[ci] refresh build workflow: live runners, floor+latest Go matrix, decoupled CLI job#129
Conversation
…n_test - gofmt: reflow doc comments to current gofmt output (comment-only; dataconv/interface.go, internal/replacecr/replace_cr.go, lib/http/server.go) - run_test.go: capture and defer the cancel functions returned by context.WithTimeout (go vet lostcancel) No behavior change; this makes the tree pass the local fmt/vet bar that the revived CI series will hold. Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
…coupled CLI job The workflow was effectively dead: ubuntu-20.04 and macos-13 runners have been retired by GitHub, so every run queued for 24h and timed out, and the coverage upload gate was bound to the retired runner so it never executed. - OS matrix: ubuntu-22.04 / macos-14 / windows-2022 - Go matrix: 1.18.x (module floor) + 1.25.x (latest); floor bump is deferred to the pin-upgrade PR at the end of this series - actions: checkout@v5, setup-go@v6, codecov-action@v5 (with the coverage gate semantics documented inline) - drop the 'export' debug step (leaked the environment into public logs) - cmd/starlet is a separate Go module with its own pins: moved out of 'make ci' into a dedicated non-gating job Refs: ENG-05 Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
Up to standards ✅🟢 Issues
|
| Metric | Results |
|---|---|
| Duplication | 0 |
🟢 Coverage ∅ diff coverage
Metric Results Coverage variation Report missing for bb483ca1 Diff coverage ✅ ∅ diff coverage Coverage variation details
Coverable lines Covered lines Coverage Common ancestor commit (bb483ca) Report Missing Report Missing Report Missing Head commit (8c03e51) 5645 5276 93.46% Coverage variation is the difference between the coverage for the head and common ancestor commits of the pull request branch:
<coverage of head commit> - <coverage of common ancestor commit>Diff coverage details
Coverable lines Covered lines Diff coverage Pull request (#129) 0 0 ∅ (not applicable) Diff coverage is the percentage of lines that are covered by tests out of the coverable lines that the pull request added or modified:
<covered lines added or modified>/<coverable lines added or modified> * 100%1 Codacy didn't receive coverage data for the commit, or there was an error processing the received data. Check your integration for errors and validate that your coverage setup is correct.
NEW Get contextual insights on your PRs based on Codacy's metrics, along with PR and Jira context, without leaving GitHub. Enable AI reviewer
TIP This summary will be updated as you push new changes.
Codecov Report✅ All modified and coverable lines are covered by tests. Additional details and impacted files@@ Coverage Diff @@
## master #129 +/- ##
==========================================
+ Coverage 91.19% 91.97% +0.78%
==========================================
Files 44 44
Lines 5494 4509 -985
==========================================
- Hits 5010 4147 -863
+ Misses 363 241 -122
Partials 121 121 ☔ View full report in Codecov by Harness. 🚀 New features to boost your workflow:
|
1 participant
Why
CI on this repo is effectively dead: the
ubuntu-20.04andmacos-13runners have been retired by GitHub, so every workflow run queues for 24 hours and times out (see the run on #128:failureafter 24h0m3s). The coverage upload gate was bound to the retiredubuntu-20.04runner, so coverage never uploaded either.What
ubuntu-22.04/macos-14/windows-2022(live runners only)1.18.x(module floor) +1.25.x(latest stable), replacing the six-version ladder. The floor bump to 1.19 is deliberately deferred to the dependency pin-upgrade PR at the end of this fix series — the currentgo.starlark.netpin is consistent with go 1.18.checkout@v5,setup-go@v6,codecov-action@v5. The coverage gate semantics are documented inline: the gate is thecodecov/project+codecov/patchcommit statuses,continue-on-erroron the upload step only tolerates upload-channel outages.exportdebug step which dumped the full environment into public logs.cmd/starletdecoupled: it is a separate Go module with its own (older) pins and extra third-party dependencies; it now builds in a dedicated non-gating job instead of insidemake ci, so a CLI-side breakage is visible without blocking library PRs.lostcancelvet fixes inrun_test.go. No behavior change.Refs: ENG-05
🤖 Generated with Claude Code