acdream/docs/architecture/code-structure.md

acdream — code structure & extraction sequence

Status: Living document. Created 2026-05-16; implementation reconciliation completed 2026-07-21; Slice 1 landed the same day. This is the active structural program before new M4 subsystems enter the App layer. Purpose: Describe the desired structural state of the App layer, explain the rules we've adopted, and lay out the safe extraction sequence from today's reality (one 15,723-line GameWindow.cs at the 2026-07-21 audit) to the target (thin GameWindow, small focused collaborators). Companion to: acdream-architecture.md (the layered architecture) and worldbuilder-inventory.md (what we take from WB vs port ourselves).

---

1. The structural problem we're solving

The layered architecture works: AcDream.Core is GL-free, the network layer is wire-compatible, the UI has a stable contract, plugins load. The structural debt is concentrated in one file:

baseline cf50ee3d                         15,723 lines / 278 fields / 205 methods
after Slice 1                             14,912 lines / 278 fields / 191 methods

GameWindow is the single object that:

• Owns the GL context, the window, input, and shaders.
• Reads ~40 different environment variables across its lifetime.
• Still owns live-session bootstrap, character entry, roughly 440 lines of event subscription, and most App-side packet handlers even though LiveSessionController owns the socket/session shell.
• Owns the adapters that hydrate canonical LiveEntityRuntime records into animation, collision, rendering, and DAT-backed appearance resources.
• Composes the shipped WorldSelectionQuery and SelectionInteractionController; it no longer owns world-picking, selection intent, Use/PickUp, or auto-walk deferral algorithms.
• Drives per-frame render orchestration (sky → terrain → opaque mesh → transparent mesh → particles → debug lines → UI).
• Builds and applies streamed landblock presentation (DAT decode, scenery, EnvCells, mesh publication, collision, and retirement) around the shipped StreamingController and GpuWorldState owners.
• Wires up every plugin hook sink, every diagnostic, every panel.

The extracted controllers are valuable and tested, but line count alone proves that creating collaborators has not yet made the window thin. A collaborator is a completed extraction only when it owns the state and behavior body; wrapping a GameWindow method in a delegate preserves ordering but remains a partial extraction.

The fix is not "rewrite GameWindow in one pass" — that is a high-risk change. The fix is to extract one collaborator at a time, verify behavior is unchanged, ship, and move on. This document defines that sequence.

---

2. Code Structure Rules — the discipline

Recap of the rules from CLAUDE.md with the rationale:

Rule 1: No new substantial feature bodies in GameWindow.cs

Why: Every line we add to GameWindow makes the eventual decomposition harder. New features that "live in" GameWindow instead of being extracted are the reason the file is 10k lines.

How to apply: A new feature gets its own class under src/AcDream.App/<Subsystem>/ (or deeper in AcDream.Core if it's pure logic). GameWindow owns a field and a wiring call, nothing more. If you find yourself adding a 200-line method to GameWindow, stop and extract.

Exemption: Trivial wiring that must stay in GameWindow because it touches GL state during OnLoad is acceptable, but should still delegate to a collaborator for the substance.

Rule 2: AcDream.Core must not depend on window / GL / backend projects

Why: Core is the GL-free, testable layer. The moment Core imports a GL or windowing namespace, we've lost the ability to test it without a graphics context, and the layer split becomes fiction.

How to apply: Phase O removed both external WorldBuilder/backend project references. The only currently allowed seams are the GL-free helpers owned in our tree under src/AcDream.Core/Rendering/Wb/: TerrainUtils, TerrainEntry, RegionInfo, SceneryHelpers, and TextureHelpers. ObjectMeshManager and every GL resource owner remain in App. If Core needs a new capability, define a narrow Core interface and implement it in App; adding a new project reference requires an inventory-doc update explaining why.

Rule 3: UI panels target AcDream.UI.Abstractions only

Why: This is the one rule that keeps D.2b (the future retail-look backend) viable. Every panel that imports ImGuiNET directly is a panel we'd have to rewrite when the backend swaps.

How to apply: A panel's using block must mention AcDream.UI.Abstractions.* and nothing from AcDream.UI.ImGui. The panel writes against IPanelRenderer. The ImGuiPanelRenderer translates those calls to ImGui at runtime. Plugin-facing UI follows the same rule.

