 25 ms/30% threshold, three-sample quarantine, five-clean release, authenticated provider arguments and rejection of client RTT fields; Steam coordinator and regional probe adapters remain |
 | 8.16 `[D:8.14,8.15]` | **IN PROGRESS.** Pure Go candidate/team selection implements the <=100 ms ceiling, pairwise widening tolerance, anchor inclusion, deterministic set/region scoring and balanced team partitioning; queue-backed formation now consumes the server-owned projection, fences duplicate player identities and rejects playlist/build/protocol mixing | `server/domain/matcher.go`, `teams.go` and adversarial fixtures cover no-common-region, tolerance boundaries, lexical ties, mean-rating balance, malformed candidates, duplicate identities, compatibility mismatches and queue-backed oldest-anchor formation; full population fixtures and durable matcher claim integration remain |
 | 8.17 `[D:8.14,8.16]` | **IN PROGRESS.** Pure Go proposal policy sends a 10-second response window to every selected human, requires unanimous acceptance, applies exact decline/timeout cooldowns and ranked escalation; authenticated API exposes revisioned accept/decline mutations; formed matches now pass through a playlist-aware proposal boundary | `server/domain/proposal.go`, `formation.go` and `server/api/service.go` plus adversarial fixtures cover partial/unanimous response, expiry, replay/conflict, stale API revision, casual lineup preparation and ranked metadata validation; queue precedence and allocation integration remain |
-| 8.18 `[D:8.5,8.14,8.17]` | **IN PROGRESS.** Go store layer defines PostgreSQL SERIALIZABLE whole-transaction retries and queue candidate/proposal claim SQL using `FOR UPDATE SKIP LOCKED` plus durable uniqueness/revision fences; proposal creation now inserts proposal/participants and promotes every ticket in one rollback-safe transaction with player- and playlist-bound claim predicates, queue creation has a durable idempotency/owner-read adapter, and participant-scoped proposal recovery now expires OPEN proposals and pending participants transactionally at read time | `server/store/serializable.go`, `queue_sql.go`, `proposal_sql.go`, `proposal_recovery_sql.go` and tests cover retry classification, claim-boundary invariants, player/ticket/playlist mapping, durable queue replay/conflict, owner-scoped queue/proposal recovery, expiry/read boundaries, zero-row claim aborts and atomic statement ordering; live PostgreSQL adapter execution, Redis candidate index/repair, worker-failure and concurrent two-matcher integration tests remain |
+| 8.18 `[D:8.5,8.14,8.17]` | **IN PROGRESS.** Go store layer defines PostgreSQL SERIALIZABLE whole-transaction retries and queue candidate/proposal claim SQL using `FOR UPDATE SKIP LOCKED` plus durable uniqueness/revision fences; proposal creation now inserts proposal/participants and promotes every ticket in one rollback-safe transaction with player- and playlist-bound claim predicates, queue creation has a durable idempotency/owner-read adapter, participant-scoped proposal recovery now expires OPEN proposals and pending participants transactionally at read time, and proposal accept/decline now uses participant/proposal locks, revision fencing and durable idempotency | `server/store/serializable.go`, `queue_sql.go`, `proposal_sql.go`, `proposal_recovery_sql.go` and tests cover retry classification, claim-boundary invariants, player/ticket/playlist mapping, durable queue replay/conflict, owner-scoped queue/proposal recovery, expiry/read boundaries, response replay/conflict, stale revisions, zero-row claim aborts and atomic statement ordering; live PostgreSQL adapter execution, Redis candidate index/repair, worker-failure and concurrent two-matcher integration tests remain |
 | 8.19 `[D:8.18]` | **IN PROGRESS.** Pure Go casual lineup requires 2–6 humans with at least one per team, fills missing slots with explicit bots, permits kickoff-only bot-slot backfill and assigns no backfill penalty/rating update; proposal preparation now derives the lineup from formed teams | `server/domain/casual.go`, `formation.go` cover both-team minimum, bot shape, live-play rejection, zero-penalty backfill and casual proposal composition; queue candidate selection, opt-in 10 s backfill proposals, reconnect/leave penalties and live integration remain |
 | 8.20 `[D:8.18]` | **IN PROGRESS.** Pure Go ranked admission requires six unique verified solo humans, rejects bots/backfill/parties, and allows only random-enabled non-elevated arenas; proposal preparation requires matching metadata for every formed player | `server/domain/ranked.go`, `formation.go` cover count, identity, party, bot/backfill, arena eligibility and formed-player metadata rejection; `ArenaRegistry` integration, allocation wiring and innocent-ticket restoration remain |
 | 8.21 `[D:8.5,8.20]` | **IN PROGRESS.** Pure Go rating core implements canonical Glicko-2, daily inactivity, ranked 1/3 and casual 1/N human-opponent weights, deterministic opponent ordering, and authoritative draw/overtime/abandon scoring | `server/domain/rating.go` has canonical/inactivity/weight/invalid-input plus draw/OT/abandon fixtures; PostgreSQL snapshot locking, rating transaction integration, seasons and concurrent result transaction tests remain |
@@ -1226,7 +1226,7 @@ the local/CI/community transport, not a silent production fallback.
 
