[M3gA-Mind]

Summary

Adds a golden-value test for key_of's hashed path (escaped keys longer than 512 bytes). The value must match what the legacy tinystoragedrivers-core Key::of produced, so dedupe (teq.outcomes) and buffer rows written before the storage-port migration stay addressable after an upgrade.

Related issue

None. Raised by the 2026-10-23 openhuman core-data audit (Q-L1), tracked from tinyhumansai/openhuman.

API or behavior changes

None. Test only.

Validation

Commands actually run, with their outcome:

• cargo fmt --all -- --check: not run workspace-wide; rustfmt --edition 2021 --check on the changed file is clean
• cargo clippy --all-targets --all-features -- -D warnings: left to CI
• cargo build --all-targets --all-features: left to CI
• cargo test --all-features: left to CI. Ran cargo test -p tinyeventqueues --lib -- ports: 4 passed. With key_of's encoding deliberately broken (to_ascii_uppercase in place of to_ascii_lowercase), the new test fails and the existing two still pass.

Tests

• ports::tests::oversized_keys_hash_exactly_like_the_legacy_key_of pins two literals:
  • a single 600-byte part;
  • a composite with an escaped / (["a/b", "y" × 600]).
• The expected values follow the legacy algorithm at tinystoragedrivers-core 0c6dcfce (key/mod.rs Key::of): ~ + lowercase unpadded RFC 4648 base32 of SHA-256 of the escaped key, with MAX_ID_LEN = 512. They were computed independently from the Rust code, and the current key_of reproduces them.
• The existing oversized_composite_keys_are_bounded_and_deterministic only checked determinism and length, so a changed encoding would have passed it.

Documentation

Not needed: no behaviour or API change.

Checklist

• The change is focused on one logical change
• No new #[allow(...)], #[ignore], or relaxed lints
• No secrets, tokens, or .env contents in the diff or the description

[coderabbitai]

Important

• 🔍 Trigger review

This repository does not receive automatic reviews because it has fewer than 10 stars.

⚙️ Run configuration

• Source mode: Auto
• Repository: Unavailable
• Branch: Unavailable
• File: No configuration file from this repository applied
• Commit: Unavailable
• Layers: Organization UI
• Review profile: CHILL
• Plan: Advanced
• Run ID: c69a081c-8cc5-4574-b087-fb212284995d

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

_{Comment @coderabbitai help to get the list of available commands.}

[tinysweeper]

Tiny Sweeper review

Reviewed head: 0d965cafff208c26a06802dc122e01c64a89d1cb

Verdict: Ready for maintainer review · Priority: none
Coverage: Reviewed 1 of 1 files.
Pull request purpose (title): test(ports): oversiz…

Diagnostics and supporting details

Full summary

Pull request purpose (title): test(ports): oversized keys hash exactly like the legacy Key::of. Tiny Sweeper completed its review; deterministic results follow.

Review snapshot

Change surface	Files	Review signal	Count
Production	0	Active findings	0
Tests	1	Noted findings	0
Documentation	0	Resolved findings	0
Configuration	0	Pending checks/questions	0

Completeness: Complete
Test assessment: No supported feature-to-test mapping was available; this does not mean tests are absent or passed.

What changed

Adds a golden-value characterization test in `crates/tinyeventqueues/src/ports_tests.rs` (`fn oversized_keys_hash_exactly_like_the_legacy_key_of()`), asserting that `key_of` on oversized keys returns two exact `~`-prefixed encoded strings: one for a single 600-byte part and one for the composite `["a/b", "y" × 600]`. Per the description, no production behavior or API changes; the encoding is pinned to what the legacy `tinystoragedrivers-core` `Key::of` produced.

Features

• Added — Golden-value test pinning oversized-key hash encoding: Locks the encoded form of keys exceeding the size limit to fixed expected literals, so a drift in the hash encoding would be caught by a failing test rather than silently orphaning pre-migration dedupe and buffer rows. Test-only; no runtime behavior change. (crates/tinyeventqueues/src/ports_tests.rs fn oversized_composite_keys_are_bounded_and_deterministic() {)

Tests

No supported feature-to-test mapping was produced. Test execution is not inferred.

Findings

No active actionable findings.

Before merge

None.

Agent review details

critique

• Conclusion: Success
• Scope reviewed: all assigned evidence
• Positive: Critique lane answered with no findings, noting the test pins the legacy key-hash layout with golden values and that the `key_of` escaping rules look consistent with the inputs used.
• Lane summary: The new test pins the legacy key-hash layout with golden values; I could not independently verify the two base32 digests without running the helper, and the escaping rules in `key_of` (including `%7E` for a leading `~` in part 0 and `%2F` in the second part) look consistent with the inputs used. No concrete defect identified in this test-only diff. _The code index is behind this pull request (indexed at `a719a6e00927`), so retrieved context may be out of date._

security

• Conclusion: Success
• Scope reviewed: all assigned evidence
• Positive: Security lane answered with no findings: a pure unit test with no production code path, no external input, and no behavior change; a wrong constant would fail the test rather than ship a defect.
• Lane summary: This adds two golden-value assertions pinning `key_of`'s oversized-key hashing to the legacy `Key::of` encoding. It is a pure unit test with no production code path, no external input, and no change to behaviour; nothing reportable. I did not independently recompute the base32 digests, but a wrong constant would fail the test it adds rather than ship a defect. _The code index is behind this pull request (indexed at `a719a6e00927`), so retrieved context may be out of date._

tests

• Conclusion: Success
• Scope reviewed: all assigned evidence
• Positive: Tests lane answered with no findings, observing the new test exercises the changed hashing path and would fail if the hash encoding drifted.
• Lane summary: The change adds a golden-value characterization test for `key_of`'s oversized-key hashing path, pinning two exact encoded strings. It exercises the changed behaviour and would fail if the hash encoding drifted. I could not independently recompute the base32 SHA-256 goldens from the evidence here, but they are exact literals rather than a truthiness assertion, so the test can genuinely fail. No reportable defect found. _The code index is behind this pull request (indexed at `a719a6e00927`), so retrieved context may be out of date._

commits

• Conclusion: Success
• Scope reviewed: all assigned evidence
• Positive: Commits lane answered with no findings: no sensitive content in committed files.
• Lane summary: No sensitive content in committed files. _The code index is behind this pull request (indexed at `a719a6e00927`), so retrieved context may be out of date._

description

• Conclusion: Success
• Scope reviewed: all assigned evidence
• Positive: Description lane answered with no findings, noting the diff matches the PR description (test-only, no behavior change) and documents why the golden values must match the legacy encoding.
• Lane summary: Adds a golden-value test pinning `key_of`'s oversized-key hash output against two literals, matching the PR description exactly (test-only, no behaviour change). One detail in the description's escape example is loose, but the description still matches and explains what the diff does; I found no reportable defect in the change itself. I could not independently verify the two base32 literals against the legacy `Key::of`, since that algorithm lives in the parent repository, not here. _The code index is behind this pull request (indexed at `a719a6e00927`), so retrieved context may be out of date._

Evidence and run details

• Models: ladder/vectors-oai3, deepseek-v4.1-flash
• Spend: $0.000242
• Tokens: 79792 input · 2248 output · 12288 cached · 167 embedding
• Updated: 2026-10-23T00:33:37Z

Head	State	Pass summary
0d965cafff20	ready for maintainer review	0 active finding(s), 0 resolved finding(s) (at 2026-10-23T00:33:37Z)

[tinysweeper]

tinysweeper found nothing blocking. Approving.

Diagnostics

Total spend: $0.0002