ship(O): Phase O — DatPath Unification — SHIPPED

ONE thing touches the DATs. WB code lives in our repo:
- src/AcDream.Core/Rendering/Wb/ — pure helpers (5 files, ~782 LOC)
- src/AcDream.App/Rendering/Wb/ — GL infra + mesh pipeline (~27 files, ~7K LOC)

Project references to WorldBuilder.Shared + Chorizite.OpenGLSDLBackend
dropped from AcDream.App.csproj and AcDream.Core.csproj.
references/WorldBuilder/ remains in-tree as read-reference only.

DefaultDatReaderWriter eliminated; DatCollection is the only dat reader.
WbMeshAdapter consumes our DatCollection via DatCollectionAdapter
(O-D7 fallback adapter; ObjectMeshManager has 26 _dats.X call sites,
exceeding the 20 refactor threshold).

Visual side-by-side passed: Holtburg town, inn interior, dungeon all
render identically to pre-O.

Doc updates:
- CLAUDE.md: rewrote WB integration cribs to point at extracted code.
  Code Structure Rules rule 2 updated to remove stale seam names.
  "Currently working toward" flipped from Phase O to M1.5 resumption.
- docs/architecture/worldbuilder-inventory.md: Phase O banner added.
  Status/integration model updated to post-O ownership. Workflow
  section updated to reference our extracted tree, not WB project ref.
- docs/plans/2026-04-11-roadmap.md: Phase O moved to shipped table.
  Phase O "ahead" block collapsed to SHIPPED note. M1.5 block updated
  to ACTIVE (Phase O shipped; resuming from 2026-05-20 baseline).
- docs/plans/2026-05-12-milestones.md: M1.5 heading updated to ACTIVE;
  Phase O ship writeup prepended to the M1.5 block.

Phase O ship closes Tasks O-T1..O-T7 shipped across this session.
Specs + audit + plan: docs/superpowers/{specs,plans}/2026-05-21-phase-o-*

Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>

docs/architecture/worldbuilder-inventory.md

@@ -1,35 +1,55 @@
-# WorldBuilder Inventory — what we take, adapt, or leave
+# WorldBuilder Inventory — what we extracted, adapted, or left behind
 
-**Status:** load-bearing reference. As of 2026-05-08 acdream's strategy is
-to **rely heavily on WorldBuilder** for rendering and dat-handling rather
-than re-port retail algorithms ourselves. WorldBuilder is MIT-licensed, is
-verified by visual inspection to render the AC world correctly (terrain,
-scenery, slabs, dungeons, slopes, particles), and uses the same Silk.NET
-+ .NET stack we already target.
+> **Phase O shipped 2026-05-21.** The ~33 WB files we actually use have
+> been extracted into our tree. `references/WorldBuilder/` stays as a
+> **read-reference only** — nothing in `src/AcDream.*` references it as a
+> project dependency. `DatCollection` is now the only dat reader in process.
+>
+> Use this document to:
+> 1. Know **where our extracted code lives** (look for the "Extracted to"
+>    column / notes in each section below).
+> 2. Know **what WB still has** that we haven't needed yet — grep
+>    `references/WorldBuilder/` if you ever need to add something.
+> 3. Know **what WB never had** (the 🔴 list) — those are always ours.
 
-**Integration model:** **fork upstream WorldBuilder** at
-`github.com/Chorizite/WorldBuilder`, depend on our fork, delete editor-only
-code, expose hooks for our network state to feed scene data in. Sync with
-upstream via merge so we inherit fixes. This document tells you, before
-you write code, whether the thing you're about to port already exists in
-WorldBuilder.
+**Pre-O status (archived for context):** As of Phase N.4 (2026-05-08)
+acdream relied heavily on WorldBuilder as a project reference for rendering
+and dat-handling. WorldBuilder is MIT-licensed, verified by visual inspection
+to render the AC world correctly (terrain, scenery, slabs, dungeons, slopes,
+particles), and uses the same Silk.NET + .NET stack we target.
 
