morluto/rea

Fork 0

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

[PR #3975] [MERGED] fix(browser): reject empty capture containers beyond the 64-level nesting budget #5976

New issue

Closed

opened 2026-10-31 01:10:53 +00:00 by bsakaguchi · 0 comments

bsakaguchi commented

2026-10-31 01:10:53 +00:00

Owner

📋 Pull Request Information

Original PR: https://github.com/morluto/rea/pull/3975
Author: @Anurag-5ingh
Created: 10/15/2026
Status: ✅ Merged
Merged: 10/22/2026
Merged by: @morluto

Base: main ← Head: fix/capture-nesting-limit

📝 Commits (1)

df9a2b3 fix(browser): reject empty capture containers beyond the 64-level nesting budget

📊 Changes

3 files changed (+45 additions, -5 deletions)

View changed files

📝 src/browser/history/CaptureJson.ts (+7 -5)
📝 src/browser/history/HarCapture.test.ts (+19 -0)
📝 src/browser/history/QlogCapture.test.ts (+19 -0)

📄 Description

Summary

Fixes #3960.

inspect-web-network-capture documents a 64-level nesting budget for complete evidence, but the shared HAR/qlog JSON parser accepted an empty array or object at container level 65, while [0] at the same level was rejected. The documented limit now holds no matter what the deepest container holds.

Problem and expected behavior

Smallest trigger: a HAR whose log._extension chain reaches container level 65 with an empty innermost [] or {} (root = level 1). It decodes successfully. The same chain ending in [0] fails with input-limit.

Cause: parseCaptureJson checked depth before pushing a container with frames.length > 64. That admitted the 65th container. Over-deep scalars were rejected only because onLiteralValue ran the same check one frame deeper.

Expected:

Containers up to level 64 are accepted.
Every container at level 65 or deeper is rejected with the existing input-limit failure, whether it is empty or holds values.

Change and scope

src/browser/history/CaptureJson.ts owns JSON structure for both decoders (HarCapture.ts, QlogCapture.ts).
onObjectBegin and onArrayBegin now admit a container only while its own level stays within the budget (frames.length >= 64 rejects).
The literal-value check is removed. A value inside an admitted container adds no level, so empty and scalar-bearing containers share one boundary.

Not included:

The HAR visitor's own depth guard is unchanged.
The SAZ/XML depth checks (SazProjection.ts, SazMetadata.ts) are a separate path and unchanged.

Follow the contribution scope: this is an ordinary fix within existing boundaries, so a maintainer scope decision is not applicable.

Contract and boundary impact

Semantic owner and earliest changed stage: parseCaptureJson in src/browser/history/CaptureJson.ts.
Contribution scope fit and linked maintainer decision: in scope; not applicable.
CLI and MCP/tool-catalog contract: no schema change. Captures with empty containers at level 65 now return the documented input-limit failure instead of a decoded capture.
Provider, bridge, target-format, or platform compatibility: HAR and qlog JSON only; platform-independent.
Evidence, artifact, provenance, or reconstruction contract: no shape change. The error message and pointer are unchanged.
Target selection, authority, upstream behavior, and cleanup details: not applicable.
Generated metadata, package, or installation impact: none.

Expected effects:

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 in src/browser/history/HarCapture.test.ts and src/browser/history/QlogCapture.test.ts. Each decoder gets a parameterized boundary table using the issue's fixture shapes (HAR log._extension, qlog trace ext):

Container level Innermost Result

64 [], {}, [0] accepted

65 [], {}, [0] nesting budget

66 [] nesting budget
Base reproduction (source-derived): on the base revision, the level-65 [] and {} cases decode instead of throwing, matching the issue's report. The existing deeper-nesting tests are unchanged.
User-visible CLI/MCP output: an affected capture now fails with HAR exceeds the 64-level complete-evidence nesting budget. (or qlog …) instead of returning a decoded capture.
Remaining proof gaps: none known beyond CI results, which are pending.

Container level	Innermost	Result
64	`[]`, `{}`, `[0]`	accepted
65	`[]`, `{}`, `[0]`	`nesting budget`
66	`[]`	`nesting budget`

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

oxfmt --check on the changed files — passed.
Full test, lint and typecheck validation relies on this PR's CI. A local npm ci is not available in my environment because the registry it uses blocks some of the locked package versions.

Compatibility, safety, and release

Breaking changes or migration steps: none. Captures with an empty container at level 65 were already outside the documented budget and are now rejected consistently.
Real Hopper/Ghidra, browser, or OS coverage: not applicable (offline capture parsing).
Package or release metadata impact: none.
Security, privacy, process, or containment review: this tightens an input resource bound; no other change.

Author readiness

Human owner: @Anurag-5ingh

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.

