fix(retrieve): preserve helper source when file evidence wins - #747
Conversation
|
No actionable comments were generated in the recent review. 🎉 ℹ️ Recent review info⚙️ Run configurationConfiguration used: Path: .coderabbit.yaml Review profile: CHILL Plan: Team Run ID: 📒 Files selected for processing (2)
Included review availability: 0 reviews are currently available. Your included PR review attempts over the past 7 days set your current allowance at 1 review per hour. 📝 WalkthroughWalkthrough
ChangesSource evidence preservation
Estimated code review effort: 3 (Moderate) | ~20 minutes Merge Risk: ⚪ Minimal · up to This change preserves selected helper source when file-level evidence is also included, with coverage for the relevant selection and rendering limits. No current merge-blocking risk remains. 🚥 Pre-merge checks | ✅ 4 | ❌ 1❌ Failed checks (1 warning)
✅ Passed checks (4 passed)
✨ Finishing Touches 💡 1📝 Generate docstrings 💡
🧪 Generate unit tests (beta)
Comment |
|
@coderabbitai review |
✅ Action performedReview finished.
|
|
Integration complete on |
|
Merged as |
When stronger matches elsewhere in a file win snippet selection, a selected helper can lose its implementation even though its declared source fits the excerpt limit. Preserve scored symbol evidence or a real bounded source fragment when the scored list is empty, then admit file context while keeping that fragment intact. Existing selection order, source attribution and excerpt limits remain.
Validation on the current head: 110 focused tests, six unchanged consumer reproductions, typecheck, build and a zero-vulnerability audit pass. The complete product diff is byte-identical to the previously reviewed source-preservation fix; the branch also includes the merged fast-uri dependency repair from
next.The prior exposed 72-case comparison improved raw helper-source retention from 57/108 to 108/108 and compact retention from 54/108 to 102/108, without selected-ID/order or frozen-fact regressions. That comparison has not been repeated as a fresh holdout. Compact allocation, owner-input completeness and discovery limitations remain. Refs #740.
Summary by CodeRabbit
Bug Fixes
Tests