[PR #5573] [CLOSED] fix(native): mark Objective-C declarations partial when the budget cuts their methods #6971

New issue

Closed

opened 2026-11-01 21:44:24 +00:00 by bsakaguchi · 0 comments

bsakaguchi commented

2026-11-01 21:44:24 +00:00

Owner

📋 Pull Request Information

Original PR: https://github.com/morluto/rea/pull/5573
Author: @kaoru0822-kitauji
Created: 10/20/2026
Status: ❌ Closed

Base: main ← Head: fix/objc-method-budget-coverage

📝 Commits (1)

23ec993 fix(native): mark Objective-C declarations partial when the budget cuts their methods

📊 Changes

4 files changed (+92 additions, -30 deletions)

View changed files

📝 src/native/AppleDispatchMetadata.test.ts (+41 -0)
📝 src/native/AppleObjcCategories.ts (+23 -16)
📝 src/native/AppleObjcDispatchFacets.ts (+12 -7)
📝 src/native/AppleObjcProtocols.ts (+16 -7)

📄 Description

Summary

Fixes #5543. When --max-records stopped decoding inside an Objective-C method list, inspect-native-dispatch-metadata still returned the protocol or class with decode: {status: "decoded"} and methods: [], so a truncated list looked like a genuinely empty one. The declaring record now reports partial with reason method_list_record_budget_exhausted. Aggregate truncation warnings are unchanged, and the methods are still returned when the budget admits them.

Problem and expected behavior

The method-list loops in AppleObjcProtocols.ts (protocol methods) and AppleObjcDispatchFacets.ts (class and category methods) break when budget.admit() fails. The declaring record never learned of that, so its decode status stayed decoded. docs/tool-design.md requires unknown coverage to remain distinguishable from absence, and AppleObjcProtocols.ts already marks a declaration partial for a skipped empty selector.

Change and scope

Protocol methods() sets a per-record budget.exhausted flag. Class/category methods() returns { selectors, budgetExhausted }.
Protocol, class and category records report partial / method_list_record_budget_exhausted when any of their method lists was cut. The recoverable budget reason takes precedence over existing record-local reasons, matching how other limit reasons are reported first.
Not included: ivar, property and adopted-protocol lists cut by the same budget. Those facets keep their existing behavior, and only aggregate truncation reports them. That is a possible follow-up, kept out of this PR to keep it focused on the reported method facts.

Contribution scope: ordinary fix within existing boundaries; maintainer scope decision not applicable.

Contract and boundary impact

Semantic owner and earliest changed stage: Objective-C metadata decoders in src/native/AppleObjc*.ts.
Contribution scope fit and linked maintainer decision: ordinary fix; not applicable.
CLI and MCP/tool-catalog contract: no schema change (decode.reason is a nullable string); affected records change from decoded to partial.
Provider, bridge, target-format, or platform compatibility: Mach-O Objective-C metadata only.
Evidence, artifact, provenance, or reconstruction contract: records now state their own omission; returned methods and evidence unchanged.
Target selection, authority, upstream behavior, and cleanup details: not applicable.
Generated metadata, package, or installation impact: none.
Read existing files or captured data.
Launch or attach to a process.
Inject instrumentation or run scripts/code.
Change process memory, registers, files, or configuration.
Contact network services.
Recover protected code or work around protections.
Documentation or usability only; no target interaction.
Not sure yet; investigation or maintainer guidance needed.

Evidence and regression coverage

Test added: src/native/AppleDispatchMetadata.test.ts decodes the synthetic fixture at every budget from 1 to 39. It asserts that every protocol/class record still marked decoded has exactly the full-budget method lists, and that some budgets yield method_list_record_budget_exhausted for both a protocol and a class.
Base reproduction: with the three source files stashed, the new test fails.

Real CLI repro (the issue's tiny.m, compiled with Apple clang, inspected statically, built CLI):

`--max-records`	TinyProtocol	TinyClass (instance)
1	0 methods, `partial` / `method_list_record_budget_exhausted`	not admitted
2	1 method, `decoded`	not admitted
3	1 method, `decoded`	0 methods, `partial` / `method_list_record_budget_exhausted`
4	1 method, `decoded`	1 method, `decoded`
5000	1 method, `decoded`	1 method, `decoded`

Remaining proof gaps: category records have no budget-cut regression case (the fixture has no categories); they share the class method reader. macOS arm64 only.
Observed, derived, and inferred claims remain distinguishable.
Artifact identity, source provenance, and failed attempts remain preserved.
Unsupported, incomplete, unavailable, or uncertain outcomes remain visible.

