This repository was archived by the owner on Jun 7, 2026. It is now read-only.
Fix deploy pipeline: retry artifact downloads, strict 4/4 filtering#66
Merged
Conversation
- deploy-dashboard.yml: Add 3-attempt retry with backoff on gh run download to handle transient GitHub API 502 errors during artifact pagination - csv-to-dashboard-json.cjs: Remove placeholder JSON generation for missing benchmarks — teams with incomplete traces are now excluded from rankings instead of getting artificially low scores - update-history.cjs: Include per-benchmark scores and metrics in history snapshots for per-trace trend analysis - .gitignore: Ignore artifacts/ and fuzz-perf/ directories Co-authored-by: Cursor <cursoragent@cursor.com>
📝 WalkthroughWalkthroughThe PR hardens the dashboard deployment pipeline with retry logic for artifact downloads, validates benchmark completeness across teams, and enriches history snapshots with per-benchmark score data. Incomplete teams are logged and excluded from rankings rather than filled with placeholders. ChangesDashboard Build and Artifact Processing
Possibly related PRs
Poem
Estimated code review effort🎯 2 (Simple) | ⏱️ ~12 minutes 🚥 Pre-merge checks | ✅ 4 | ❌ 1❌ Failed checks (1 warning)
✅ Passed checks (4 passed)
✏️ Tip: You can configure your own custom pre-merge checks in the settings. ✨ Finishing Touches🧪 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 |
![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
🧹 Nitpick comments (1)
scripts/csv-to-dashboard-json.cjs (1)
69-90: ⚡ Quick winConsider removing unused placeholder function.
The
makePlaceholderJsonfunction appears to be unused after removing the placeholder generation behavior. Removing it would improve code clarity.♻️ Proposed cleanup
-function makePlaceholderJson(targetName) { - return { - info: { - name: targetName, - app_version: { major: 0, minor: 0, patch: 0 }, - jam_version: { major: 0, minor: 7, patch: 2 }, - }, - stats: { - steps: 0, - imported: 0, - import_max_step: 0, - import_min: MISSING_BENCHMARK_MS, - import_max: MISSING_BENCHMARK_MS, - import_mean: MISSING_BENCHMARK_MS, - import_p50: MISSING_BENCHMARK_MS, - import_p75: MISSING_BENCHMARK_MS, - import_p90: MISSING_BENCHMARK_MS, - import_p99: MISSING_BENCHMARK_MS, - import_std_dev: 0, - }, - }; -} -Additionally,
MISSING_BENCHMARK_MS(line 37) could also be removed if no longer needed.🤖 Prompt for 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. In `@scripts/csv-to-dashboard-json.cjs` around lines 69 - 90, Remove the unused placeholder generator to clean up dead code: delete the makePlaceholderJson function and, if no other references exist, also remove the MISSING_BENCHMARK_MS constant declared earlier; search for usages of makePlaceholderJson and MISSING_BENCHMARK_MS in the file to confirm they are unused before deleting to avoid breaking behavior.
🤖 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 `@scripts/update-history.cjs`:
- Around line 65-73: The benchmarkScores serialization uses inconsistent
null-coalescing: update the metrics assignment inside the loop that builds
entry.benchmarkScores so it uses nullish coalescing like score does (replace
info.metrics || null with info.metrics ?? null) to preserve legitimate falsy
values such as 0; locate the loop where t.benchmarkScores is iterated and adjust
the metrics expression accordingly.
---
Nitpick comments:
In `@scripts/csv-to-dashboard-json.cjs`:
- Around line 69-90: Remove the unused placeholder generator to clean up dead
code: delete the makePlaceholderJson function and, if no other references exist,
also remove the MISSING_BENCHMARK_MS constant declared earlier; search for
usages of makePlaceholderJson and MISSING_BENCHMARK_MS in the file to confirm
they are unused before deleting to avoid breaking behavior.
🪄 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: Path: .coderabbit.yaml
Review profile: CHILL
Plan: Pro
Run ID: 9e304f09-f11f-4afe-943b-b686e2713d6c
📒 Files selected for processing (4)
.github/workflows/deploy-dashboard.yml.gitignorescripts/csv-to-dashboard-json.cjsscripts/update-history.cjs
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 subscribe to this conversation on GitHub.
Already have an account?
Sign in.
2 participants
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.
Summary
gh run downloadto handle transient GitHub API 502 errors during artifact pagination. With 3700+ artifacts in the repo, the API intermittently returns 502 during pagination — currently silently swallowed by2>/dev/null, causing valid artifacts to be skipped.benchmarkScores(score + metrics per trace) in history snapshots, enabling per-trace trend analysis.Problem
The current
gh run download --name <artifact> 2>/dev/null || echo "(not found, skipping)"pattern treats transient 502 errors identically to genuinely missing artifacts. This causes:csv-to-dashboard-json.cjscreates placeholder JSONs with 0 values for the "missing" artifactsgenerate-aggregated-data.jscalculates scores usinggeometricMeanwhich filters out zeros, giving teams a score based on fewer benchmarks (e.g., 3/4 instead of 4/4)Verified locally: 5 teams currently affected by false positives (artifacts exist on GitHub but were missed by the pipeline). With the retry fix, all 5 are recovered — 4 artifacts needed a second attempt.
Test plan
deploy-dashboardworkflow run after merge to verify corrected scoresMade with Cursor
Summary by CodeRabbit
Bug Fixes
Chores