Repository navigation
Unblock the current BAML language release - #4215
Conversation
|
The latest updates on your projects. Learn more about Vercel for GitHub.
|
|
No actionable comments were generated in the recent review. 🎉 ℹ️ Recent review info⚙️ Run configurationConfiguration used: Repository UI Review profile: CHILL Plan: Pro Plus Run ID: 📒 Files selected for processing (1)
🚧 Files skipped from review as they are similar to previous changes (1)
📝 WalkthroughWalkthroughThe NuGet publisher now sources its username from a repository variable, with contract coverage updated accordingly. The C# cancellation test now polls for asynchronous cancellation using a timeout helper. ChangesRelease credential wiring
C# cancellation validation
Estimated code review effort: 2 (Simple) | ~10 minutes Possibly related PRs
Poem
🚥 Pre-merge checks | ✅ 5✅ Passed checks (5 passed)
✨ Finishing Touches📝 Generate docstrings
🧪 Generate unit tests (beta)
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 |
⏭️ Performance benchmarks were skippedPerf benchmarks (CodSpeed) are opt-in on pull requests — they no longer run on every push. They always run automatically after merge to To run them on this PR, do any of the following, then push a commit (or re-run CI):
|
![coderabbitai[bot]](https://avatars.githubusercontent.com/in/347564?s=60&v=4){kind=small}
There was a problem hiding this comment.
Actionable comments posted: 1
🤖 Prompt for all review comments with AI agents
Verify each finding against current code. Fix only still-valid issues, skip the
rest with a brief reason, keep changes minimal, and validate.
Inline comments:
In `@baml_language/sdk_tests/crates/csharp/phase12_resources/Program.cs`:
- Around line 545-561: Update RequireEventuallyAsync so each loop checks elapsed
time before invoking condition(), preventing polls after timeout. Compute the
remaining timeout and bound Task.Delay to that duration instead of always
waiting 10 ms; after the bounded wait, throw InvalidOperationException only when
the timeout has actually been exceeded.
🪄 Autofix (Beta)
Fix all unresolved CodeRabbit comments on this PR:
- Push a commit to this branch (recommended)
- Create a new PR with the fixes
ℹ️ Review info
⚙️ Run configuration
Configuration used: Repository UI
Review profile: CHILL
Plan: Pro Plus
Run ID: 6e94fef7-9348-4b31-a291-a5e24f50cea1
⛔ Files ignored due to path filters (1)
pnpm-lock.yamlis excluded by!**/pnpm-lock.yaml
📒 Files selected for processing (3)
.github/workflows/publish2-csharp-sdk.yamlbaml_language/sdk_tests/crates/csharp/phase12_resources/Program.csscripts/tests/test_release_pipeline_contract.py
Binary size checks passed✅ 7 passed
Generated by |
Summary
CancelToken.anyconsumer assert its asynchronous propagation with a bounded wait instead of a same-tick racevars.NUGET_USERinstead of the environment secret that repeatedly resolved emptyRoot causes
Current canary CI could not attest a release because
typescript/apps/beps/package.jsonaddedmermaidwithout updatingpnpm-lock.yaml. Independently, the C# macOS Phase 12 consumer cancelled an input token and immediately sampled the asynchronously linked composite token; macOS repeatedly exposed that scheduling race.The two latest properly attested releases both reached NuGet with no
userinput even though theNUGET_USERenvironment secret existed and had been refreshed. The NuGet profile name is public configuration, so this PR reads the existingboundary-tools-prodenvironment variable instead.vars.NUGET_USER=boundarymlis configured on that existing environment.Impact
A green canary CI can once again create the immutable source tag and dispatch the language release. The C# test retains a strict one-second propagation bound, and NuGet OIDC can receive the required profile name without depending on secret delivery.
Validation
corepack pnpm install --frozen-lockfileactionlintonpublish2-csharp-sdk.yamlgit diff --checkmacOS CI remains the decisive proof for the scheduling fix; the subsequent production release remains the decisive proof for NuGet trusted publishing.
Summary by CodeRabbit
Bug Fixes
Tests