 | # | Task | Acceptance |
 |---|---|---|
-| 8.39 `[D:8.3,8.14,8.17]` | **IN PROGRESS.** `MatchmakingState` now projects queue → proposal → allocation/process-ready/assignment-ready/connect/live plus terminal failure states; autoload `ControlPlaneClient` provides authenticated queue create/recovery/heartbeat/cancel and proposal response requests with idempotency/revision headers; `matchmaking.tscn`/`matchmaking.gd` expose the state and authoritative actions from the main menu; authenticated proposal recovery now reconciles missed proposal events and expires them at read time; the Go API publishes targeted authenticated revisioned queue/proposal/assignment events and the Godot client consumes state/proposal/assignment events, fetching the authoritative assignment after assignment readiness; the API can now use the durable participant-scoped proposal provider | `test_matchmaking_state.gd`, `test_control_plane_client.gd`, `test_matchmaking_ui.gd`, `TestProposalRecoveryIsParticipantScopedAndExpiresAtReadBoundary`, `TestAuthenticatedWebSocketDeliversOnlyTargetedRevisionedEvents`, `TestStateChangingAPIActionsPublishTargetedEvents` and `TestProposalRecoveryUsesDurableBackendAndRemainsParticipantScoped` reject stale/gapped/conflicting updates, validate endpoint/token/payload normalization, preserve idempotent duplicates, and guarantee visible phase/terminal copy; wiring durable allocation events into the outbox, wait/latency explanations and Godot runtime verification remain |
+| 8.39 `[D:8.3,8.14,8.17]` | **IN PROGRESS.** `MatchmakingState` now projects queue → proposal → allocation/process-ready/assignment-ready/connect/live plus terminal failure states; autoload `ControlPlaneClient` provides authenticated queue create/recovery/heartbeat/cancel and proposal response requests with idempotency/revision headers; `matchmaking.tscn`/`matchmaking.gd` expose the state and authoritative actions from the main menu; authenticated proposal recovery now reconciles missed proposal events and expires them at read time; the Go API publishes targeted authenticated revisioned queue/proposal/assignment events and the Godot client consumes state/proposal/assignment events, fetching the authoritative assignment after assignment readiness; the API can now use the durable participant-scoped proposal provider for both recovery and accept/decline mutations | `test_matchmaking_state.gd`, `test_control_plane_client.gd`, `test_matchmaking_ui.gd`, `TestProposalRecoveryIsParticipantScopedAndExpiresAtReadBoundary`, `TestAuthenticatedWebSocketDeliversOnlyTargetedRevisionedEvents`, `TestStateChangingAPIActionsPublishTargetedEvents`, `TestProposalRecoveryUsesDurableBackendAndRemainsParticipantScoped` and store proposal SQL tests reject stale/gapped/conflicting updates, validate endpoint/token/payload normalization, preserve idempotent duplicates, and guarantee visible phase/terminal copy; wiring durable allocation events into the outbox, wait/latency explanations and Godot runtime verification remain |
 | 8.40 `[D:8.3,8.14]` | **IN PROGRESS.** Pure Go revisioned replica reducer rejects gaps for REST resync, makes duplicate/out-of-order events idempotent, and resumes from the authoritative snapshot revision; authenticated queue-ticket recovery now has an owner-checked REST read; Godot client projection persists non-secret ticket/proposal state, forces authoritative recovery after restart, and can replay a lost queue-create response with the original ticket/idempotency key; the Go API now exposes an authenticated `/v1/events` WebSocket with bounded per-player queues, strict upgrade/vocabulary validation and REST-resync-safe slow-client failure; Godot `ControlPlaneClient` can connect to the stream, consume validated events, automatically reconnect with bounded backoff, and trigger REST recovery on projection gaps and stream return | `server/domain/sync.go`, `server/api/events.go`, `server/api/service.go`, `service_test.go`, `matchmaking_state.gd` and `control_plane_client.gd` cover gap, snapshot, replay, same-revision conflict, owner-only ticket recovery, expired-ticket terminal handling, malformed restart snapshots, API-level duplicate-create replay/conflict, authenticated handshake/key rejection, targeted event delivery, exactly-once slow-subscriber closure and invalid-event rejection; durable outbox fan-out and live Godot verification remain |
 | 8.41 `[D:7.8,8.9,8.31,8.40]` | **IN PROGRESS.** Authenticated `GET /v1/assignments/{matchId}` now exposes only a validated, player-scoped assignment view; Godot `AssignmentState`/`ControlPlaneClient.fetch_assignment()` bind the response to the authenticated player, recheck expiry, preserve explicit transport/slot/join authorisation and do not connect before assignment-ready; migration 0002 and the Go store adapter now persist/recover the complete player-scoped assignment projection with conflict-safe identical replay; `AssignmentProviderFromStore` wires that durable projection into the API injection point | `server/api/service.go`, `store_adapters.go`, `service_test.go`, `assignment_state.gd`, `test_assignment_state.gd`, `server/migrations/0002_assignments.sql` and `server/store/assignment_sql.go` cover participant/identity/expiry/shape/transport boundaries, player-scoped schema keys, expiry-filtered reads, assignment upsert conflict handling and nil-store fail-closed behavior; signed manifest-to-player persistence wiring, SDR relay-ticket installation, `hello` join-authorisation wiring, fencing integration and live Godot/PostgreSQL verification remain |
 | 8.42 `[D:8.22,8.23,8.24,8.40]` | **IN PROGRESS.** `RankedProfileState` and `ControlPlaneClient.fetch_ranked_profile()` expose the backend-authoritative rating/RD/volatility/games/tier/provisional/season view; matchmaking UI displays provisional/tier status without client-side rating math | `test_control_plane_client.gd` validates profile shape, numeric safety and provisional display; ranked profile fetch/display, committed revision after reconnect, abandon status and season countdown remain dependent on live auth/backend events and Godot runtime verification |