Validation performed

npm run test:focused -- src/native/AppleDispatchMetadata.test.ts tests/boundary/cli/nativeDispatchPropertyAttributes.test.ts tests/boundary/mcp/nativeDispatchAvailability.test.ts: 71 passed
npm run test:focused -- src/native/AppleDispatchMetadata.test.ts with the source fix stashed: 1 failed (new test), 52 passed
npm run check:fast: passed
npm run build:cached, then the issue's five-budget CLI loop above: all exited 0 with the results shown

Compatibility, safety, and release

Breaking changes or migration steps: none; budget-cut records now report partial instead of decoded.
Real Hopper/Ghidra, browser, or OS coverage: macOS arm64 with Apple clang; no provider involved; fixture never executed.
Package or release metadata impact: none.
Security, privacy, process, or containment review: no new reads, processes or side effects.

Author readiness

Human owner:

I reviewed the final diff and relevant existing code, understand the change and its edge cases, and take responsibility for the complete PR, including AI-assisted work.
I checked the evidence above, confirmed it supports the current change, and recorded meaningful validation gaps.
I declare this PR ready for initial maintainer review.

Before merge

This PR addresses a concrete problem or an agreed enhancement.
Material scope and authority questions are resolved, and any required maintainer scope decision covers the final diff.
The PR has one focused outcome and the title follows type(scope): outcome.
Related issue is linked, or the reason for not linking one is stated above.
Tests cover changed observable behavior and meaningful failure paths.
Owning docs, contracts, and generated metadata are updated where needed.
User-visible CLI/MCP changes include representative output.
I checked the final diff for secrets, unrelated cleanup, and unsupported claims.

🤖 Generated with Claude Code

_{🔄 This issue represents a GitHub Pull Request. It cannot be merged through Gitea due to API limitations.}

