Implements server-authoritative per-word occlusion for NPC conversations.
ConversationEventBuffer drains into ObserverSnapshot each tick so the client
receives only words audible from the player's position. Updates test fixtures
to include the new conversation_events and conversation_ended fields.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Add 3 zone enrichment tests to observer pipeline (tile inside zone,
tile outside zone, absent ZoneMap resource). Fix misleading Decay
Observatory doc comment that claimed Direct start when test starts at
KnowsDetails. Add production population TBD note on ZoneMap.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Server-tracked zone_id on VisibleTile for D-073 zone crossfade and
D-059 deep fog temperature tint. ZoneMap resource backed by BTreeMap,
observer enrichment in snapshot assembly. Backwards-compatible: v10
payloads deserialize with zone_id: None.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Three new gauntlet test rooms for Sprint 13, each targeting a specific
system under test. StableIds are additive-only (66–74).
Sound Lab (66-68): Three NPCs at calibrated tile distances (2, 6, 12
tiles) verify D-018 three-range sound model. Tests confirm Close ≤3,
Medium ≤8, Long ≤20 tile thresholds via SoundEvent::audible_at().
Decay Observatory (69): Single NPC observed in LOS then broken. Tests
confirm D-041 confidence decay: KnowsDetails → KnowsOf → Suspects
(floor), and Stale state when stale_after threshold is exceeded.
Shift Change (70-71): Two NPCs with DailyRoutine components. Test
confirms D-031 phase-boundary check_phase_transition issues PathRequest
only to NPCs not already at their target location.
Sprint 13 reset plates (72-74) wired per existing pattern.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
validate_movement now inserts SoundEventEmitter with Footstep events on
every successful move. Intensity scales by stance: Sprint 0.8, Walk 0.5,
Careful 0.3, Crouch 0.15. Range is Close (3 tiles) for all stances.
This completes the sound event pipeline end-to-end: movement produces
events → collect_sound_events drains to queue → observer snapshot
includes audible events → client bridge receives them.
Addresses Tyre critical review item #1 on PR #42.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
1. Sound producers: document empty v0.1 pipeline explicitly (critical)
2. Routine tests: add ActiveSim to 3 tests that passed trivially
3. Rename _observer_pos → observer_pos (used at line 191)
4. Add FactionOnly positive test case (matching faction_id)
5. Fix stale doc comment "Current: 9" → 10 in ObserverSnapshot
6. Remove orphaned SimulationTier/LastInteraction/ScopeTag types
7. Add tracing::warn on FactionOnly non-numeric parse failure
8. Document Medium-range occlusion gap as TODO in audible_at
9. Insert SoundEventQueue in observer test setup_world
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Review fixes: deduplicate dump_schedule_graph doc comment,
add rng_seed round-trip test and v9→v10 backward compat test.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Tracing (#344):
- Add 'json' feature to tracing-subscriber dependency
- Emit JSON log format when CI=true or RUST_LOG_FORMAT=json is set
(structured log ingestion in CI pipelines)
- Add tracing::debug! with tick_ms/budget_ms/over_budget fields on each
tick for performance profiling and tier system debugging prerequisite
Schedule dump (#346):
- Add --dump-schedule CLI flag that prints bevy_ecs schedule graph and exits
without requiring TCP bridge or world setup
- Add make debug-schedule target for CI artifact generation and diff-based
regression detection of unintended system reordering
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Adds rng_seed: Option<u64> to ObserverSnapshot. The WRONG button (#507) captures
inputs.jsonl and seed.txt for replay, but seed.txt was writing "unavailable"
because the server did not include the RNG seed in ObserverSnapshot.
Changes:
- bridge/types.rs: PROTOCOL_VERSION 9→10, rng_seed field with serde(default,
skip_serializing_if = "Option::is_none") for backward compatibility
- perception/observer/mod.rs: inject Res<SimRng> into compute_observer_snapshot,
populate rng_seed: Some(rng.seed()) each tick
- All test files: add rng_seed: None to ObserverSnapshot constructors
- tests/serialization.rs: bump protocol_version_constant assertion 9→10
- Regenerate msgpack fixtures and golden file for protocol v10
Completes the WRONG button capture loop: replays can now fully reproduce
observed bugs with the exact RNG seed from the capture.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
#94 — Active tier simulation (complete):
- Add ActiveSim marker to all 9 test world room NPC spawns
- Fix test entities in routine.rs and path_follow.rs to include ActiveSim
so With<ActiveSim> queries match correctly in unit tests
#99 — Tier transition logic (complete):
- Implement update_tier_markers system in tier.rs
- Promotes/demotes tier markers by manhattan distance from PlayerCharacter:
≤40 tiles → ActiveSim, ≤120 → BackgroundSim, beyond → StateSaved
- Handles cross-z-level as u32::MAX (effectively unreachable)
- No-op when no PlayerCharacter entity present (headless tests safe)
- 11 new unit tests covering all distance bands and boundary cases
- TierPlugin now registers the system after movement::validate_movement
Also picks up extended test coverage added by hoshe:
- observer/tests.rs — 230 lines of perception observer tests
- sound.rs — additional sound event integration tests
All 548 tests pass.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Add rng_seed: Option<u64> to ObserverSnapshot (protocol v10). Populated
from SimRng state each tick. Completes the WRONG button capture loop —
seed.txt now writes a valid u64 instead of "unavailable", enabling
deterministic replay from bug reports.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Add clippy::disallowed_types for std::collections::HashMap scoped to
the simulation crate. Replace HashMap with BTreeMap in movement.rs for
deterministic iteration order. Allow exception in perception/query.rs
where iteration order is irrelevant (per-frame scratch buffer).