server/api/service.go

@@ -45,6 +45,9 @@ type SessionIssuer interface {
 type ProposalBackend interface {
 	Get(context.Context, string, string, time.Time) (domain.Proposal, error)
 }
+type ProposalMutationBackend interface {
+	Respond(context.Context, string, string, string, bool, uint64, time.Time) (domain.Proposal, error)
+}
 
 type AssignmentView struct {
 	MatchID           string    `json:"match_id"`
@@ -451,6 +454,16 @@ func (s *Service) proposalMutation(w http.ResponseWriter, r *http.Request) {
 		writeError(w, http.StatusBadRequest, "invalid_revision")
 		return
 	}
+	now := s.now()
+	var updated domain.Proposal
+	if s.ProposalBackend != nil {
+		mutator, supportsMutation := s.ProposalBackend.(ProposalMutationBackend)
+		if !supportsMutation {
+			writeError(w, http.StatusServiceUnavailable, "proposal_unavailable")
+			return
+		}
+		updated, err = mutator.Respond(r.Context(), playerID, parts[0], key, parts[1] == "accept", revision, now)
+	} else {
 		s.proposalMu.Lock()
 		defer s.proposalMu.Unlock()
 		proposal, exists := s.Proposals[parts[0]]
@@ -458,8 +471,8 @@ func (s *Service) proposalMutation(w http.ResponseWriter, r *http.Request) {
 			writeError(w, http.StatusNotFound, "not_found")
 			return
 		}
-	now := s.now()
+		updated, err = proposal.Respond(playerID, key, parts[1] == "accept", revision, now)
-	updated, err := proposal.Respond(playerID, key, parts[1] == "accept", revision, now)
+	}
 	if err != nil {
 		writeDomainError(w, err)
 		return

