Anchor the ERR-04 baseline to a commit, not a date - #278
Merged
Conversation
The 1,138 figure was measured on a working tree that happened to contain the then-uncommitted hwped.rs. The number was right, but its provenance was accidental: 19 of the 1,138 matches live in that file, so on a tree without it the count reads 1,119 and the post-migration count lands at 1,109 - below the [1118, 1138] band the plan asserts against. Re-confirmed 1,138 on the committed tree at 34f96ee, and recorded that commit as the provenance in 03-02 and 03-04. The band itself is unchanged. - 03-04 now requires re-measuring the baseline at the start of the plan rather than only "if concurrent sessions touched src-tauri/src" - the hwp-editor track is still active, so drift is expected, not exceptional - 03-04's full-gate step drops the Phase 1/2 escape hatch for uncommitted hwped.rs/lib.rs: that work is committed and make verify is green on it, so a red gate is now this phase's to diagnose - ROADMAP success criterion 4 and requirement ERR-04 cited CONCERNS.md's stale 1,118; both now cite the measured 1,138 and name the stale figure as stale (RESEARCH Pitfall 6 had already flagged it)
![chatgpt-codex-connector[bot]](https://avatars.githubusercontent.com/in/1144995?s=60&v=4){kind=small}
There was a problem hiding this comment.
💡 Codex Review
Here are some automated review suggestions for this pull request.
Reviewed commit: 2ab0a7e7ad
ℹ️ About Codex in GitHub
Your team has set up Codex to review pull requests in this repo. Reviews are triggered when you
- Open a pull request for review
- Mark a draft as ready
- Comment "@codex review".
If Codex has suggestions, it will comment; otherwise it will react with 👍.
Codex can also answer questions or update the PR. Try commenting "@codex address that feedback".
…after Two review findings on #278, both correct. P2 — the re-measure instruction landed in the wrong plan. 03-04 runs after 03-01 and 03-02 have already flipped signatures, so "re-measure the baseline at the start of this plan" observes the post-migration count (~1,128), not a baseline. Recording that as one and shifting the band would let unintended conversions pass unnoticed. The measurement moves to where the tree is still pristine: 03-01's action now opens by running the pinned command and recording the result as the phase's pre-migration baseline B, before any file is edited (03-01 flips evidence_binder_mutate, so it cannot be taken later). 03-02 records its count against B minus 10; 03-04 asserts within [B-20, B] and derives the band from B rather than re-deriving a baseline. The literal [1118, 1138] stays as the expected values, with an explicit substitution note in the automated check. P2 — 03-CONTEXT.md still prescribed 1,118 in two places while the requirement now says 1,138. The plans consume that context as implementation guidance, so the two sources were prescribing incompatible guards. Context now cites the measured baseline, and the CONCERNS.md quotation is marked stale in place (RESEARCH Pitfall 6 had already established that).
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Planning-docs only. No source changes.
The problem
Phase 3's ERR-04 gate asserts the
Result<T, String>signature count landsinside
[1118, 1138]. The 1,138 baseline was measured on a working tree thathappened to contain the then-uncommitted
hwped.rs.The number was right. Its provenance was accidental — 19 of the 1,138 matches
live in that file. On a tree without it the pinned command reads 1,119, and
after the planned 10-signature migration the count lands at 1,109, below the
band. Anyone re-deriving the baseline from a checkout that predated #277 would
have set a band Phase 3 could not satisfy.
What changed
Re-confirmed 1,138 on the committed tree at
34f96eeand recorded thatcommit as the provenance in
03-02-PLAN.mdand03-04-PLAN.md. The band itselfis unchanged.
03-04now requires re-measuring the baseline at the start of the plan,rather than only "if concurrent sessions touched
src-tauri/src". Thehwp-editor track is still active, so drift is expected, not exceptional.
03-04's full-gate step drops the Phase 1/2 escape hatch for uncommittedhwped.rs/lib.rs. That work is committed andmake verifyis green on it,so a red gate there is now Phase 3's to diagnose rather than to write off.
stale 1,118. They now cite the measured 1,138 and name the stale figure as
stale —
03-RESEARCH.mdPitfall 6 had already flagged it, but the roadmap andrequirements had never been updated to match.
row is aligned.
Verification
Measured on
34f96ee. No code touched, so no gate run is implicated.