Breakwater

Backtest / #1029

perf: use fullmatch for markers too

pypa/packaging#1029 · merged 2026-01-04 · base c39b6da9ab → head 8a805e3baa · ran on Docker

PASS

No behaviour divergence found by any probe.

Performance · Marker

No realistic input family measured slower.

Measured in backtest packaging-v2 (same commits, run bt-1029-8689c7, image bw-pair:d03dcb2e7566ace6). The semantic results above are from a later run; perf doesn't use the semantic planner, so it was not re-measured.

head slowerhead faster1×1001,00010,000input size (characters)F1F1: 1.00× at ~42 charsF1: 1.00× at ~84 charsF1: 1.00× at ~342 charsF1: 1.00× at ~1,398 charsF1: 1.00× at ~5,778 charsF1: 1.00× at ~23,466 charsF1: 1.00× at ~97,194 chars
  • F1
Head/base cachegrind instructions per call, by input size (log-log). Each line is one input family Nemotron proposed; points above 1× are where the "faster" code does more work.
FamilyInputsRatio range Slope base → headVerdict
F1many short extra clauses combined with 'or'claimed speedup · plan1.00× – 1.00×0.99 → 0.99neutral

Receipt

TargetBackendImageRun AuditAdjudication
packaging.markers:Markerlocal Dockerbw-pair:8f442c507f13d5d7bt-1029-95d888$0.018$0.000

Nemotron cost on Nebius Token Factory. Backend: where the untrusted code ran. local Docker (an image cited by its content-addressed tag). Every image is content-addressed (repo at each commit + harness), so every card re-runs from the image and seed 0. Raw data: audit.