server/api/service_test.go

@@ -24,6 +24,16 @@ type sessionBackendSpy struct{ calls int }
 type proposalBackendSpy struct {
 	proposal  domain.Proposal
 	calls     int
+	mutations int
+}
+
+func (b *proposalBackendSpy) Respond(_ context.Context, playerID, _ string, key string, accept bool, revision uint64, now time.Time) (domain.Proposal, error) {
+	b.mutations++
+	updated, err := b.proposal.Respond(playerID, key, accept, revision, now)
+	if err == nil {
+		b.proposal = updated
+	}
+	return updated, err
 }
 
 func (b *proposalBackendSpy) Get(_ context.Context, playerID, _ string, _ time.Time) (domain.Proposal, error) {
@@ -389,6 +399,21 @@ func TestProposalRecoveryUsesDurableBackendAndRemainsParticipantScoped(t *testin
 	if status := get(participantSession, participantToken); status != http.StatusOK {
 		t.Fatalf("participant recovery status = %d", status)
 	}
+	respond, err := http.NewRequest(http.MethodPost, server.URL+"/v1/proposals/"+proposal.ProposalID+"/accept", nil)
+	if err != nil {
+		t.Fatal(err)
+	}
+	respond.Header.Set("Authorization", "Bearer "+participantSession.SessionID+":"+participantToken)
+	respond.Header.Set("Idempotency-Key", "proposal-durable-response-123456")
+	respond.Header.Set("If-Match-Revision", "0")
+	response, err := server.Client().Do(respond)
+	if err != nil {
+		t.Fatal(err)
+	}
+	_ = response.Body.Close()
+	if response.StatusCode != http.StatusOK || backend.mutations != 1 {
+		t.Fatalf("durable response status = %d, mutations = %d", response.StatusCode, backend.mutations)
+	}
 	if status := get(outsiderSession, outsiderToken); status != http.StatusNotFound {
 		t.Fatalf("outsider recovery status = %d", status)
 	}

server/api/store_adapters.go

@@ -37,6 +37,10 @@ func (p postgresProposalBackend) Get(ctx context.Context, playerID, proposalID s
 	return store.GetProposal(ctx, p.db, playerID, proposalID, now)
 }
 
+func (p postgresProposalBackend) Respond(ctx context.Context, playerID, proposalID, idempotencyKey string, accept bool, expectedRevision uint64, now time.Time) (domain.Proposal, error) {
+	return store.RespondToProposal(ctx, p.db, playerID, proposalID, idempotencyKey, accept, expectedRevision, now)
+}
+
 func ProposalProviderFromStore(db *sql.DB) ProposalBackend {
 	return postgresProposalBackend{db: db}
 }

server/store/proposal_recovery_sql.go

@@ -1,8 +1,11 @@
 package store
 
 import (
+	"bytes"
 	"context"
+	"crypto/sha256"
 	"database/sql"
+	"encoding/json"
 	"fmt"
 	"time"
 
@@ -27,6 +30,50 @@ FROM proposal_participants
 WHERE proposal_id = $1
 ORDER BY player_id`
 
