morluto/rea

Fork 0

mirror of https://github.com/morluto/rea.git synced 2026-11-05 06:45:22 +00:00

[PR #2364] [MERGED] fix(native): parse renamed dyld_info re-exports #5189

New issue

Closed

opened 2026-10-31 01:04:40 +00:00 by bsakaguchi · 0 comments

bsakaguchi commented

2026-10-31 01:04:40 +00:00

Owner

📋 Pull Request Information

Original PR: https://github.com/morluto/rea/pull/2364
Author: @Chaoqi31
Created: 10/12/2026
Status: ✅ Merged
Merged: 10/13/2026
Merged by: @morluto

Base: main ← Head: fix/macho-renamed-reexport

📝 Commits (5)

a04874f fix(native): parse renamed dyld_info re-exports
1574be4 fix(native): split dyld re-export suffix without nested quantifiers
3c3eb57 fix: reconcile captured main and owned findings for PR #2364
1ff17d7 test(native): cover nested re-export library names
89bfda5 fix(native): retain reexports missing from nm inventory

📊 Changes

5 files changed (+196 additions, -4 deletions)

View changed files

📝 src/native/NativeMachoInspection.ts (+5 -0)
📝 src/native/parsers/dyldInfo.ts (+81 -4)
📝 tests/boundary/providers/native/nativeInventory.golden.test.ts (+92 -0)
📝 tests/fixtures/native-macos/dyld-inventory/README.md (+13 -0)
➕ tests/fixtures/native-macos/dyld-inventory/renamed-reexports.txt (+5 -0)

📄 Description

Summary

inspect_macho misreported a dylib that re-exports a symbol under a new name. It listed an export named _foo_renamed (_foo from libfoo) that does not exist, with no source library, and kept nm's _foo_renamed as a second, unattributed export. The export now appears once, as _foo_renamed re-exported from libfoo, and the exports coverage states that the source library's own symbol name is not reported.

No issue is linked. I found this while comparing inspect_macho exports with xcrun dyld_info -exports on synthetic dylibs.

Problem and expected behavior

Smallest trigger: a dylib linked against libfoo.dylib with -Wl,-alias,_foo,_foo_renamed. xcrun dyld_info -exports prints the renamed re-export with its source symbol:

libre.dylib [arm64]:
    -exports:
        offset      symbol
        [re-export] _foo_renamed (_foo from libfoo)
        0x00000308  _own

parseDyldSymbols in src/native/parsers/dyldInfo.ts only accepted (from libfoo), so the whole remainder became the name and the source was null. NativeMachoInspection merges dyld rows with nm rows by name, so nm's _foo_renamed matched nothing and stayed as a second row with reexport: null.

Expected: one row per exported name, flagged as a re-export with its source library, the same as an ordinary [re-export] _sym (from lib) row.

Change and scope

The re-export row pattern accepts an optional source symbol before from. It tries the plain (from lib) form first, so a library whose install name contains from keeps its whole name, and a source symbol may itself contain parentheses, such as operator(). The row is named by its exported name and replaces the nm row like any other re-export.

The export item schema has no field for the source symbol (_foo), and adding one is a contract change. This PR keeps it out of the items and adds an exports.limitations entry whenever a renamed re-export is present, so the omission is visible. Ordinary re-export rows and import rows parse as before.

Contribution scope: not applicable (ordinary fix within existing boundaries).

Contract and boundary impact

Semantic owner and earliest changed stage: parseDyldSymbols (src/native/parsers/dyldInfo.ts), which now also returns the source symbol, and NativeMachoInspection, which states the limitation.
Contribution scope fit and linked maintainer decision: ordinary fix; not applicable.
CLI and MCP/tool-catalog contract: no schema change. exports.items changes only for renamed re-exports, and exports.limitations gains one entry when they are present.
Provider, bridge, target-format, or platform compatibility: macOS native provider only. Other dyld_info row shapes keep their existing golden coverage.
Evidence, artifact, provenance, or reconstruction contract: removes a fabricated export name. The source symbol name is not represented, and the limitation says so.
Target selection, authority, upstream behavior, and cleanup details (if applicable): not applicable.
Generated metadata (docs/public/product-catalog.json), package, or installation impact: none.

