Repository navigation
ci(attach): run real-daemon replay regression - #31
Merged
Merged
Conversation
Kody Review CompleteGreat news! 🎉 Keep up the excellent work! 🚀 Kody Guide: Usage and ConfigurationInteracting with Kody
Current Kody ConfigurationReview OptionsThe following review options are enabled or disabled:
|
External contributor checklistThanks for contributing. A maintainer will handle the trusted checks and final merge.
|
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Why
Catch attach replay regressions in hosted Linux CI before they reach main.
What's Included
UI / UX
Not applicable. This changes CI and contributor guidance.
Notes
Models: GPT-6.1 Sol
Merge risk: low. The two-file change passes local workflow checks and hosted Linux CI, including all 96 attach cases. Recovery: revert this workflow change; completed runs keep their results.
Local checks pass:
git diff --check, actionlint, and all 6 PR automation tests. Before this two-file change, a clean macOS checkout of mainb8eb0f6passed Zig 0.16 formatting, unit tests, build, replay tests, and all 96 attach cases. Hosted main CI also passed. Hosted Linux CI run 42342847334 passed at2f2e47e, including the new Attach regression step with 96/96 cases.Open in [code]smith · Need help on this PR? Tag
@codesmith-botwith what you need.Summary
CI now runs the real-daemon attach/replay regression that previously existed only as a manual, documented step. The attach regression is promoted from "run this locally when you touch attach code" to a blocking CI check on Linux, and
CONTRIBUTING.mdis updated to match the new CI reality.Changes
.github/workflows/ci.ymlAttach regressionstep to the Linux job:zig build && zig build test-replay && python3 test/attach_scrollback.py zig-out/bin/zmxzmxbinary againsttest/attach_scrollback.py, so scrollback/attach behavior regressions are caught in CI rather than only during manual verification.CONTRIBUTING.mdSecurity Impact
None. No production code, input handling, or trust boundaries are touched. The only new execution is a CI step invoking the project's own test binary and a repo-local Python script.
Performance Impact
Review Callouts
zig buildimmediately after theUnit testsstep (zig build test), which already compiles the artifactstest-replayand the Python harness consume. Dropping the leadingzig buildwould shorten CI without changing coverage — worth confirming whether it was added deliberately for thezmxbinary specifically.ci.ymlandCONTRIBUTING.md. Extracting it into a single script (e.g.test/attach_regression.sh) would keep the docs and CI from drifting, and gives contributors one entry point to run.Verification
zig build test, then the new CI step locally:zig build && zig build test-replay && python3 test/attach_scrollback.py zig-out/bin/zmx.