morluto/rea

Fork 0

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

[PR #7039] [MERGED] fix(analysis-view): preserve current observation paths in summaries #7596

New issue

Closed

opened 2026-11-03 13:38:59 +00:00 by bsakaguchi · 0 comments

bsakaguchi commented

2026-11-03 13:38:59 +00:00

Owner

📋 Pull Request Information

Original PR: https://github.com/morluto/rea/pull/7039
Author: @WenyuChiou
Created: 10/24/2026
Status: ✅ Merged
Merged: 10/24/2026
Merged by: @morluto

Base: main ← Head: fix/retained-summary-current-path-20261024

📝 Commits (2)

10795a4 fix(analysis-view): preserve current observation paths in summaries
eee7940 merge: integrate upstream Evidence sealing repair

📊 Changes

3 files changed (+209 additions, -23 deletions)

View changed files

📝 src/application/analysisView/AnalysisViewService.test.ts (+101 -1)
📝 src/application/analysisView/AnalysisViewService.ts (+49 -14)
📝 tests/boundary/mcp/analyzeFunctionDetail.test.ts (+59 -8)

📄 Description

Summary

Preserve the current observation's path when returning a retained summary instead of substituting metadata from an earlier same-ID ledger record.

Complete and summary delivery should describe the same current observation. Explicit later lookup by retained ID should continue to show the historical record.

Problem and expected behavior

A production MCP analyze_function regression returned /selected/native.exe for complete delivery but /old/native.exe for summary delivery. The production EvidenceLedger correctly preserves original metadata when a semantic ID is recorded again; the summary path incorrectly came from that historical lookup.

The previous MCP fixture used an overwrite-map and hid this behavior. The regression now uses the production ledger and real MCP SDK/server/registration/delivery boundaries with a synthetic provider dossier.

Change and scope

Keep ledger duplicate-ID policy and historical by-ID lookup unchanged.
After recording the current Evidence, retain the existing missing-record and identity validation.
Keep the authenticated retained record as payload/sealing owner while using a guarded presentation envelope for the current subject/path.
Preserve retained-source admission and the existing inline accounting accumulator; do not re-admit a large retained payload as generic inline data.
Test complete/summary path parity and later historical lookup separately.

This is a focused application-layer presentation fix, not a general metadata rewrite. It does not copy caller_input_path into later retained views. It does not edit the domain sealing owner in evidence.ts or duplicate #7036's separate JSON ownership correction.

Contract and boundary impact

Semantic owner: retained-summary projection in AnalysisViewService.
CLI/MCP and public Evidence schemas: unchanged.
Retained payload identity, ledger history and source admission: preserved.
Provider, target selection, process/network/mutation authority: unchanged.
Dependencies, installation, generated metadata and release behavior: unchanged.
Contribution scope: routine correction within existing Evidence delivery; no separate advance capability decision required.

Evidence and regression coverage

Before the fix, the owner regression failed with historical same-ID metadata while its clean-ledger control passed. The actual MCP regression likewise failed current-path parity while its control passed.

Final coverage includes a large 40,000-character-header capture: retained projections remain admitted while the equivalent generic inline envelope is refused. It also covers unchanged retained root/payload ownership, record failure, missing-record and wrong-ID guards, current MCP paths, later historical lookup and existing oversized MCP views.

Independent automated static/log review approved the final three-file patch; the reviewer did not independently execute tests. No human readiness attestation is implied.

Two fixture mistakes were corrected and their failed attempts preserved: missing parsed pagination defaults, and an incorrect expected error tag. Exact assertions and failure reasons remain checked. A separate environment-metadata hypothesis was not established as a defect and is excluded from this patch.

Validation performed

On pinned base 31fc9762a34f30aae1832994bfad6209a023f03e, serial bounded checks:

Five focused files — 99 tests passed: AnalysisViewService, analyzeFunctionDetail MCP, EvidenceLedger, analysisViewParent and analysisViewMcp. Actual MCP SDK in-memory transport and production record/projection owners; synthetic provider data.
node scripts/run-typescript.mjs --noEmit --checkers 1 --maxMemory 2G — whole-project typecheck passed.
Scoped lint, format, module boundaries and git diff --check — passed without warnings.

Local merge-tree against 425776174c1826d09b629cf05d2a5a79eb729239 was clean but not executed. Before publication, main advanced to a834ed1cab5327485fdd8845108b63ffcd8654fe; its changes do not touch these three files. Runtime evidence applies to the pinned-base patch, not an independently tested latest-main integration.

Compatibility and limits

No real Ghidra/native engine, browser, newly compiled CLI, full build/CI/coverage or package qualification was run locally. No performance improvement is claimed. The scope is current summary path presentation, not all historical metadata.

The API-published commit differs in metadata from reviewed local commit a3b14683b88c401133ab8d9ab8f8bdc9b202e5c3; its complete tree is identical: c14767a10eb6a975082d68d137ff7800a1f7c3dc. All three public file blob hashes were verified against the reviewed files.

Author readiness

Human owner: @WenyuChiou

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

Required current-head checks and relevant verification are satisfied or their gaps are explicitly resolved by maintainers.
The human owner has completed the readiness declarations for this PR.
Material scope and authority questions are resolved.