Rule 4: Startup env vars enter through RuntimeOptions

Why: Environment variables are global mutable state. Reading them at random call sites means (a) duplicated Environment.GetEnvironmentVariable boilerplate, (b) no single place to see "what flags does the client respond to?", (c) impossible to unit-test parsing.

How to apply: src/AcDream.App/RuntimeOptions.cs is the typed options object. Program.cs builds it once from args + env and passes it to GameWindow. New startup flags add a field to RuntimeOptions and a parser in RuntimeOptions.FromEnvironment. They don't add Environment.GetEnvironmentVariable reads.

Scope: RuntimeOptions is for startup-time configuration — things that don't change once the window is up. Runtime diagnostic toggles are Rule 5's domain.

Rule 5: Runtime diagnostic toggles live in diagnostic owner classes

Why: Diagnostic flags (ACDREAM_DUMP_MOTION, ACDREAM_PROBE_*, etc.) need to be both env-readable at startup and runtime-toggleable from the DebugPanel. Per-call-site env reads can't be runtime-toggled.

How to apply: Today's template is src/AcDream.Core/Physics/PhysicsDiagnostics.cs — one static class with typed Probe* properties read from env vars once at startup, plus runtime setters that the DebugPanel binds. New diagnostic flags follow this shape, not the per-call-site pattern that dominates GameWindow.cs.

Cleanup direction: The dozens of existing ACDREAM_DUMP_* reads inside GameWindow.cs are tech debt. We do NOT bulk-migrate them as part of this refactor — they're working, they're scattered, and moving them carries risk without a current acceptor. We migrate them opportunistically: when a GameWindow extraction lands and a diagnostic moves with it, route it through the new owner's diagnostic class.

Rule 6: Tests live in the project matching the layer

Why: Test discoverability + dependency hygiene. A test for a Core class belongs next to other Core tests; a test for an App class belongs in an App test project. Co-locating tests across layers makes the dependency graph dishonest.

How to apply: One test project per source project that has tests. Today:

• tests/AcDream.Core.Tests/ ← src/AcDream.Core/
• tests/AcDream.Core.Net.Tests/ ← src/AcDream.Core.Net/
• tests/AcDream.UI.Abstractions.Tests/ ← src/AcDream.UI.Abstractions/
• tests/AcDream.App.Tests/ ← src/AcDream.App/

tests/AcDream.App.Tests/ now exists and owns App-layer controller, streaming, render-resource lifetime, retained-UI, and RuntimeOptions tests. New App tests belong there; do not place GL-free Core behavior in that project merely because App currently wires it.

---

3. Target structure of the App layer

The end state — not what we're shipping in one pass, but the shape we're aiming at.

