Functional correctness
How completely the refactor preserves both public functions' observable behavior.
Clean-room nomination · Refactoring
Refactoring test: extract shared event filtering
This test measures whether a model can improve internal structure exactly as requested while preserving already-correct observable behavior.
Nomination prepared · results pendingPublic testsbehavior preservation
Objective evaluatorstructure and behavior
4 criteriaseparate blind review
1 source filestrict change boundary
Test assignment
Share validation and filtering without changing outputs
The existing event-summary functions already behave correctly. The model must remove duplicated selection logic by introducing exactly one required private helper and routing both exports through it.
What the refactor must preserve and change
- Keep
countByTypeandlatestByTypeas the only exports. - Add exactly one non-exported helper named
matchingValidEvents(events, type). - Use that helper from both public functions for validation and event selection.
- Count only valid matching events and select the latest timestamp, keeping the first event when timestamps tie.
- Return the documented fallback for invalid top-level input and never mutate the input array or events.
How each reviewed solution will be scored
Each criterion receives 0–10. Public and objective checks remain separate from the manual score.
Reliability / edge cases
How safely it preserves invalid-input, timestamp-tie, and immutability behavior.
Maintainability / clarity
How well the shared helper removes duplication and makes the module easier to understand.
Scope discipline
How precisely it follows the required helper name, export boundary, and allowed file.
Overall is the arithmetic mean of the four scores.
Results are not published yet
This page defines the test before model execution. The future table will show public and objective results, four review scores, time, cost, and runtime route.