_{🔄 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/7039 **Author:** [@WenyuChiou](https://github.com/WenyuChiou) **Created:** 10/24/2026 **Status:** ✅ Merged **Merged:** 10/24/2026 **Merged by:** [@morluto](https://github.com/morluto) **Base:** `main` ← **Head:** `fix/retained-summary-current-path-20261024` --- ### 📝 Commits (2) - [`10795a4`](https://github.com/morluto/rea/commit/10795a4b37e123e250f67095857494981ffd95c2) fix(analysis-view): preserve current observation paths in summaries - [`eee7940`](https://github.com/morluto/rea/commit/eee794035ea4248e644506f1b10117cb53f45c59) merge: integrate upstream Evidence sealing repair ### 📊 Changes **3 files changed** (+209 additions, -23 deletions) <details> <summary>View changed files</summary> 📝 `src/application/analysisView/AnalysisViewService.test.ts` (+101 -1) 📝 `src/application/analysisView/AnalysisViewService.ts` (+49 -14) 📝 `tests/boundary/mcp/analyzeFunctionDetail.test.ts` (+59 -8) </details> ### 📄 Description ## Summary Preserve the current observation's path when returning a retained summary instead of substituting metadata from an earlier same-ID ledger record. Complete and summary delivery should describe the same current observation. Explicit later lookup by retained ID should continue to show the historical record. ## Problem and expected behavior A production MCP `analyze_function` regression returned `/selected/native.exe` for complete delivery but `/old/native.exe` for summary delivery. The production EvidenceLedger correctly preserves original metadata when a semantic ID is recorded again; the summary path incorrectly came from that historical lookup. The previous MCP fixture used an overwrite-map and hid this behavior. The regression now uses the production ledger and real MCP SDK/server/registration/delivery boundaries with a synthetic provider dossier. ## Change and scope - Keep ledger duplicate-ID policy and historical by-ID lookup unchanged. - After recording the current Evidence, retain the existing missing-record and identity validation. - Keep the authenticated retained record as payload/sealing owner while using a guarded presentation envelope for the current subject/path. - Preserve retained-source admission and the existing inline accounting accumulator; do not re-admit a large retained payload as generic inline data. - Test complete/summary path parity and later historical lookup separately. This is a focused application-layer presentation fix, not a general metadata rewrite. It does not copy `caller_input_path` into later retained views. It does not edit the domain sealing owner in `evidence.ts` or duplicate #7036's separate JSON ownership correction. ## Contract and boundary impact - Semantic owner: retained-summary projection in AnalysisViewService. - CLI/MCP and public Evidence schemas: unchanged. - Retained payload identity, ledger history and source admission: preserved. - Provider, target selection, process/network/mutation authority: unchanged. - Dependencies, installation, generated metadata and release behavior: unchanged. - Contribution scope: routine correction within existing Evidence delivery; no separate advance capability decision required. ## Evidence and regression coverage Before the fix, the owner regression failed with historical same-ID metadata while its clean-ledger control passed. The actual MCP regression likewise failed current-path parity while its control passed. Final coverage includes a large 40,000-character-header capture: retained projections remain admitted while the equivalent generic inline envelope is refused. It also covers unchanged retained root/payload ownership, record failure, missing-record and wrong-ID guards, current MCP paths, later historical lookup and existing oversized MCP views. Independent automated static/log review approved the final three-file patch; the reviewer did not independently execute tests. No human readiness attestation is implied. Two fixture mistakes were corrected and their failed attempts preserved: missing parsed pagination defaults, and an incorrect expected error tag. Exact assertions and failure reasons remain checked. A separate environment-metadata hypothesis was not established as a defect and is excluded from this patch. ## Validation performed On pinned base `31fc9762a34f30aae1832994bfad6209a023f03e`, serial bounded checks: - Five focused files — **99 tests passed**: AnalysisViewService, analyzeFunctionDetail MCP, EvidenceLedger, analysisViewParent and analysisViewMcp. Actual MCP SDK in-memory transport and production record/projection owners; synthetic provider data. - `node scripts/run-typescript.mjs --noEmit --checkers 1 --maxMemory 2G` — whole-project typecheck passed. - Scoped lint, format, module boundaries and `git diff --check` — passed without warnings. Local merge-tree against `425776174c1826d09b629cf05d2a5a79eb729239` was clean but not executed. Before publication, main advanced to `a834ed1cab5327485fdd8845108b63ffcd8654fe`; its changes do not touch these three files. Runtime evidence applies to the pinned-base patch, not an independently tested latest-main integration. ## Compatibility and limits No real Ghidra/native engine, browser, newly compiled CLI, full build/CI/coverage or package qualification was run locally. No performance improvement is claimed. The scope is current summary path presentation, not all historical metadata. The API-published commit differs in metadata from reviewed local commit `a3b14683b88c401133ab8d9ab8f8bdc9b202e5c3`; its complete tree is identical: `c14767a10eb6a975082d68d137ff7800a1f7c3dc`. All three public file blob hashes were verified against the reviewed files. ## Author readiness Human owner: @WenyuChiou - [x] 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. - [x] I checked the evidence above, confirmed it supports the current change, and recorded meaningful validation gaps. - [x] I declare this PR ready for initial maintainer review. ## Before merge - [ ] Required current-head checks and relevant verification are satisfied or their gaps are explicitly resolved by maintainers. - [ ] The human owner has completed the readiness declarations for this PR. - [ ] Material scope and authority questions are resolved. --- <sub>🔄 This issue represents a GitHub Pull Request. It cannot be merged through Gitea due to API limitations.</sub>

bsakaguchi

2026-11-03 13:38:59 +00:00

closed this issue
added the
pull-request
label

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#7596

No description provided.

Rows
Columns