+const ProposalResponseIdempotencyScope = "proposal.respond"
+
+const ProposalResponseIdempotencyInsertSQL = `INSERT INTO idempotency_keys
+    (scope, idempotency_key, payload_digest, result)
+VALUES ($1, $2, $3, $4)
+ON CONFLICT (scope, idempotency_key) DO NOTHING`
+
+const ProposalResponseIdempotencySelectSQL = `SELECT payload_digest, result
+FROM idempotency_keys
+WHERE scope = $1 AND idempotency_key = $2
+FOR UPDATE`
+
+const ProposalLockSQL = `SELECT playlist, state, revision, expires_at
+FROM proposals
+WHERE proposal_id = $1
+FOR UPDATE`
+
+const ProposalParticipantLockSQL = `SELECT response
+FROM proposal_participants
+WHERE proposal_id = $1 AND player_id = $2
+FOR UPDATE`
+
+const ProposalParticipantRespondSQL = `UPDATE proposal_participants
+SET response = $3, responded_at = $4
+WHERE proposal_id = $1 AND player_id = $2 AND response = 'PENDING'`
+
+const ProposalCountPendingSQL = `SELECT COUNT(*)
+FROM proposal_participants
+WHERE proposal_id = $1 AND response = 'PENDING'`
+
+const ProposalAcceptSQL = `UPDATE proposals
+SET state = 'ACCEPTED', revision = revision + 1
+WHERE proposal_id = $1 AND state = 'OPEN'`
+
+const ProposalDeclineSQL = `UPDATE proposals
+SET state = 'DECLINED', revision = revision + 1
+WHERE proposal_id = $1 AND state = 'OPEN'`
+
+const ProposalRevisionBumpSQL = `UPDATE proposals
+SET revision = revision + 1
+WHERE proposal_id = $1 AND state = 'OPEN'`
+
+var ErrProposalResponseConflict = fmt.Errorf("proposal response conflict")
+
 // GetProposal recovers the full proposal only after proving the caller is a
 // participant. Expiry is advanced in the same transaction as the read so a
 // missed event cannot leave a durable proposal indefinitely OPEN.
@@ -75,3 +122,121 @@ func GetProposal(ctx context.Context, db *sql.DB, playerID, proposalID string, n
 	}
 	return proposal, nil
 }
+
+// RespondToProposal is the durable mutation counterpart to GetProposal. The
+// proposal row and participant row are locked in one transaction; the result
+// is stored under the idempotency key before the transaction commits.
+func RespondToProposal(ctx context.Context, db *sql.DB, playerID, proposalID, idempotencyKey string, accept bool, expectedRevision uint64, now time.Time) (domain.Proposal, error) {
+	if db == nil || playerID == "" || proposalID == "" || len(idempotencyKey) < 16 || len(idempotencyKey) > 128 || now.IsZero() {
+		return domain.Proposal{}, fmt.Errorf("invalid proposal response arguments")
+	}
+	digest := sha256.Sum256([]byte(fmt.Sprintf("%s|%s|%t|%d", playerID, proposalID, accept, expectedRevision)))
+	var proposal domain.Proposal
+	err := RunSerializable(ctx, db, DefaultSerializableAttempts, func(ctx context.Context, tx *sql.Tx) error {
+		result, err := tx.ExecContext(ctx, ProposalResponseIdempotencyInsertSQL, ProposalResponseIdempotencyScope, idempotencyKey, digest[:], []byte("{}"))
+		if err != nil {
+			return err
+		}
+		inserted, err := result.RowsAffected()
+		if err != nil {
+			return err
+		}
+		if inserted == 0 {
+			var priorDigest, priorResult []byte
+			if err := tx.QueryRowContext(ctx, ProposalResponseIdempotencySelectSQL, ProposalResponseIdempotencyScope, idempotencyKey).Scan(&priorDigest, &priorResult); err != nil {
+				return err
+			}
+			if !bytes.Equal(priorDigest, digest[:]) {
+				return ErrProposalResponseConflict
+			}
+			if err := json.Unmarshal(priorResult, &proposal); err != nil {
+				return fmt.Errorf("invalid stored proposal response: %w", err)
+			}
+			return nil
+		}
+
+		var playlist, state string
+		var revision uint64
+		var expiresAt time.Time