src/AcDream.App/
├── Program.cs                          # parse args + env → RuntimeOptions, build GameWindow
├── RuntimeOptions.cs                   # typed startup options (Rule 4)
├── Rendering/
│   ├── GameWindow.cs                   # thin: GL/window lifecycle + delegates per-frame to RenderFrameOrchestrator
│   ├── RenderFrameOrchestrator.cs      # per-frame draw order (sky → terrain → opaque → trans → particles → debug → UI)
│   ├── LiveEntityAnimationScheduler.cs # shipped: ordinary live-object update workset
│   ├── LiveEntityAnimationPresenter.cs # final part-pose/mesh/effect composition after scheduler output
│   ├── RetailStaticAnimatingObjectScheduler.cs # shipped: separate static-animation workset
│   ├── StaticLiveRootCommitter.cs       # live static root → pose + collision boundary
│   ├── TerrainModernRenderer.cs        # (already exists)
│   ├── TextureCache.cs                 # (already exists)
│   ├── ParticleRenderer.cs             # (already exists)
│   ├── Sky/                            # (already exists)
│   ├── Wb/                             # WB seam + EnvCellLandblockBuild transaction
│   └── Vfx/                            # (already exists)
├── Net/
│   ├── LiveSessionController.cs        # owns complete WorldSession connect/enter/logout/reconnect lifecycle
│   └── LiveSessionEventRouter.cs       # typed subscriptions → focused domain handlers
├── Physics/
│   ├── ProjectileController.cs          # canonical live-record projectile orchestration
│   ├── RemotePhysicsUpdater.cs          # ordinary/Hidden remote narrow-tick integration
│   ├── LiveEntityOrdinaryPhysicsUpdater.cs # manager-less canonical body Transition path
│   ├── LiveEntityNetworkUpdateController.cs # Position/Vector/State/Motion App integration
│   ├── LiveEntityShadowPublisher.cs     # authoritative exact-owner/residency collision gate
│   ├── RemoteInboundMotionDispatcher.cs # shared animated/headless UpdateMotion funnel
│   ├── RemoteTeleportController.cs      # loaded/pending teleport placement ownership
│   ├── RemoteTeleportHook.cs            # ordered retail teleport teardown actions
│   └── RemoteTeleportPlacement.cs       # collision-seated SetPosition transition commit
├── World/
│   ├── InboundPhysicsStateController.cs # timestamps + accepted spawn snapshots
│   ├── LiveEntityRuntime.cs             # shipped: logical lifetime + ServerGuid↔entity.Id translation
│   ├── LiveEntityHydrationController.cs # spawn/appearance/parent/delete resource integration
│   ├── RetailInboundEventDispatcher.cs  # update-thread packet/frame FIFO barrier
│   ├── RetailLiveFrameCoordinator.cs    # shipped: object/network/command/reconcile phase order
│   ├── LiveEntityPresentationController.cs # ordered Hidden/NoDraw/effect/collision side effects
│   ├── LiveEntityTeardown.cs            # failure-isolated multi-owner lifecycle drain
│   └── ParentAttachmentState.cs         # parent generations + pending ParentEvent relations
├── Interaction/
│   ├── WorldSelectionQuery.cs           # read-only picking/classification/description queries
│   └── SelectionInteractionController.cs # owns selection intents, Use/PickUp, auto-walk deferral
├── Streaming/
│   ├── LandblockPresentationPipeline.cs # DAT build/apply/retire transaction and resource publication
│   └── ...                              # shipped streamer/world-state/reveal owners
├── Input/                              # (already exists)
├── Audio/                              # (already exists)
└── Plugins/                            # (already exists)

What GameWindow keeps:

• IWindow / GL / IInputContext lifecycle (constructor + OnLoad + Run + OnClosing).
• RuntimeOptions reference (the typed startup config).
• GL resource construction and top-level collaborator composition. Construction is allowed here; feature algorithms and mutable subsystem state are not.
• One field per top-level collaborator (_liveSessionController, _liveEntityRuntime, _selectionInteraction, _streamingPresentation, _liveObjectFrame, _renderFrameOrchestrator).
• The Silk.NET event-handler stubs that delegate to collaborators.

What GameWindow loses:

• Live connect/enter/logout and event-subscription bodies → completed inside LiveSessionController / LiveSessionEventRouter.
• Live-object hydration adapters and final animated-part presentation → focused world/render owners over canonical LiveEntityRuntime records.
• WorldPicker, target queries, and selection-driven Use/PickUp/auto-walk → WorldSelectionQuery + SelectionInteractionController. Core SelectionState remains the injected session owner.
• Landblock DAT build/apply/retirement presentation → LandblockPresentationPipeline; StreamingController remains the residency scheduler and GpuWorldState remains the spatial registry.
• Per-frame draw orchestration and its frame-local scratch state → RenderFrameOrchestrator.

The eventual GameEntity aggregation (target state described in acdream-architecture.md §"GameEntity: The Unified Entity") happens after LiveEntityRuntime is the single owner of entity state. Until then, the parallel-dicts problem is bounded inside one class instead of spread across GameWindow.

LiveEntityRuntime is now that single boundary. It composes InboundPhysicsStateController for the nine-channel retail timestamp gates and latest accepted immutable CreateObject snapshot, owns the canonical local ID and optional runtime components, and separates logical registration from spatial projection. It also retains raw PhysicsState separately from the final state produced by retail's ordered side effects; LiveEntityPresentationController projects those transitions into draw, collision, effect, child-NoDraw, and target visibility without becoming a second lifetime or GUID owner. ParentAttachmentState is runtime-owned and keys unresolved relations by child and parent generation. Rendering/Vfx/EntityEffectController owns the focused mixed F754/F755 pending FIFO, effect profiles, typed-table resolution, and a readiness set; canonical ServerGuid-to-local-ID translation always stays in LiveEntityRuntime. EntityScriptActivator uses the same canonical WorldEntity.Id as rendering and physics; the disjoint static ID allocators fail fast instead of wrapping into another landblock's namespace. All other non-Parent packet families still need the future general queue tracked by divergence AD-32.