Expected effects (check known effects; use documentation-only on its own).
Unchecked items do not establish absence; leave uncertainties visible.

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

Tests added or updated: tests/boundary/providers/native/nativeInventory.golden.test.ts adds two tests. "reports a renamed re-export once under its exported name" runs inspect_macho through FixtureRunner with a new golden capture, tests/fixtures/native-macos/dyld-inventory/renamed-reexports.txt, and checks the items and limitations. The fixture README records the commands that produce it. "splits re-export rows at the source library" parses rows captured from synthetic dylibs whose source symbol is operator() and whose library install name is lib from foo.
Base reproduction or other evidence: with dyldInfo.ts and NativeMachoInspection.ts restored from main (fd958808), both new tests fail and the other 14 tests in the file pass.

User-visible CLI/MCP output (if applicable): rea inspect-macho libre.dylib --format json on the dylib above, normalized_result.exports.

Before (main):

{"name": "_foo_renamed", "address": null, "weak": null, "reexport": null, "source": "nm"}
{"name": "_foo_renamed (_foo from libfoo)", "address": null, "weak": null, "reexport": true, "source": null}
{"name": "_own", "address": "0x308", "weak": null, "reexport": false, "source": null}

After:

{"name": "_foo_renamed", "address": null, "weak": null, "reexport": true, "source": "libfoo"}
{"name": "_own", "address": "0x308", "weak": null, "reexport": false, "source": null}
limitations: ["Merged nm/dyld_info results may be incomplete for stripped binaries.", "Renamed re-exports are listed under their exported name; the source library's symbol name is not reported."]

Remaining proof gaps: captured on arm64 with the Xcode toolchain on this Mac; I did not capture x86_64 output. dyld_info text cannot distinguish an exported name that itself contains (… from …); such a name would be split inside the name.

For evidence-bearing changes:

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

On the branch rebased onto main fd958808, with dependencies installed by npm ci: npm run check:fast passes.
npm run test:focused over the 24 test files in tests/boundary/providers/native/native*.test.ts and src/native/. 279 pass and 3 fail. The 3 failures are the real-hdiutil cases in nativeDmgArtifactReader.test.ts; they fail the same way on main fd958808 on this Mac, and this change does not touch DMG code.
The before/after CLI run above, with npm run build:cached on each checkout.

Compatibility, safety, and release

Breaking changes or migration steps: none.
Real Hopper/Ghidra, browser, or OS coverage: real xcrun dyld_info and nm on macOS (arm64) for synthetic dylibs.
Package or release metadata impact: none.
Security, privacy, process, or containment review: none; parser change only.

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.