-**Workflow change:** Before re-implementing any AC-specific rendering or
-dat-handling algorithm, **check this inventory first**. If WorldBuilder
-has it, port from WorldBuilder (or call into our fork once it's wired
-up), not from retail decomp. Retail decomp remains the oracle for things
-WorldBuilder lacks — animation, motion, physics collision, networking.
+**Post-O integration model:** Extracted WB code lives in two locations in
+our tree (see CLAUDE.md for the full breakdown):
+- `src/AcDream.Core/Rendering/Wb/` — pure helpers (no GL): `TerrainUtils`,
+  `TerrainEntry`, `RegionInfo`, `SceneryHelpers`, `TextureHelpers`.
+- `src/AcDream.App/Rendering/Wb/` — GL infrastructure + mesh pipeline:
+  `ObjectMeshManager`, `WbMeshAdapter`, `WbDrawDispatcher`, texture cache,
+  shader infra, EnvCell/portal/scenery/terrain-blending pipeline classes.
+
+`DatCollectionAdapter` bridges our `IDatCollection` to the `IDatReaderWriter`
+interface WB's internals expect (O-D7 fallback; `ObjectMeshManager` has 26
+internal `_dats.*` call sites — above the 20-site inline-swap threshold).
+
+**Workflow:** Before re-implementing any AC-specific rendering or dat-handling
+algorithm, **check this inventory first**. If we already extracted it (🟢
+sections), it's in `src/AcDream.App/Rendering/Wb/` — use our copy. If WB has
+it but we haven't extracted it yet, grep `references/WorldBuilder/` and extract
+as needed. Retail decomp remains the oracle for things WB never had (🔴 list).
+
+Attribution: WorldBuilder is MIT-licensed. `NOTICE.md` includes WB attribution.
 
 ---
 
-## Repo layout (as of cloned snapshot under `references/WorldBuilder/`)
+## Read-reference layout (under `references/WorldBuilder/`, not project-referenced)
 
-- **`Chorizite.OpenGLSDLBackend/`** — full OpenGL renderer (Silk.NET).
+- **`Chorizite.OpenGLSDLBackend/`** — full OpenGL renderer (Silk.NET). The
+  components we use are extracted into `src/AcDream.App/Rendering/Wb/`.
 - **`WorldBuilder.Shared/`** — data models, dat parsers, landscape module.
-- **`WorldBuilder/`** — Avalonia desktop app shell (NOT taken).
-- **`WorldBuilder.{Windows,Linux,Mac}/`** — platform entry points (NOT taken).
-- **`WorldBuilder.Server/`** — collab editing backend (NOT taken).
-- **`Tests/` + `WorldBuilder.Shared.Benchmarks/`** — test harness (study).
+  The helpers we use are extracted into `src/AcDream.Core/Rendering/Wb/`.
+- **`WorldBuilder/`** — Avalonia desktop app shell (not taken).
+- **`WorldBuilder.{Windows,Linux,Mac}/`** — platform entry points (not taken).
+- **`WorldBuilder.Server/`** — collab editing backend (not taken).
+- **`Tests/` + `WorldBuilder.Shared.Benchmarks/`** — test harness (study only).
 
 **Upstream NuGet dependencies** (these stay as NuGet packages, we don't
 vendor them):
@@ -234,17 +254,28 @@ WorldBuilder is a dat editor; it does not have:
 
 ---
 
-## What this means for the workflow
+## What this means for the workflow (post-Phase O)
 
 The CLAUDE.md "grep named → decompile → verify → port" workflow stays
 the rule for everything in the 🔴 list (network, physics, animation,
-movement, UI, plugin, audio, chat). For anything in 🟢, the new rule is:
-**check this inventory FIRST**. If WB has it, port from WB. Re-porting
-from retail decomp when WB already has a tested port is no longer
-appropriate — that's how we got the scenery edge-vertex bug.
+movement, UI, plugin, audio, chat).
 
-When the inventory says "take wholesale or adapt" and we discover a
-behavior mismatch with retail (rare — WB is verified), the resolution
-is: reconcile WB ↔ retail decomp ↔ holtburger ↔ ACE ↔ ACViewer (the
-existing reference hierarchy in CLAUDE.md). WorldBuilder ranks at the
-top of that hierarchy for anything 🟢.
+For anything in 🟢 that we've already extracted: **the code is in our
+tree at `src/AcDream.{Core,App}/Rendering/Wb/`**. Read it there — don't
+grep `references/WorldBuilder/` unless you want to compare against the
+original. Re-porting from retail decomp when we already have a tested
+port is still how we'd get the scenery edge-vertex bug back.
+
+For anything in 🟢 that we have NOT yet extracted: grep
+`references/WorldBuilder/` to find the source, then extract it using the
+Phase O pattern (verbatim copy → adapt constructor to accept
+`IDatCollection` via `DatCollectionAdapter` where needed → add to
+`src/AcDream.App/Rendering/Wb/`). Do NOT add a new project reference back
+to `WorldBuilder.Shared` or `Chorizite.OpenGLSDLBackend` — Phase O
+permanently removed those.
+
+When we discover a behavior mismatch with retail (rare — the extracted
+code is the same as the original), the resolution is: reconcile extracted
+code ↔ retail decomp ↔ holtburger ↔ ACE ↔ ACViewer (the existing
+reference hierarchy in CLAUDE.md). Our extracted code ranks at the top
+of that hierarchy for anything 🟢.