The per-frame object scheduler is extracted, but final animated-part presentation is not. LiveEntityAnimationScheduler snapshots canonical spatial root records and advances the incarnation-stable object clock, PartArray, hooks, one selected movement owner, and manager tail in retail order. The remaining GameWindow.TickAnimations loop still composes part transforms, drawable MeshRefs, and effect poses from the scheduler output; moving that body into LiveEntityAnimationPresenter is an explicit pending slice. Manager-less bodies delegate their candidate/Transition/cell/shadow commit to LiveEntityOrdinaryPhysicsUpdater; retained projectile bodies and remote MovementManagers remain mutually exclusive movement owners. Static animation is deliberately separate: RetailStaticAnimatingObjectScheduler owns the CPhysics::static_animating_objects workset for DAT and live PhysicsState- Static owners with Setup DefaultAnimation. Both schedulers are wired by GameWindow, but neither owns GUID identity or logical resources. The typed animation view fills its reusable snapshot and render-ID set from LiveEntityRuntime's concrete spatial dictionary, avoiding interface-enumerator boxing on both update and render hot paths. RemoteInboundMotionDispatcher similarly keeps UpdateMotion protocol behavior outside GameWindow: GameWindow resolves the canonical record/body and the optional PartArray sink, while one dispatcher owns retail's interrupt, style, MoveTo/type-0, sticky, and standing-long-jump order. Static root projection is bounded by StaticLiveRootCommitter, which synchronizes changed roots to effects and collision without rebuilding zero-omega shadows or resurrecting a Hidden/withdrawn registration.

Synchronous network and lifecycle callbacks are bounded by RetailInboundEventDispatcher. It owns no wire state and no identity; it only serializes nested live-object operations until the current packet or full object-frame tail completes. State-bearing direct dispatch is allocation-free; only a genuinely nested operation allocates its retained queue wrapper. LiveEntityRecord then supplies exact-incarnation and per-channel authority versions at callback boundaries. Position, State, Vector, and Movement remain independent, while a separate velocity version invalidates only an older operation that would overwrite a newer velocity installed by Position, Vector, or Movement. This prevents re-entrant App observers from creating call-stack ordering that retail's update-thread packet FIFO cannot produce.

Pose-dependent hook deferral is similarly incarnation-scoped rather than GUID- or local-ID-scoped. EntityEffectPoseRegistry publishes a monotonic pose-owner lifetime, and AnimationHookFrameQueue captures it before semantic callbacks and rechecks it before each semantic AnimationDone and each routed hook. Static animation retains process_hooks until its root, live parts, and children are published; withdrawal invalidates both its prepared pose and pending hook tail.

Resolved ordinary motion commits its body/root/contact state before the canonical full-cell setter enters LiveEntityRuntime.RebucketLiveEntity. Because that setter may synchronously move the projection to pending or replace the GUID, both RemotePhysicsUpdater and LiveEntityOrdinaryPhysicsUpdater revalidate the exact incarnation before collision or manager-tail publication. Collision residency itself is projection-owned, not updater-owned: LiveEntityPresentationController suspends retained non-projectile shadows on every unavailable-projection edge, restores them on hydration, and reconciles the pending-first case at OnLiveEntityReady after collision registration. Local projection and both authoritative remote UpdatePosition tails commit the complete root before rebucketing, then publish collision only through an exact-record/spatial-residency gate. Remote reflood tracks translation, sign-invariant complete orientation, and cell changes so an in-place turn or a same-pose EnvCell crossing cannot leave offset Setup shapes stale. The projectile controller retains its separate body/InWorld/shadow edge owner.