+		if err := tx.QueryRowContext(ctx, ProposalLockSQL, proposalID).Scan(&playlist, &state, &revision, &expiresAt); err != nil {
+			return err
+		}
+		if state != string(domain.Open) || !now.Before(expiresAt) {
+			return domain.ErrProposalClosed
+		}
+		if revision != expectedRevision {
+			return domain.ErrStaleRevision
+		}
+		var response string
+		if err := tx.QueryRowContext(ctx, ProposalParticipantLockSQL, proposalID, playerID).Scan(&response); err != nil {
+			return domain.ErrNotParticipant
+		}
+		if response != string(domain.Pending) {
+			return domain.ErrConflict
+		}
+		response = string(domain.DeclinedResponse)
+		if accept {
+			response = string(domain.AcceptedResponse)
+		}
+		changed, err := tx.ExecContext(ctx, ProposalParticipantRespondSQL, proposalID, playerID, response, now)
+		if err != nil {
+			return err
+		}
+		if count, err := changed.RowsAffected(); err != nil || count != 1 {
+			return ErrProposalResponseConflict
+		}
+		targetState := string(domain.Declined)
+		if accept {
+			var pending int
+			if err := tx.QueryRowContext(ctx, ProposalCountPendingSQL, proposalID).Scan(&pending); err != nil {
+				return err
+			}
+			if pending == 0 {
+				targetState = string(domain.Accepted)
+			} else {
+				targetState = state
+			}
+		}
+		if targetState != state {
+			if targetState == string(domain.Accepted) {
+				_, err = tx.ExecContext(ctx, ProposalAcceptSQL, proposalID)
+			} else {
+				_, err = tx.ExecContext(ctx, ProposalDeclineSQL, proposalID)
+			}
+			if err != nil {
+				return err
+			}
+			revision++
+		} else {
+			if _, err := tx.ExecContext(ctx, ProposalRevisionBumpSQL, proposalID); err != nil {
+				return err
+			}
+			revision++
+		}
+		proposal = domain.Proposal{ProposalID: proposalID, Playlist: domain.Playlist(playlist), State: domain.State(targetState), Revision: revision, ExpiresAt: expiresAt}
+		rows, err := tx.QueryContext(ctx, ProposalParticipantsSelectSQL, proposalID)
+		if err != nil {
+			return err
+		}
+		for rows.Next() {
+			var participant domain.ProposalParticipant
+			if err := rows.Scan(&participant.PlayerID, &participant.Response); err != nil {
+				rows.Close()
+				return err
+			}
+			proposal.Participants = append(proposal.Participants, participant)
+		}
+		if err := rows.Err(); err != nil {
+			rows.Close()
+			return err
+		}
+		rows.Close()
+		stored, err := json.Marshal(proposal)
+		if err != nil {
+			return err
+		}
+		_, err = tx.ExecContext(ctx, `UPDATE idempotency_keys SET result = $3 WHERE scope = $1 AND idempotency_key = $2`, ProposalResponseIdempotencyScope, idempotencyKey, stored)
+		return err
+	})
+	return proposal, err
+}

server/store/proposal_recovery_sql_test.go

@@ -11,6 +11,11 @@ func TestProposalRecoverySQLBindsParticipantAndExpiresAtReadBoundary(t *testing.
 		ProposalParticipantExpireSQL:         {"response = 'PENDING'", "response = 'TIMED_OUT'"},
 		ProposalRecoverySelectSQL:            {"proposal_id = $1", "player_id = $2", "EXISTS"},
 		ProposalParticipantsSelectSQL:        {"proposal_id = $1", "ORDER BY player_id"},
+		ProposalResponseIdempotencyInsertSQL: {"ON CONFLICT (scope, idempotency_key) DO NOTHING", "payload_digest"},
+		ProposalLockSQL:                      {"proposal_id = $1", "FOR UPDATE"},
+		ProposalParticipantLockSQL:           {"proposal_id = $1", "player_id = $2", "FOR UPDATE"},
+		ProposalParticipantRespondSQL:        {"response = 'PENDING'", "responded_at"},
+		ProposalRevisionBumpSQL:              {"revision = revision + 1", "state = 'OPEN'"},
 	} {
 		for _, fragment := range fragments {
 			if !contains(query, fragment) {