docs/plans/2026-04-11-roadmap.md

@@ -74,6 +74,7 @@
 | Indoor walking Phase 1 — BSP cluster (partial) | 2026-05-19. Probe + WorldPicker cell-BSP occlusion (#86 closed) + CellId promotion via AABB containment (partial #84 fix). Seven commits across 5 phases: `18a2e28` plan, `27d7de1` Phase A `[indoor-bsp]` probe + toggle, `3764867` Phase B CellBspRayOccluder in WorldPicker, `4e308d5` Phase B screen-rect tests, `c19d6fb` Phase D AABB containment + L.2e bare-low-byte fix, `fda6af7` Phase E `[cell-cache]` diagnostic, `1f11ba9` Phase E extended AABB/bsphere/poly-count fields. **#86 closed** (picker occlusion). **#84 partially closed** (spawn-in-building stuck-above-floor resolved; threshold/doorway walls remain open under #87). **#85 open** (wall pass-through root cause confirmed as same as #84 remaining symptom — CellId doesn't stay promoted during outdoor→indoor walking). **#87 filed** (portal-based indoor cell tracking — retail-faithful follow-up). `[indoor-bsp]` + `[cell-cache]` probes stay in place as scaffolding for the follow-up phase. Handoff: [`docs/research/2026-05-19-cluster-a-shipped-handoff.md`](../research/2026-05-19-cluster-a-shipped-handoff.md). Plan: [`docs/superpowers/plans/2026-05-19-indoor-walking-phase1-bsp-cluster.md`](../superpowers/plans/2026-05-19-indoor-walking-phase1-bsp-cluster.md). | Tests ✓ |
 | Indoor walking Phase 2 — Portal-based cell tracking | 2026-05-19. Portal-graph traversal replaces Phase D's AABB containment. Six commits: `1969c55` CellBSP+Portals wired into CellPhysics; `aad6976` CellTransit.FindCellList + FindTransitCellsSphere + AddAllOutsideCells + ResolveCellId rename; `069534a` BuildingPhysics + CheckBuildingTransit for outdoor→indoor entry; `702b30a` code-review polish; `3ffe1e4` pass foot-sphere center to ResolveCellId (critical fix — was passing CheckPos instead of GlobalSphere[0].Origin, causing PointInsideCellBsp to return false at floor level); `eb0f772` TryFindIndoorWalkablePlane synthesizes walkable plane from cell floor poly so the resolver doesn't fall through to outdoor SampleTerrainWalkable. **Closes #87, #85, and the wall-pass-through portion of #84 (fully closes #84).** Files #88 (indoor static object vibration — pre-existing) and #89 (BSPQuery.SphereIntersectsCellBsp — approximation in CheckBuildingTransit). `[cell-transit]`, `[indoor-bsp]`, `[check-bldg]`, `[cell-cache]` probes stay in place. Handoff: [`docs/research/2026-05-19-indoor-walking-phase2-shipped-handoff.md`](../research/2026-05-19-indoor-walking-phase2-shipped-handoff.md). | Live ✓ |
 | Indoor walking Phase A4 — Multi-cell BSP iteration | 2026-05-20. Ports retail's `CTransition::check_other_cells` (`acclient_2013_pseudo_c.txt:272717-272798`). After the primary cell's BSP returns OK, every other cell the foot-sphere overlaps is queried via `BSPQuery.FindCollisions`. Halt on first Collided/Adjusted/Slid; Slid clears the contact-plane fields. Three commits land the slices: `e6369e2` `CellTransit.FindCellSet` overload (refactor `FindCellList` to expose the candidate set); `493c5e5` `Transition.CheckOtherCells` + `ApplyOtherCellResult` combine helper; `691493e` (orig `967d065`, then `3add110` revert, then this reapply) wires `CheckOtherCells` into `FindEnvCollisions`. 10 new unit tests; 1139 + 8 baseline maintained. **Visual verification surfaced a separate, pre-existing M2 blocker:** at the Holtburg inn doorway the CellId ping-pongs between outdoor `0xA9B40022` and indoor vestibule `0xA9B40164` rapidly because indoor BSP push-back exits the indoor CellBSP volume → ResolveCellId reclassifies as outdoor → wall checks bypassed on outdoor ticks. Bug reproduces fully with A4 reverted (`launch-revert2.log`), confirming A4 is not the cause. A4 is correct and tested but **dormant in practice** until the ping-pong is fixed. Handoff: [`docs/research/2026-05-20-phase-a4-shipped-cell-pingpong-finding.md`](../research/2026-05-20-phase-a4-shipped-cell-pingpong-finding.md). Spec: [`docs/superpowers/specs/2026-05-20-phase-a4-multi-cell-bsp-design.md`](../superpowers/specs/2026-05-20-phase-a4-multi-cell-bsp-design.md). Plan: [`docs/superpowers/plans/2026-05-20-phase-a4-multi-cell-bsp.md`](../superpowers/plans/2026-05-20-phase-a4-multi-cell-bsp.md). | Live ✓ (dormant pending cell-tracking fix) |
+| Phase O — DatPath Unification | 2026-05-21. ONE thing touches the DATs. Extracted ~33 WB files / ~7.7K LOC into `src/AcDream.{Core,App}/Rendering/Wb/`. Dropped project references to `WorldBuilder.Shared` + `Chorizite.OpenGLSDLBackend` from `AcDream.App.csproj` and `AcDream.Core.csproj`. `DefaultDatReaderWriter` eliminated; `DatCollection` is the only dat reader. `WbMeshAdapter` consumes it via `DatCollectionAdapter` (O-D7 fallback; `ObjectMeshManager` has 26 internal `_dats.*` call sites exceeding the 20-site inline-swap threshold). `references/WorldBuilder/` stays in-tree as read-reference. **Visual side-by-side passed**: Holtburg town, inn interior, dungeon all render identically to pre-O. `NOTICE.md` includes WB MIT attribution. Spec: [`docs/superpowers/specs/2026-05-21-phase-o-dat-path-unification-design.md`](../superpowers/specs/2026-05-21-phase-o-dat-path-unification-design.md). Audit: [`docs/superpowers/specs/2026-05-21-phase-o-t1-wb-audit.md`](../superpowers/specs/2026-05-21-phase-o-t1-wb-audit.md). Plan: [`docs/superpowers/plans/2026-05-21-phase-o-plan.md`](../superpowers/plans/2026-05-21-phase-o-plan.md). | Live ✓ |
 