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Implements 4 completed tickets + partial progress on 2 more:
- #93 Tier marker components (ActiveSim, BackgroundSim, StateSaved + TierPlugin)
- #138 Information tag schema (ObserverAccess enum in knowledge/types.rs)
- #124 Sound event system (SoundEventEmitter, SoundEventQueue, bridge wiring)
- #193 Line previewer CLI (line_preview binary with filter/explain/sequence modes)
- #94 Active tier simulation (in progress — With<ActiveSim> filters)
- #139 Component-level access control (in progress — filter_by_access)
Updates snapshot fixtures and test golden files for new sound_events field.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Warnings fixed:
- contraband.rs: scan event now always emits even when NPC already
knows (was skipped by early `continue`). Contract matches doc.
- test_world/mod.rs: ScanEventBuffer added to player spawn bundle
so check_contraband_scan doesn't silently no-op in gauntlet mode.
- npc/mod.rs → simulation/mod.rs: moved check_contraband_scan
registration to SimulationPlugin (operates on player inventory and
snapshot pipeline, consistent with process_talk_interaction).
Suggestions addressed:
- cross_room_transitions.rs T1: clarified standalone position vs
constants.rs observer position in comment.
- dialogue.rs: Vec<&str> dedup replaced with BTreeSet<&str> for
deterministic iteration (project convention).
- contraband.rs: added test for multiple simultaneous ScanAuthority
NPCs in range (564 tests total).
- dialogue.rs: doc-comment on relationship_to_trust explaining
KnowledgeConfidence ordering and Suspects default.
- cross_room_transitions.rs T5: noted direct KG API usage vs full
perception system.
- sprint_gauntlet.rs: documented intentional Contentment { level: 0 }.
- content_scaling.rs: noted GAUNTLET_NPC_COUNT is manually maintained.
- contraband.rs: doc-comment on cross-plugin registration rationale.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Cross-room transition tests (server/tests/cross_room_transitions.rs):
- T1: sprint suppresses interaction buffer, restores on Walk (D-055)
- T2: CarriedBy survives room transition — no TilePosition leak (D-065)
- T3: pause mid-corridor discards movement, Unpause resumes (D-031)
- T4: KnowledgeGraph persists across player position change (D-041)
- T5: entity knowledge downgrades Direct→KnowsDetails on LOS exit (D-060)
- T6: eavesdrop cut immediately on first movement out of corner (D-071)
- T7: confrontation verb disappears on retreat beyond MID_RANGE=5 (D-057/D-070)
- T8: Sprint blocks eavesdrop accumulation, Careful enables it (D-055+D-071)
All 8 tests pass. Test suite grows from 545 → 563 (18 tests added across sprint).
Tests use direct ECS World + Schedule pattern; T3 uses full App + SimulationPlugin.
Test suite expansion:
- content_scaling.rs: max_npc_pack_behavioral_regression + stress tests (#513)
- golden/proof_room_tick_10.json: updated golden file for gauntlet world changes
- golden_suite.rs, serialization.rs, bridge_ipc.rs, bridge_tcp.rs: adapted to
new world entity count and wire types
- gen_fixtures.rs, perf_bench.rs, content_runtime.rs: minor test adaptations
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Adds three new test rooms to the gauntlet layout at coordinates that tile
correctly with the existing Sprint 11 world. All rooms follow the canonical
entity-registration pattern (StableId 55-62 assigned in spawn order).
- sprint_gauntlet.rs: 32×22 room (StableId 55-57) — Npc_pacing, Npc_guard,
Readable sign. Validates sprint suppression (D-055) at close range.
- eavesdrop_alcove.rs: 24×16 room (StableId 58-60) — two NPC speakers,
corner Readable marker. Validates eavesdrop positioning (D-071).
- confrontation_stage.rs: 32×24 room (StableId 61-62) — Npc_target and
peripheral passer-by NPC. Validates confrontation verb range (D-070).