## 📋 Pull Request Information **Original PR:** https://github.com/morluto/rea/pull/5573 **Author:** [@kaoru0822-kitauji](https://github.com/kaoru0822-kitauji) **Created:** 10/20/2026 **Status:** ❌ Closed **Base:** `main` ← **Head:** `fix/objc-method-budget-coverage` --- ### 📝 Commits (1) - [`23ec993`](https://github.com/morluto/rea/commit/23ec9931ba07810bbe6b9068c06c98bae2809bfc) fix(native): mark Objective-C declarations partial when the budget cuts their methods ### 📊 Changes **4 files changed** (+92 additions, -30 deletions) <details> <summary>View changed files</summary> 📝 `src/native/AppleDispatchMetadata.test.ts` (+41 -0) 📝 `src/native/AppleObjcCategories.ts` (+23 -16) 📝 `src/native/AppleObjcDispatchFacets.ts` (+12 -7) 📝 `src/native/AppleObjcProtocols.ts` (+16 -7) </details> ### 📄 Description ## Summary Fixes #5543. When `--max-records` stopped decoding inside an Objective-C method list, `inspect-native-dispatch-metadata` still returned the protocol or class with `decode: {status: "decoded"}` and `methods: []`, so a truncated list looked like a genuinely empty one. The declaring record now reports `partial` with reason `method_list_record_budget_exhausted`. Aggregate truncation warnings are unchanged, and the methods are still returned when the budget admits them. ## Problem and expected behavior The method-list loops in `AppleObjcProtocols.ts` (protocol methods) and `AppleObjcDispatchFacets.ts` (class and category methods) `break` when `budget.admit()` fails. The declaring record never learned of that, so its decode status stayed `decoded`. `docs/tool-design.md` requires unknown coverage to remain distinguishable from absence, and `AppleObjcProtocols.ts` already marks a declaration partial for a skipped empty selector. ## Change and scope - Protocol `methods()` sets a per-record `budget.exhausted` flag. Class/category `methods()` returns `{ selectors, budgetExhausted }`. - Protocol, class and category records report `partial` / `method_list_record_budget_exhausted` when any of their method lists was cut. The recoverable budget reason takes precedence over existing record-local reasons, matching how other limit reasons are reported first. - Not included: ivar, property and adopted-protocol lists cut by the same budget. Those facets keep their existing behavior, and only aggregate truncation reports them. That is a possible follow-up, kept out of this PR to keep it focused on the reported method facts. Contribution scope: ordinary fix within existing boundaries; maintainer scope decision not applicable. ## Contract and boundary impact - Semantic owner and earliest changed stage: Objective-C metadata decoders in `src/native/AppleObjc*.ts`. - Contribution scope fit and linked maintainer decision: ordinary fix; not applicable. - CLI and MCP/tool-catalog contract: no schema change (`decode.reason` is a nullable string); affected records change from `decoded` to `partial`. - Provider, bridge, target-format, or platform compatibility: Mach-O Objective-C metadata only. - Evidence, artifact, provenance, or reconstruction contract: records now state their own omission; returned methods and evidence unchanged. - Target selection, authority, upstream behavior, and cleanup details: not applicable. - Generated metadata, package, or installation impact: none. - [x] Read existing files or captured data. - [ ] Launch or attach to a process. - [ ] Inject instrumentation or run scripts/code. - [ ] Change process memory, registers, files, or configuration. - [ ] Contact network services. - [ ] Recover protected code or work around protections. - [ ] Documentation or usability only; no target interaction. - [ ] Not sure yet; investigation or maintainer guidance needed. ## Evidence and regression coverage - Test added: `src/native/AppleDispatchMetadata.test.ts` decodes the synthetic fixture at every budget from 1 to 39. It asserts that every protocol/class record still marked `decoded` has exactly the full-budget method lists, and that some budgets yield `method_list_record_budget_exhausted` for both a protocol and a class. - Base reproduction: with the three source files stashed, the new test fails. - Real CLI repro (the issue's `tiny.m`, compiled with Apple clang, inspected statically, built CLI): | `--max-records` | TinyProtocol | TinyClass (instance) | |---|---|---| | 1 | 0 methods, `partial` / `method_list_record_budget_exhausted` | not admitted | | 2 | 1 method, `decoded` | not admitted | | 3 | 1 method, `decoded` | 0 methods, `partial` / `method_list_record_budget_exhausted` | | 4 | 1 method, `decoded` | 1 method, `decoded` | | 5000 | 1 method, `decoded` | 1 method, `decoded` | - Remaining proof gaps: category records have no budget-cut regression case (the fixture has no categories); they share the class method reader. macOS arm64 only. - [x] Observed, derived, and inferred claims remain distinguishable. - [x] Artifact identity, source provenance, and failed attempts remain preserved. - [x] Unsupported, incomplete, unavailable, or uncertain outcomes remain visible. ## Validation performed - `npm run test:focused -- src/native/AppleDispatchMetadata.test.ts tests/boundary/cli/nativeDispatchPropertyAttributes.test.ts tests/boundary/mcp/nativeDispatchAvailability.test.ts`: 71 passed - `npm run test:focused -- src/native/AppleDispatchMetadata.test.ts` with the source fix stashed: 1 failed (new test), 52 passed - `npm run check:fast`: passed - `npm run build:cached`, then the issue's five-budget CLI loop above: all exited 0 with the results shown ## Compatibility, safety, and release - Breaking changes or migration steps: none; budget-cut records now report `partial` instead of `decoded`. - Real Hopper/Ghidra, browser, or OS coverage: macOS arm64 with Apple clang; no provider involved; fixture never executed. - Package or release metadata impact: none. - Security, privacy, process, or containment review: no new reads, processes or side effects. ## Author readiness Human owner:  - [ ] I reviewed the final diff and relevant existing code, understand the change and its edge cases, and take responsibility for the complete PR, including AI-assisted work. - [ ] I checked the evidence above, confirmed it supports the current change, and recorded meaningful validation gaps. - [ ] I declare this PR ready for initial maintainer review. ## Before merge - [ ] This PR addresses a concrete problem or an agreed enhancement. - [ ] Material scope and authority questions are resolved, and any required maintainer scope decision covers the final diff. - [ ] The PR has one focused outcome and the title follows `type(scope): outcome`. - [ ] Related issue is linked, or the reason for not linking one is stated above. - [ ] Tests cover changed observable behavior and meaningful failure paths. - [ ] Owning docs, contracts, and generated metadata are updated where needed. - [ ] User-visible CLI/MCP changes include representative output. - [ ] I checked the final diff for secrets, unrelated cleanup, and unsupported claims. 🤖 Generated with [Claude Code](https://claude.com/claude-code) --- <sub>🔄 This issue represents a GitHub Pull Request. It cannot be merged through Gitea due to API limitations.</sub>

bsakaguchi

2026-11-01 21:44:24 +00:00