pp-list-plus-rebind — xs = xs + [i] vs append
Claim
Measured on this machine: primary hot case list_plus_rebind ratio before/after ~1872.93× (faster after). Default severity stays hint. Do not escalate to error from this fill alone — ratio is strong but policy/default and G3 OK-idiom history keep severity unchanged.
How measured
- Harness:
packages/pyperf/examples/bench/evidence_pp.py - Protocol: warm=3; median of 9 rounds; GC/collect where applicable
- Raw lines:
pp-list-plus-rebind.jsonl
Runtime (this evidence file)
| Field | Value |
|---|---|
| When | 2026-10-23 Asia/Shanghai (machine clock; JSONL ts is UTC) |
| Toolchain | CPython 3.15.1 |
Median after (list_append) | 0.302584 ms (ops=20000, ratio=1.000000) |
Median before (list_plus_rebind) | 566.717875 ms (ops=20000, ratio=1872.927988) |
Median after (small_append) | 0.006208 ms (ops=100, ratio=1.000000) |
Median before (small_list_plus) | 0.013167 ms (ops=100, ratio=2.120912) |
| Primary ratio before/after | ~1872.93× (list_plus_rebind) |
Direction and absolute ms are N/machine/opt-sensitive. Re-run the harness after toolchain or machine changes.
Lint mapping
pyperf-lint pp-list-plus-rebind default stays hint. Escalate recommended: no (strong microbench, keep current default; escalate needs policy candidate + corpus agreement).
Limits
- Microbench; not an app profile.
- Small-N rows may sit near timer noise; prefer hot (non-small) cases for ratio claims.
- I/O rules (print/echo/puts/writeln) discard to
/dev/null— env-dependent.