Remote teleport placement is bounded in Physics/RemoteTeleportController, not GameWindow: it retains at most one pending request per materialized incarnation, scopes it by the live generation and accepted PositionSequence, and asks RemoteTeleportPlacement to collision-seat the current body when the destination projection is available. GameWindow supplies lifecycle and shadow-sync callbacks only; canonical identity remains in LiveEntityRuntime. Failed hydration restores the captured source and delegates an incarnation-scoped shadow restore to LiveEntityPresentationController while that source is unloaded, so Hidden/UnHide and teleport never become competing restore owners. A newer placement transfers that restore into an explicit generation-scoped active-placement state before its rebucket visibility edge even while Hidden. All intervening Hidden/UnHide and projection edges defer to that owner until stable success or rollback completes; only then can it restore, re-defer the source, or hand a Hidden result back for UnHide. The ILiveEntityRemotePlacementRuntime seam keeps the complete cell/contact handoff available across same-body runtime-wrapper replacement; replacing the canonical body or dropping the placement contract within one incarnation is rejected even after an operational component clear. RemoteMotion.Body is constructor-owned; hydration compares pending/current wrappers directly to the record body rather than trusting wrapper-to-wrapper equality. Binding reads an interface Body getter once and reuses that snapshot. GpuWorldState performs remove+place as one spatial rebucket, then commits and serially drains visibility edges; LiveEntityRuntime filters delayed duplicates. A rollback inside an observer cannot race the outer destination-visible notification or expose an intermediate false pulse. LiveEntityTeardown executes those independent owner callbacks to completion and aggregates failures afterwards, so a throwing effect/plugin sink cannot strand teleport, movement, shadow, light, or GUID-scoped state.

---

4. Reconciled ownership ledger — 2026-07-21

This section replaces the original six-step sketch. That sketch correctly identified the target but overstated some extractions and omitted large bodies added during M2/M3. The current audit is based on implementation ownership, not whether a class with the planned name exists.

4.1 What counts as extracted

An extraction is complete only when:

1. the collaborator owns the mutable state and behavior body;
2. GameWindow constructs it and delegates through a narrow method;
3. the collaborator does not call back into arbitrary GameWindow methods;
4. focused tests exercise the collaborator without constructing a Silk window;
5. the accepted connected/visual behavior is unchanged.

A class that stores delegates back to substantial GameWindow methods is a useful ordering seam, but its ownership status is partial.

4.2 Current implementation truth

Area	Status	Current truth
Startup options	Complete	RuntimeOptions owns startup configuration (eda936dc). Remaining direct environment reads are legacy runtime diagnostics, not startup configuration.
Network session	Partial	LiveSessionController owns endpoint resolution, WorldSession construction, ticking, and disposal (0b25df53). TryStartLiveSession, character choice/entry, session clearing, about 440 lines of subscriptions, and most App packet handlers remain in GameWindow.
Live identity/lifetime	Complete core owner	LiveEntityRuntime owns incarnation identity, accepted snapshots/timestamps, runtime components, logical/spatial lifetime, and teardown. GameWindow still owns large hydration and wire-to-presentation adapter bodies.
Inbound/object-frame order	Partial	RetailInboundEventDispatcher, RetailLiveFrameCoordinator, LiveEntityAnimationScheduler, static scheduler, remote/projectile updaters, and teleport owners are shipped. AdvanceLiveObjectRuntimeCore, final part/mesh/effect composition, and several integration tails remain in GameWindow.
World reveal	Complete	WorldRevealCoordinator owns login/portal readiness and reveal lifetime (a4ef5788). The accepted deterministic lifecycle trace did not change after extraction.
Retained gameplay UI	Mostly complete feature ownership	RetailUiRuntime and focused panel/controllers own layout and behavior. GameWindow.OnLoad still performs substantial service composition, which is allowed until the final composition cleanup.
Selection/interaction	Complete	WorldSelectionQuery owns read-only picking/classification/range queries; SelectionInteractionController owns selection intent, Use/PickUp transport, exact-incarnation queues, and auto-walk deferral; ItemInteractionController owns ItemHolder policy plus the shared retail inventory-request transaction. GameWindow retains construction and narrow lifecycle forwarding only.
Landblock presentation	Not extracted	Residency scheduling is extracted, but DAT build, scenery/EnvCell construction, apply, collision/resource publication, and retirement glue remain a large GameWindow body.
Render-frame orchestration	Not extracted	OnRender plus its portal/PView/alpha/particle helpers still own the draw graph and frame-local scratch state. There is no RenderFrameOrchestrator yet.
Unified GameEntity	Deferred target	LiveEntityRuntime already supplies the critical canonical owner. Full type aggregation is a separate high-risk migration and is not required to make GameWindow thin.