_{🔄 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/2364 **Author:** [@Chaoqi31](https://github.com/Chaoqi31) **Created:** 10/12/2026 **Status:** ✅ Merged **Merged:** 10/13/2026 **Merged by:** [@morluto](https://github.com/morluto) **Base:** `main` ← **Head:** `fix/macho-renamed-reexport` --- ### 📝 Commits (5) - [`a04874f`](https://github.com/morluto/rea/commit/a04874ffe4eb1385a799e11757b83efd42b60130) fix(native): parse renamed dyld_info re-exports - [`1574be4`](https://github.com/morluto/rea/commit/1574be4876f0ee1a85afb5d5fc27c9b81ffb4ac3) fix(native): split dyld re-export suffix without nested quantifiers - [`3c3eb57`](https://github.com/morluto/rea/commit/3c3eb579db02d191cc481ead4ab311e3280af3f9) fix: reconcile captured main and owned findings for PR #2364 - [`1ff17d7`](https://github.com/morluto/rea/commit/1ff17d73fe4ab2c7311c1e667de19e32764211ad) test(native): cover nested re-export library names - [`89bfda5`](https://github.com/morluto/rea/commit/89bfda597426bdfcc737b7e515e771ed5f62f4d6) fix(native): retain reexports missing from nm inventory ### 📊 Changes **5 files changed** (+196 additions, -4 deletions) <details> <summary>View changed files</summary> 📝 `src/native/NativeMachoInspection.ts` (+5 -0) 📝 `src/native/parsers/dyldInfo.ts` (+81 -4) 📝 `tests/boundary/providers/native/nativeInventory.golden.test.ts` (+92 -0) 📝 `tests/fixtures/native-macos/dyld-inventory/README.md` (+13 -0) ➕ `tests/fixtures/native-macos/dyld-inventory/renamed-reexports.txt` (+5 -0) </details> ### 📄 Description ## Summary `inspect_macho` misreported a dylib that re-exports a symbol under a new name. It listed an export named `_foo_renamed (_foo from libfoo)` that does not exist, with no source library, and kept `nm`'s `_foo_renamed` as a second, unattributed export. The export now appears once, as `_foo_renamed` re-exported from `libfoo`, and the exports coverage states that the source library's own symbol name is not reported. No issue is linked. I found this while comparing `inspect_macho` exports with `xcrun dyld_info -exports` on synthetic dylibs. ## Problem and expected behavior Smallest trigger: a dylib linked against `libfoo.dylib` with `-Wl,-alias,_foo,_foo_renamed`. `xcrun dyld_info -exports` prints the renamed re-export with its source symbol: ``` libre.dylib [arm64]: -exports: offset symbol [re-export] _foo_renamed (_foo from libfoo) 0x00000308 _own ``` `parseDyldSymbols` in `src/native/parsers/dyldInfo.ts` only accepted `(from libfoo)`, so the whole remainder became the name and the source was null. `NativeMachoInspection` merges dyld rows with `nm` rows by name, so `nm`'s `_foo_renamed` matched nothing and stayed as a second row with `reexport: null`. Expected: one row per exported name, flagged as a re-export with its source library, the same as an ordinary `[re-export] _sym (from lib)` row. ## Change and scope The re-export row pattern accepts an optional source symbol before `from`. It tries the plain `(from lib)` form first, so a library whose install name contains ` from ` keeps its whole name, and a source symbol may itself contain parentheses, such as `operator()`. The row is named by its exported name and replaces the `nm` row like any other re-export. The export item schema has no field for the source symbol (`_foo`), and adding one is a contract change. This PR keeps it out of the items and adds an `exports.limitations` entry whenever a renamed re-export is present, so the omission is visible. Ordinary re-export rows and import rows parse as before. Contribution scope: not applicable (ordinary fix within existing boundaries). ## Contract and boundary impact - Semantic owner and earliest changed stage: `parseDyldSymbols` (`src/native/parsers/dyldInfo.ts`), which now also returns the source symbol, and `NativeMachoInspection`, which states the limitation. - Contribution scope fit and linked maintainer decision: ordinary fix; not applicable. - CLI and MCP/tool-catalog contract: no schema change. `exports.items` changes only for renamed re-exports, and `exports.limitations` gains one entry when they are present. - Provider, bridge, target-format, or platform compatibility: macOS native provider only. Other `dyld_info` row shapes keep their existing golden coverage. - Evidence, artifact, provenance, or reconstruction contract: removes a fabricated export name. The source symbol name is not represented, and the limitation says so. - Target selection, authority, upstream