[apotema]

Refs labelle-toolkit/labelle-engine#922 (labelle-toolkit/labelle-cli#528 WP2b). Draft. This is the inspector half; the startup and save-loader half is labelle-toolkit/labelle-engine#925, which is stacked on engine #921.

Why

In Flying Platform's ReleaseSafe WASM build, the debug plugin's drawEntityDetail (~165 KB) and drawEntityBrowser (~85 KB) were the 2nd- and 4th-largest functions. Both walk the whole component registry with inline for. That expands every component's code into the two functions: a runtime bufPrintZ of its name, membership checks, tag text, and the detail tree with every field's formatting.

Changes

• test(debug) (first commit, on the unchanged code): pins the exact widget transcript of both windows against a mock game and a call-recording Gui. It covers:

  • Position first, then registry order with fields;
  • the ? label for an over-long name;
  • filters and the dirty flag, and the tag column;
  • Select/Deselect and a dead entity;
  • the 50-row cap.

  The plugin's test build has no labelle-core, so the position type is resolved lazily through PositionOf(Game). Only the tests declare DebugInspectorPosition; real games still get core.Position.

• perf(debug):

  • Labels are now built at comptime, with the same ? fallback.
  • The browser's filter, membership and tag walks are runtime loops over comptime tables: labels, plus one noinline "has component" check per type.
  • Each component's detail tree is drawn by its own noinline drawComponentNode.

  The engine registry's names() returns a slice of a comptime var, so the names are copied into a constant first. The mock now reproduces that.

  The transcript tests pass unmodified, in Debug and ReleaseSafe.

Result

In FP, with engine#925, both functions are gone from the top list. They are now small enough to inline into drawGui, which is 26 KB. The before/after table for the whole of #922 is in labelle-toolkit/labelle-engine#925:

• game compile (edit rebuild): 131 → 99 s
• peak memory: 5.67 → 4.9 GiB
• game.wasm: −394 KB

Tests

cd plugins/debug && zig build test: 15/15 pass, in Debug and ReleaseSafe. On main it is 12/12; the 3 new tests are the inspector tests. Nothing outside plugins/debug changed, so I did not run the assembler's own suite.

Follow-up

Flying Platform vendors this plugin as plugins/debug (a byte-identical copy apart from line endings). It needs re-vendoring once this merges. My FP measurement clone used this version locally; nothing was committed to FP.

Not validated

The inspector was not exercised in a browser or on desktop. Behaviour is covered only by the transcript tests.

https://claude.ai/code/session_01Hpt1haxcboZPznpG1KqDRS

Summary by CodeRabbit

• Improvements
  • The inspector now uses a game-specific debug position type when available, with the standard position type as a fallback.
  • Component browsing and detail views handle long labels and display up to 50 rows.

[coderabbitai]

No actionable comments were generated in the recent review.

Recent review info

Run configuration

• Source mode: Auto
• Repository: Unavailable
• Branch: Unavailable
• File: No configuration file from this repository applied
• Commit: Unavailable
• Layers: Repository UI
• Review profile: CHILL
• Plan: Team
• Run ID: b5fc327c-b009-48d1-9d5f-30cbfaf7ad36

Commits

Reviewing files that changed from the base of the PR and between 5ce90db and a81adda.

Files selected for processing: 1

• plugins/debug/src/root.zig

Included review availability: This review used your included allowance. 1 included review remains after this review. Your included PR review attempts over the past 7 days set your current allowance at 10 reviews per hour.

Walkthrough

The inspector now resolves Position per game, iterates component labels and membership checks at runtime, and renders component details through a bounded helper. New tests cover browser filtering and selection, detail output, and display limits.

Changes: Debug Inspector

Layer / File(s)	Summary
Position resolution and component browser plugins/debug/src/root.zig	The browser uses the game’s DebugInspectorPosition when available and otherwise uses core.Position. It iterates component labels and membership checks at runtime.
Component detail rendering and tests plugins/debug/src/root.zig	The detail window uses the resolved Position type and renders each registered component through drawComponentNode. Tests cover detail output, browser filtering and selection, long labels, missing Position, stale selections, and the 50-row limit.

Estimated code review effort: 3 (Moderate) | ~20 minutes

Change: Refactor

Merge Risk: Minimal · up to a81ad

No actionable merge-blocking issue was established. The inspector change is mergeable after normal checks.

Pre-merge checks: 5

Check name	Status	Explanation
Description Check		Check skipped - CodeRabbit’s high-level summary is enabled.
Title check		The title clearly identifies the debug performance change and matches the main objective of reducing per-component code in the entity inspector.
Docstring Coverage		No eligible functions found in the changed files. Skipping docstring coverage check. Docstring coverage is scoped to functions touched by this diff with cyclomatic complexity of at least 2. Documentat…
Linked Issues check		Check skipped because no linked issues were found for this pull request.
Out of Scope Changes check		Check skipped because no linked issues were found for this pull request.

Full details: Docstring Coverage

Explanation

No eligible functions found in the changed files. Skipping docstring coverage check. Docstring coverage is scoped to functions touched by this diff with cyclomatic complexity of at least 2. Documentation-only edits and deletions within surviving functions are included; unchanged diff context and fully deleted functions are excluded. Analyzed 0 functions across 0 files. (1 skipped: 1 unsupported.)

Finishing Touches

Generate unit tests (beta)

• Commit to this branch
• Create a new PR

• Autofix · Keep fixing CodeRabbit findings and required CI, and resolving merge conflicts

_{I’m a rabbit with a debug view,
I hop through labels, one by one.
I check each component in the list,
Then peek at fields beneath the sun.
Fifty rows, and then I’m done!}

[chatgpt-codex-connector]

Codex Review Summary

This comment shows the latest Codex review activity on this pull request.

Review	Status	Commit	Review trigger
📝 Code Review	✅ Completed 2026-10-31T02:57:20.245074Z	a81adda	Draft marked ready

ℹ️ About Codex in GitHub

Your team has set up Codex to review pull requests in this repo. Reviews are triggered when you

• Open a pull request for review
• Mark a draft as ready
• Comment "@codex review" or "@codex security review".

Codex reacts with 👀 while any review is running, comments if it has suggestions, and reacts with 👍 once all reviews finish with no findings.

[apotema]

Codex, OpenAI LLM — review of a81addad9e9d0b765340bb0947dc4f84e7886ce8.

No correctness findings in the inspector outlining diff. The fixed UI labels, truncated-name fallback, runtime row dispatch, component ordering, filtering, 50-entity cap, selection, and dead-entity handling remain consistent with the old path. PositionOf(Game) retains the fallback while allowing the test game override.

Validation: local plugin zig build test -j2 --summary all passes all 15 Debug tests; current-head GitHub CI run 41111827084 is successful. I did not perform an interactive FP inspector smoke test or repeat the WASM memory benchmark.

Consumption note: FP vendors plugins/debug; a toolkit/assembler pin bump alone will not deliver this optimization to FP. Re-vendor that plugin as part of consumer adoption, as the PR description already notes.