4.3 Revised extraction sequence

Every numbered slice is behavior-preserving and independently committed. A slice does not include feature work or opportunistic gameplay fixes.

Gate 0 — deterministic baseline — COMPLETE

The Release suite, connected R6 soak, world-lifecycle screenshots/checkpoints, graceful reconnect, local locomotion/collision/projectile/teleport gate, and two-client portal observer gate form the pre-refactor baseline. Each later slice runs the subset capable of detecting its risk; render/session slices run the complete connected lifecycle gate.

Slice 1 — finish selection/interaction ownership — COMPLETE 2026-07-21

Detailed execution plan: docs/plans/2026-07-21-gamewindow-slice-1-selection-interaction.md.

Split the old Step 4 into three reviewable commits:

1. WorldSelectionQuery receives camera/scene/live-object read seams and owns picking, target classification, selection bounds, closest-target lookup, names, and descriptions. It cannot mutate selection or send packets.
2. SelectionInteractionController owns selection intents, double-click Use, Use/PickUp packet requests, range decisions, speculative facing, and the pending auto-walk action. It composes the existing item/combat controllers rather than duplicating their rules.
3. Selection/Use/PickUp/combat-target InputAction cases delegate to the new controller; the old methods and fields are deleted from GameWindow.

Tests cover read-only query classification separately from stateful intents, including direct use, distant auto-walk completion, corpse/container opening, pickup placement, hostile-only targeting, Hidden objects, and same-GUID reuse. The cutover landed in 047a4c83, 52dbb574, e74f2ca9, fa8d5232, d2bb5af4, and review-correction commit 5acc3f01. The final review also ported the shared ACCWeenieObject::prevRequest transaction shape across inventory, toolbar, paperdoll, external-container, give, split, merge, and drop routes. Queued actions, pending placements, and responses are bound to exact object incarnations/tokens; optimistic projection, rollback projection, and authoritative response notices are distinct.

Measured against cf50ee3d, GameWindow fell from 15,723 to 14,912 lines and from 205 to 191 methods; field count stayed at 278 because the slice exchanged legacy state for composed owners. Three independent retail, architecture, and adversarial review loops finished clean. Release build passed; 6,558 tests passed and five fixture/conformance tests skipped intentionally.

Slice 2 — finish live animation presentation — NEXT

Move TickAnimations, final rigid/visual part composition, effect-pose publication, and motion-done binding into LiveEntityAnimationPresenter. LiveEntityAnimationScheduler remains the time/movement owner; the presenter only consumes one scheduler result and publishes the final draw/effect pose. Move the surviving motion diagnostics with this owner instead of adding more environment reads to GameWindow.

Tests pin legacy and sequencer paths, Hidden/static eligibility, ObjScale versus rigid effect poses, part availability, same-frame hook pose publication, and incarnation replacement. Run the R6 object-frame and projectile/effect suites plus the connected locomotion/projectile gate.

Slice 3 — complete live-session ownership

Expand LiveSessionController to own connect, character-list validation, character selection, EnterWorld, graceful character logout, reconnect, and session-scoped disposal. Add LiveSessionEventRouter for typed subscriptions; it routes events to focused domain handlers and owns unsubscription, but owns no entity or UI state.

Move TryStartLiveSession, ClearInboundEntityState, and WireLiveSessionEvents out of GameWindow. Do not create a giant context bag or a second world-state owner. The connected gate must prove login, fresh world reveal, command/chat, portal, exact 0xF653 logout, ACE endpoint disconnect, and fresh-process reconnect.

Slice 4 — extract live-entity App integration

This is two owners, not one replacement god object:

• LiveEntityHydrationController owns CreateObject/ObjDesc/Parent/Pickup/Delete integration, DAT-backed appearance hydration, resource registration, and exact teardown callbacks over LiveEntityRuntime.
• LiveEntityNetworkUpdateController owns App routing for accepted Position/Vector/State/Movement updates into the existing remote physics, motion, projectile, presentation, and teleport controllers.

Neither may own a GUID dictionary. Both resolve the current incarnation only through LiveEntityRuntime and must preserve the inbound FIFO/authority-token rules. Run the complete live-entity stress suite, R6 connected route, inventory equip/parenting, death/corpse, and portal gates.