 Plus polish that doesn't get its own phase number:
 - FlyCamera default speed lowered + Shift-to-boost
@@ -85,48 +86,17 @@ Plus polish that doesn't get its own phase number:
 
 ## Phases ahead — agreed order
 
-### Phase O — DatPath Unification (ACTIVE — pre-empts M1.5 by user direction 2026-05-21)
+### Phase O — DatPath Unification — SHIPPED 2026-05-21
 
 **Tagline:** ONE thing touches the DATs.
 
-**Filed:** 2026-05-21. **Status:** active. **Spec:** [`docs/superpowers/specs/2026-05-21-phase-o-dat-path-unification-design.md`](../superpowers/specs/2026-05-21-phase-o-dat-path-unification-design.md).
+**Filed:** 2026-05-21. **Status:** SHIPPED. **Spec:** [`docs/superpowers/specs/2026-05-21-phase-o-dat-path-unification-design.md`](../superpowers/specs/2026-05-21-phase-o-dat-path-unification-design.md).
 
-**Problem:** As of Phase N.4 (2026-05-08) we run TWO dat readers in
-process — our `DatCollection` and WorldBuilder's
-`DefaultDatReaderWriter` (constructed inside `WbMeshAdapter.cs:79`).
-Both open the same four `.dat` files independently. Costs:
-~50-100 MB duplicated index cache, double seeks on shared blocks, and
-the cross-check awkwardness in `WbMeshAdapter.cs:224-229`. AC community
-feedback flagged WB's dat-touching as "built for tools, not runtime";
-the double-reader is the most concrete manifestation.
+See the shipped-table entry above for the full summary. Phase O is
+complete; M1.5 (indoor walking, paused for Phase O) resumes from
+the 2026-05-20 baseline.
 
