[patch] Write benchmark results where the workflow looks for them - #73
Merged
Merged
Conversation
The Benchmarks workflow has never published an artifact. BenchmarkDotNet writes its reports below the working directory, and `dotnet run --project` leaves that at the repository root rather than moving into the project, so the reports landed in ./BenchmarkDotNet.Artifacts/results while the summary and upload steps looked in PreciseNumber.Benchmarks/BenchmarkDotNet.Artifacts/results. All four runs so far succeeded and uploaded nothing; the most recent spent thirteen minutes benchmarking and ended with "No files were found with the provided path". The run step now passes --artifacts explicitly, so the location is stated rather than inherited from whatever the working directory happens to be, and the summary and upload steps read that same path. if-no-files-found is also raised from warn to error. The upload step warning was the only sign this was broken, and a warning on a green run is what let it go unnoticed across four runs. Co-Authored-By: Claude Opus 5 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_017jrnV7N94UGL8fDRRE8Xt8
|
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.



The Benchmarks workflow has never published an artifact.
BenchmarkDotNet writes its reports below the working directory, and
dotnet run --projectleaves that at the repository root rather than moving into the project. So the reports landed in./BenchmarkDotNet.Artifacts/resultswhile the summary and upload steps looked inPreciseNumber.Benchmarks/BenchmarkDotNet.Artifacts/results.All four runs of the workflow so far are green and uploaded nothing. Run 34747735864 spent thirteen minutes benchmarking and ended with:
and the API reports
0artifacts for that run.The change
The run step now passes
--artifactsexplicitly, so the location is stated rather than inherited from whatever the working directory happens to be, and the summary and upload steps read that same path.if-no-files-foundis also raised fromwarntoerror. That upload warning was the only sign this was broken, and a warning on an otherwise green run is what let it go unnoticed across four runs. This is slightly beyond the path fix, and it is the part that stops the failure mode from being silent again — worth calling out in case you would rather keep it atwarn.Verified
Both behaviours were measured in this session rather than assumed:
./BenchmarkDotNet.Artifacts/results/…-report-github.md, confirming the mismatch.--artifacts DIR, reports land atDIR/results/…-report-github.mdand nothing is written at the repository root.-report-github.md,-report-full.json,-report.csv,-report.html) — the JSON being what the existing comment says a later comparison would be built from.No library code is touched, so nothing here affects the published package.
🤖 Generated with Claude Code
https://claude.ai/code/session_017jrnV7N94UGL8fDRRE8Xt8
Generated by Claude Code