Repository navigation
release: idryx 1.2.1 - #104
Merged
Merged
Conversation
…known Release notes for v1.2.1: #103 (the stderr line names event types and sources no detector reads, instead of calling them unknown), the teeth case in 4403c42, and the CI action pins in dfe360e. No code change in this commit. A patch, not a minor: nothing is read, emitted or refused differently, and no name in compat/1.0.json moved. Co-Authored-By: Claude Haiku 5.5 <noreply@anthropic.com>
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Release prep for idryx 1.2.1:
docs/releases/v1.2.1.mdonly. No code change in this PR.What and why
knownTypesandevents.consumedare unchanged.release.ymlandscorecard.yml).compat/1.0.jsonmoved; the stderr wording is not in the frozen surface. The README badge (381) was already updated by fix(ingest): name the event types and sources no detector reads, not unknown #103.Tier: T0 (release notes; no code, no new test). The red-first step does not apply here; #103 carries its own tests, and I did not re-run them against the unfixed code. Coverage is not reported because no code changed.
Gates run locally
gofmt -l .: emptygo vet ./...: okstaticcheck ./...(GOTOOLCHAIN=go1.27.1, as CI does): run withGOOS=linuxit is clean. Run on darwin it reportsconst procPidInitIno is unusedininternal/ebpfcapture/self.go, because the only use is in the linux-onlycapture_linux.go. That is a darwin-only false positive, present onmainas well, and CI runs on Linux.go test -race -count=1 ./...: all 20 packages with tests okgosec -quiet ./...: exit 0 (gosecv2.29.0 is the CI pin; the installed binary reportsScannerfrom the same family, not re-pinned here)govulncheck ./...: No vulnerabilities foundscripts/compat-surface.sh: 29 frozen names present,COMPATIBILITY.mdis their renderingscripts/detectors-complete.sh: 27 detectors, each registered and testedscripts/diagrams-match-detectors.sh: exit 0scripts/features-are-bound.sh: 17 scenarios, 17 bindings, 0 brokenscripts/ebpf-optional.sh: exit 0scripts/readme-numbers.sh: 381 test functions, badge says 381scripts/reproducible-build.sh: reproducible, digest3e7339c5457e2090185758b4a7bedc318b7448d667602819307db8a6c300a4fascripts/gates-have-teeth.sh: 24 cases, every gate fails on its own fault and passes on a non-fault; tree clean afterwardspreflight.sh pr: every check okNot proven
object-matches-its-source.sh, the BPF teeth cases and the integration job run only in CI's Linux jobs and were not run here.🤖 Generated with Claude Code