Repository navigation
docs: record the declined modernizer analyzer in both guidance files - #451
Conversation
The Go 1.27 modernizer pass here rejected the embedlit analyzer; the SDK's pass applied it. The same question went on record with two different answers, and undoing it cost 43 hunks in the SDK. Nothing mechanical objects to embedlit -- it is behaviour-identical and gorelease is indifferent -- so the only thing that can keep the two repositories agreeing is a written reason where the next run will look. Running `go fix ./...` without the flag proposes all 119 hunks here again, silently. Both CLAUDE.md and AGENTS.md carry it, since they are maintained as copies of each other, and bomly-sdk's AGENTS.md carries the matching note. [skip release] Co-Authored-By: Claude Opus 5 <[email protected]>
|
Warning Review limit reachedNext included review available in 49 minutes. View limit detailsLimit details: You’ve used the included review currently available. You've used all free OSS reviews for now. Wait for the free limit to reset to keep reviewing this public repository. Review configuration: ⚙️ Run configurationConfiguration used: Path: .coderabbit.yaml Review profile: CHILL Plan: Advanced Run ID: 📒 Files selected for processing (2)
Thanks for using CodeRabbit! It's free for OSS, and your support helps us grow. If you like it, consider giving us a shout-out. Comment |
Bomly Diff SummaryCompared Overview
Dependency Changes✅ No dependency changes. Vulnerabilities✅ No vulnerability changes. License Changes✅ No license changes. Project Posture✅ No project posture changes ( Policy Findings✅ No policy differences were identified. |
![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: e4d2a5d7ae
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".
Codex checked the merged pass and found my note wrong. #448 ran `go fix -embedlit=false -omitzero=false -stringsbuilder=false ./...` and I recorded one of the three, so an agent following the note would have reintroduced the other two -- a note that is worse than none, because it looks authoritative. omitzero is the one that matters. It drops omitempty from struct-valued JSON fields in golden- and schema-backed output types. The bytes do not move, so no test, linter or API gate objects -- and the tag is the published schema, so a consumer generating one by reflection starts reading the field as required. The SDK took that rewrite in its own pass and spent four review rounds undoing the consequences. The same analyzer was one command away from doing it here. Also says what `go fix` does: it applies in place, and `-diff` prints instead. A reader who assumed it previewed would find out afterwards. Co-Authored-By: Claude Opus 5 <[email protected]>
1 participant
The CLI half of "the two repos should agree on
embedlit". The SDK half is bomly-dev/bomly-sdk#82, which reverts 43 hunks.Why a doc change is the deliverable
The revert makes the two repositories agree today. Nothing keeps them agreeing tomorrow:
embedlitis behaviour-identical, so tests do not object.goreleaseis indifferent, so the API gate does not object.The next
go fix ./...in either repo proposes every hunk again — 119 here, 43 there — and whoever runs it has no way to know the question was already settled.What it says
The flag (
go fix -embedlit=false ./...) and the reason:Coordinatesis a named identity concept under ADR-0041 anddev-docs/MODELS.md, and the wrapper at a construction site is what keeps the identity visible where a package is built.It also records that the disagreement already happened once and what it cost, so the note reads as a decision with a price rather than a preference.
Both
CLAUDE.mdandAGENTS.mdget it, since this repo maintains them as copies.make verifygreen.[skip release]— documentation only.🤖 Generated with Claude Code