Updated test_world: mod.rs registers all new rooms, constants.rs adds GAUNTLET
region constant and room spawn points, reset.rs clears all gauntlet rooms.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
- Document non-blocking receive contract in perf_bench.rs docstring,
confirming no TCP deadlock race (Hoshe #1, critical)
- Make shadowcast parser order-independent — flush on new config header
instead of requiring Recursive after Symmetric (Hoshe #2)
- Fix p95 calculation: use floor(0.95*(N-1)) nearest-rank instead of
ceil(0.95*N)-1 which was off-by-one at N=50 (Hoshe #4)
- Error on --compare when no baseline file exists (Hoshe #5)
- Add D-031 10tps assumption comment to TICK_BUDGET_US (Tyre #1)
- Strengthen snapshot assertion: require warmup + half measurement
window instead of warmup + 1 (Tyre #2)
- Regenerate baseline with corrected p95 (356µs, was 526µs)
Hoshe #3 (.PHONY) was already addressed — perf-baseline is in the
.PHONY declaration on Makefile line 10.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Adds `make perf-baseline` — boots the full server plugin stack with
real content, measures 50 ticks (5 warmup), captures per-tick timing,
entity counts, and process RSS. Includes shadowcast benchmarks. Saves
structured JSON to tests/perf/baseline.json for regression detection.
Supports --compare mode (>20% threshold).
First baseline: mean 366µs, p95 526µs (0.5% of D-026 100ms budget).
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Technical project descriptor with game architecture, simulation
parameters, and content model. Version field (0.1.9) is now the
canonical version — Cargo.toml synced, CLAUDE.md and start-sprint
skill updated to reference it during release workflow.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
- Fail on encode errors instead of silently writing empty .msgpack files
- Fail test on missing/empty fixture dir instead of silent skip
- Add all missing action variants (MoveSouth, MoveEast, MoveWest,
Unpause, ToggleStanceDown, WalkAway) to GDScript fixture generator
- Add GDScript fixture staleness check to make pre-pr
- Validate repo root detection before writing outside client/
- Add file.flush() before close in headless mode
- Document fixture failure recovery in DEVOPS.md
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
format_snapshot_text(&ObserverSnapshot) -> String pub-exported from
server crate. Entity labels as kind:entity_id sorted by distance,
room name stubbed as (unknown) until Gauntlet constants land.
10 unit tests. Consumed by tooling/test-client for --text mode.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Implement full dialogue selection pipeline (D-028): 4-layer filtering
engine with access tier, situation derivation, trust tier, and weighted
topic+mood scoring via SimRng. Add ContentSlug component for stable
content identity across save/load. Add walk-away KG recording with
IncompleteInteraction events per D-064 three-phase consequences. Bump
protocol to v8 with DialogueResponseEvent. Fixes#305, #427, #452.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Add AnomalyMarker component and detect_anomalies() system that flags
entities with KG relationship PersonOfInterest or Contradicted state
for urgent cognitive delay (0.3s vs 0.6s normal). Add
trigger_recognition_monologue() that fires monologue at delay START
(when grey blob appears), not at completion — the monologue IS the
recognition process per D-060. Includes v0.1 fallback recognition
lines and cooldown tracking. Fixes#450, #451.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Add CLI argument parsing for test infrastructure: --test-mode enables
deterministic seed (42) and warn-level tracing to stderr, --port allows
OS-assigned ports (--port 0), --seed overrides RNG seed. Prints
LISTENING:{port} to stdout after bind for test harness discovery.
Extracts setup_proof_room() for reuse. Fixes#459.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Replace HashSet with BTreeSet for visible_ids, sort visible_tiles by
coordinates, sort visible entities in snapshot by entity_id, and sort
movers by Entity bits in validate_movement. Required by D-010 principle 4
(deterministic simulation). Fixes#456, #457, #458.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
process_player_input processed movement commands regardless of tick
rate, causing the camera to drift from the player when the game was
paused and the player held direction keys. Add PlayerAction::is_movement()
helper and a paused guard that skips movement actions while TickRate is
Paused. Pause/Unpause commands still process normally.
Fixes bug #3 (player moves while game is paused).
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
read_framed() called read_exact() which blocked the entire bevy Update
schedule waiting for client input — no systems ran until a keystroke
arrived. Switch TcpStream to non-blocking mode so receive_inputs()
returns Ok(vec![]) on WouldBlock instead of blocking. Toggle to
blocking for snapshot writes (reliable delivery). Add 50ms frame
throttle (~20 ticks/sec) since the non-blocking loop would otherwise
spin. Downgrade input receive logging to trace, add error logging for
missing bridge resource and failed observer queries.
Fixes bug #1 (server never sends snapshots) and #2 (camera doesn't
center until first keystroke).
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Critical fixes:
- Add PendingRecognitionWire serialization roundtrip test
- Check Option return from delay.cancel() before logging
- InputQueue capacity limit (1000) with drop-oldest and warning
- TODO in observation.rs references ticket #450
Warning fixes:
- Hot-reload guards against invalid/empty content root
- Location header mismatch warning in line pool indexing
- walk_yaml() depth limit (100) against symlink loops
- Consecutive reload failure counter (warns after 5+)
Test additions:
- Negative prerequisite filtering test for monologue lines
- Integration test for pending_recognitions in observer snapshot
- Eavesdrop threshold ordering assertion (Careful < default)
Documentation:
- Playtesting expectation comments on delay constants
- is_pending() scalability note for future NPC cognitive delay
- ID format regex validation in line pool spec
- BTreeMap vs sort() ordering clarification in loader
- Multiplayer TODO in relationships.rs references D-010
- ContentSlug ticket #452 filed for entity slug resolution
394 tests, 0 failures.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