_{🔄 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/3975 **Author:** [@Anurag-5ingh](https://github.com/Anurag-5ingh) **Created:** 10/15/2026 **Status:** ✅ Merged **Merged:** 10/22/2026 **Merged by:** [@morluto](https://github.com/morluto) **Base:** `main` ← **Head:** `fix/capture-nesting-limit` --- ### 📝 Commits (1) - [`df9a2b3`](https://github.com/morluto/rea/commit/df9a2b35251bb32a8c7c00fd35593b9da512e072) fix(browser): reject empty capture containers beyond the 64-level nesting budget ### 📊 Changes **3 files changed** (+45 additions, -5 deletions) <details> <summary>View changed files</summary> 📝 `src/browser/history/CaptureJson.ts` (+7 -5) 📝 `src/browser/history/HarCapture.test.ts` (+19 -0) 📝 `src/browser/history/QlogCapture.test.ts` (+19 -0) </details> ### 📄 Description ## Summary Fixes #3960. `inspect-web-network-capture` documents a 64-level nesting budget for complete evidence, but the shared HAR/qlog JSON parser accepted an empty array or object at container level 65, while `[0]` at the same level was rejected. The documented limit now holds no matter what the deepest container holds. ## Problem and expected behavior Smallest trigger: a HAR whose `log._extension` chain reaches container level 65 with an empty innermost `[]` or `{}` (root = level 1). It decodes successfully. The same chain ending in `[0]` fails with `input-limit`. Cause: `parseCaptureJson` checked depth before pushing a container with `frames.length > 64`. That admitted the 65th container. Over-deep scalars were rejected only because `onLiteralValue` ran the same check one frame deeper. Expected: - Containers up to level 64 are accepted. - Every container at level 65 or deeper is rejected with the existing `input-limit` failure, whether it is empty or holds values. ## Change and scope - `src/browser/history/CaptureJson.ts` owns JSON structure for both decoders (`HarCapture.ts`, `QlogCapture.ts`). - `onObjectBegin` and `onArrayBegin` now admit a container only while its own level stays within the budget (`frames.length >= 64` rejects). - The literal-value check is removed. A value inside an admitted container adds no level, so empty and scalar-bearing containers share one boundary. Not included: - The HAR visitor's own depth guard is unchanged. - The SAZ/XML depth checks (`SazProjection.ts`, `SazMetadata.ts`) are a separate path and unchanged. Follow the [contribution scope](https://github.com/morluto/rea/blob/main/CONTRIBUTING.md#contribution-scope): this is an ordinary fix within existing boundaries, so a maintainer scope decision is not applicable. ## Contract and boundary impact - Semantic owner and earliest changed stage: `parseCaptureJson` in `src/browser/history/CaptureJson.ts`. - Contribution scope fit and linked maintainer decision: in scope; not applicable. - CLI and MCP/tool-catalog contract: no schema change. Captures with empty containers at level 65 now return the documented `input-limit` failure instead of a decoded capture. - Provider, bridge, target-format, or platform compatibility: HAR and qlog JSON only; platform-independent. - Evidence, artifact, provenance, or reconstruction contract: no shape change. The error message and pointer are unchanged. - Target selection, authority, upstream behavior, and cleanup details: not applicable. - Generated metadata, package, or installation impact: none. Expected effects: - [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 in `src/browser/history/HarCapture.test.ts` and `src/browser/history/QlogCapture.test.ts`. Each decoder gets a parameterized boundary table using the issue's fixture shapes (HAR `log._extension`, qlog trace `ext`): | Container level | Innermost | Result | | --- | --- | --- | | 64 | `[]`, `{}`, `[0]` | accepted | | 65 | `[]`, `{}`, `[0]` | `nesting budget` | | 66 | `[]` | `nesting budget` | - Base reproduction (source-derived): on the base revision, the level-65 `[]` and `{}` cases decode instead of throwing, matching the issue's report. The existing deeper-nesting tests are unchanged. - User-visible CLI/MCP output: an affected capture now fails with `HAR exceeds the 64-level complete-evidence nesting budget.` (or `qlog …`) instead of returning a decoded capture. - Remaining proof gaps: none known beyond CI results, which are pending. 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 - `oxfmt --check` on the changed files — passed. - Full test, lint and typecheck validation relies on this PR's CI. A local `npm ci` is not available in my environment because the registry it uses blocks some of the locked package versions. ## Compatibility, safety, and release - Breaking changes or migration steps: none. Captures with an empty container at level 65 were already outside the documented budget and are now rejected consistently. - Real Hopper/Ghidra, browser, or OS coverage: not applicable (offline capture parsing). - Package or release metadata impact: none. - Security, privacy, process, or containment review: this tightens an input resource bound; no other change. ## Author readiness Human owner: @Anurag-5ingh - [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 - [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 01:10:53 +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#5976

No description provided.

Rows
Columns