Slice 5 — extract landblock presentation

Create LandblockPresentationPipeline around the existing immutable LandblockBuild transaction. It owns DAT build context, scenery/EnvCell construction, mesh/collision/light publication, demotion/full-retirement, and the single-reader DAT lock contract. StreamingController continues deciding what is resident; GpuWorldState continues owning spatial buckets.

Tests pin loaded/pending/demoted/unloaded symmetry, stale build generations, resource pin/release balance, collision footprints across landblock seams, and first-login bootstrap replacement. Run the deterministic world-lifecycle gate and the seven-destination resource soak.

Slice 6 — extract update-frame orchestration

After the stateful bodies above have owners, make the update path a real orchestrator instead of delegates back into the window. It owns the fixed retail phase order: input/local object → ordinary/static objects → hooks/ particles/scripts → inbound network → command interpreter → non-advancing spatial reconcile → streaming/UI updates. GameWindow.OnUpdate becomes a short time/input handoff.

Frame-order tests and the existing R6 gate must produce the same lifecycle and movement traces before and after extraction.

Slice 7 — extract RenderFrameOrchestrator

Move the complete draw graph and its reusable frame-local scratch state into a GL-owning App collaborator. Preserve the exact modern pipeline order, clip routing, PView flood, landscape/opaque/shared-alpha flush boundaries, particles, debug draw, paperdoll, retained UI, and frame fences. Do not pass a hundred individual delegates or let the orchestrator reach back into GameWindow; inject a small immutable service set plus explicit per-frame input.

Automated acceptance uses framebuffer artifacts and render/resource checkpoints. Visual acceptance compares outdoor, building, dungeon, portal exit, translucent lifestone/particles, and UI at the same camera positions.

Slice 8 — composition and shutdown cleanup

Keep GL/window construction in GameWindow.OnLoad, but group creation into small composition functions and delete feature state left behind by prior slices. OnClosing delegates to the existing retryable shutdown transaction. Silk callbacks become narrow calls into the input, update, render, resize, focus, and shutdown owners.

4.4 Exit criteria

The campaign is complete when:

• GameWindow contains no AC gameplay algorithm, entity scan, packet builder, DAT landblock builder, animation-part composer, or draw-graph body;
• OnUpdate, OnRender, OnInputAction, and live-session callbacks are short delegation methods;
• every extracted owner has App-layer tests and symmetric teardown;
• the Release suite and connected lifecycle/resource gates stay green after every slice;
• the final local visual matrix passes unchanged.

Line count is a progress signal, not the acceptance test. The expected result is below roughly 5,000 lines, but ownership and dependency direction decide completion. Full GameEntity type aggregation is evaluated only afterwards as a separate migration; it is not folded into this campaign.

---

5. Rules of the road during the extraction

1. One slice at a time. Each commit ships one ownership boundary or mechanical call-site cutover. Bundling slices makes failures hard to isolate.
2. Behavior preservation is the acceptance criterion. Every slice must build clean, all tests pass, and visual verification at the appropriate accepted milestone scenarios must succeed. We're moving code, not changing it.
3. No new features during an extraction step. If you spot a real bug while extracting, file it in docs/ISSUES.md and address it in a separate commit (before or after the extraction, not folded into it).
4. Diagnostic toggle migrations are opportunistic. When a method moves to a new owner, the diagnostic flag inside it can move to a diagnostic class as part of the same commit. We do not do a bulk diagnostic-cleanup pass.
5. Update this document when the plan changes. If a slice turns out to need a different ownership shape than described above, update §4 in the same session you discover the divergence.
6. No façade-only completion claims. A delegate from a new class back to a substantial GameWindow method is a useful intermediate seam, not a completed extraction.
7. No duplicate ownership. New collaborators query LiveEntityRuntime, GpuWorldState, SelectionState, or the relevant existing owner. They do not create replacement GUID, visibility, session, or resource maps.

---

6. What this document is not

• Not a full rewrite plan. The point is the opposite — small steps, verified at each boundary.
• Not a feature phase. Following the 2026-07-21 user decision, this behavior-preserving campaign is the structural prerequisite before new M4 subsystem bodies are added. Severe regressions remain fixable in separate commits; ordinary feature work waits.
• Not a substitute for the milestones / roadmap. Those drive the feature work. This drives the structural work that runs underneath.