behavior, and cleanup details (if applicable): not applicable. - Generated metadata (`docs/public/product-catalog.json`), package, or installation impact: none. Expected effects (check known effects; use documentation-only on its own). Unchecked items do not establish absence; leave uncertainties visible. - [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 - Tests added or updated: `tests/boundary/providers/native/nativeInventory.golden.test.ts` adds two tests. "reports a renamed re-export once under its exported name" runs `inspect_macho` through `FixtureRunner` with a new golden capture, `tests/fixtures/native-macos/dyld-inventory/renamed-reexports.txt`, and checks the items and limitations. The fixture README records the commands that produce it. "splits re-export rows at the source library" parses rows captured from synthetic dylibs whose source symbol is `operator()` and whose library install name is `lib from foo`. - Base reproduction or other evidence: with `dyldInfo.ts` and `NativeMachoInspection.ts` restored from `main` (fd958808), both new tests fail and the other 14 tests in the file pass. - User-visible CLI/MCP output (if applicable): `rea inspect-macho libre.dylib --format json` on the dylib above, `normalized_result.exports`. Before (`main`): ``` {"name": "_foo_renamed", "address": null, "weak": null, "reexport": null, "source": "nm"} {"name": "_foo_renamed (_foo from libfoo)", "address": null, "weak": null, "reexport": true, "source": null} {"name": "_own", "address": "0x308", "weak": null, "reexport": false, "source": null} ``` After: ``` {"name": "_foo_renamed", "address": null, "weak": null, "reexport": true, "source": "libfoo"} {"name": "_own", "address": "0x308", "weak": null, "reexport": false, "source": null} limitations: ["Merged nm/dyld_info results may be incomplete for stripped binaries.", "Renamed re-exports are listed under their exported name; the source library's symbol name is not reported."] ``` - Remaining proof gaps: captured on arm64 with the Xcode toolchain on this Mac; I did not capture x86_64 output. `dyld_info` text cannot distinguish an exported name that itself contains ` (… from …)`; such a name would be split inside the name. For evidence-bearing changes: - [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 - On the branch rebased onto `main` fd958808, with dependencies installed by `npm ci`: `npm run check:fast` passes. - `npm run test:focused` over the 24 test files in `tests/boundary/providers/native/native*.test.ts` and `src/native/`. 279 pass and 3 fail. The 3 failures are the real-`hdiutil` cases in `nativeDmgArtifactReader.test.ts`; they fail the same way on `main` fd958808 on this Mac, and this change does not touch DMG code. - The before/after CLI run above, with `npm run build:cached` on each checkout. ## Compatibility, safety, and release - Breaking changes or migration steps: none. - Real Hopper/Ghidra, browser, or OS coverage: real `xcrun dyld_info` and `nm` on macOS (arm64) for synthetic dylibs. - Package or release metadata impact: none. - Security, privacy, process, or containment review: none; parser change only. ## Before merge - [x] This PR addresses a concrete problem or an agreed enhancement. - [x] Material scope and authority questions are resolved, and any required maintainer scope decision covers the final diff. - [x] The PR has one focused outcome and the title follows `type(scope): outcome`. - [x] Related issue is linked, or the reason for not linking one is stated above. - [x] Tests cover changed observable behavior and meaningful failure paths. - [x] Owning docs, contracts, and generated metadata are updated where needed. - [x] User-visible CLI/MCP changes include representative output. - [x] I checked the final diff for secrets, unrelated cleanup, and unsupported claims. --- <sub>🔄 This issue represents a GitHub Pull Request. It cannot be merged through Gitea due to API limitations.</sub>

bsakaguchi

2026-10-31 01:04:40 +00:00

closed this issue
added the
pull-request
label

bsakaguchi referenced this issue

2026-11-01 21:43:15 +00:00

[PR #5214] [MERGED] fix(hopper): preserve an explicit display number zero in the X11 helper #6819

No milestone

No project

No assignees

1 participant

Notifications

Due date

The due date is invalid or out of range. Please use the format "yyyy-mm-dd".

No due date set.

Dependencies

No dependencies set.

Reference

morluto/rea#5189

No description provided.

Rows
Columns