-**Approach:** Extract the WB pieces we actually use (mesh pipeline,
-texture decode, GL state, scenery, terrain blending, EnvCell/portal
-decode — roughly 3-5K LOC) into `src/AcDream.Core/Rendering/Wb/`.
-Verbatim copy first, no "improvements" while extracting. Adapt
-constructors to consume our `DatCollection` instead of
-`IDatReaderWriter`. Drop the two project references
-(`WorldBuilder.Shared` + `Chorizite.OpenGLSDLBackend`) at the end.
-WB is MIT-licensed; attribute in `NOTICE.md`. Keep `references/WorldBuilder/`
-in-tree for read-reference (not as a project dependency).
-
-**Why pre-empt M1.5:** User direction 2026-05-21 — we want this
-foundational decoupling done before further milestone work compounds
-the WB dependency. M1.5's Phase A6/A7 indoor work doesn't touch dat
-infrastructure, so it can resume cleanly after Phase O lands.
-
-**Tasks** (full breakdown in spec §5; ~7-8 working days):
-- **O-T1** — WB call-graph audit. Produce the exact "must extract" list.
-- **O-T2** — Create `src/AcDream.Core/Rendering/Wb/` + `NOTICE.md` attribution.
-- **O-T3** — Extract texture / GL infrastructure (no dat dependency, low risk).
-- **O-T4** — Extract mesh pipeline (`ObjectMeshManager` + types) + `DatCollection` swap.
-- **O-T5** — Extract scenery + terrain blending pipelines + `DatCollection` swap.
-- **O-T6** — Extract EnvCell + portal renderers + `DatCollection` swap.
-- **O-T7** — Drop the two project references. Drop `_wbDats` and cross-check code.
-- **O-T8** — Verification gate: three-scene visual side-by-side with main (Holtburg, inn, dungeon).
-- **O-T9** — Ship. Update CLAUDE.md + `worldbuilder-inventory.md` to reflect new ownership.
-
-**Acceptance** (full list in spec §6):
+**Acceptance met** (full list in spec §6):
 - `dotnet build` + `dotnet test` green; ~1147 test count maintained.
 - Zero `using WorldBuilder.*` or `Chorizite.OpenGLSDLBackend.*` in `src/AcDream.*`.
 - `DefaultDatReaderWriter` referenced in zero places in our source.
@@ -141,7 +111,7 @@ copy + swap to `DatCollection` only. Refactors in follow-up phases.
 **After Phase O ships:** M1.5 resumes from its 2026-05-20 baseline
 with no code changes lost — M1.5 doesn't touch WB-extracted territory.
 
-### Milestone M1.5 — "Indoor world feels right" (PAUSED for Phase O — see above)
+### Milestone M1.5 — "Indoor world feels right" (ACTIVE — Phase O shipped; resuming from 2026-05-20 baseline)
 
 The current top of the work order. Two phases (A6 + A7) inside one
 milestone. M2 ("kill a drudge") is deferred until M1.5 lands —

docs/plans/2026-05-12-milestones.md

@@ -185,7 +185,18 @@ close range and the player sees "You pick up the X." in chat.
 
 ---
 
-### M1.5 — "Indoor world feels right" — 🔵 NEXT (active, opened 2026-05-20)
+### M1.5 — "Indoor world feels right" — 🔵 ACTIVE (resumed 2026-05-21 after Phase O ship)
+
+**Phase O — DatPath Unification — shipped 2026-05-21.** ONE thing
+touches the DATs. ~33 WB files (~7.7K LOC) extracted into
+`src/AcDream.{Core,App}/Rendering/Wb/`; project references to
+`WorldBuilder.Shared` + `Chorizite.OpenGLSDLBackend` dropped;
+`DefaultDatReaderWriter` eliminated. Visual side-by-side passed at
+Holtburg town, inn interior, and dungeon. Phase O pre-empted M1.5
+per user direction 2026-05-21; M1.5 now resumes from the 2026-05-20
+baseline with no code changes lost — Phase O did not touch dat-loading
+infrastructure for physics or collision, only the rendering pipeline.
+M1.5's planned phases (A6 + A7) are unaffected.
 
 **Demo scenario:** Enter the Holtburg Sewer dungeon through the in-town entry
 portal. Navigate to the end (5–7 rooms with stairs + a multi-Z chamber).