Skip to content

Commit e7d3a51

Browse files
committed
docs: drop machine hostname from benchmarks and notes
Stop hardcoding host name in README generator and local review/PERF docs. Agent: Grok 4.5 (xAI).
1 parent 2f325c1 commit e7d3a51

4 files changed

Lines changed: 4 additions & 4 deletions

File tree

‎PERF_2026-07-17.md‎

Lines changed: 1 addition & 1 deletion
Original file line numberDiff line numberDiff line change
@@ -1,7 +1,7 @@
11
# Performance: before/after the 2026-07-17 review fixes
22

33
Author: Claude Code (agent), model `claude-fable-5`; implementation by a Sonnet subagent,
4-
measured on `bleepblop` against `benchmark_data/SwissProt.xml` (110 MB) and the synthetic
4+
measured against `benchmark_data/SwissProt.xml` (110 MB) and the synthetic
55
`benchmark.py` corpus. Raw numbers persisted in `benchmark_data/benchmark_results.json`.
66

77
## The headline finding: everyone was benchmarking a debug build

‎README.md‎

Lines changed: 1 addition & 1 deletion
Original file line numberDiff line numberDiff line change
@@ -39,7 +39,7 @@ From a clone: `make develop` (release extension; needed for honest benches/tests
3939
<!-- BEGIN BENCHMARKS -->
4040
Release builds only (`make develop` / `make build`). Debug extensions are ~9× slower.
4141

42-
Numbers: **2026-07-17**, machine `bleepblop`. Source of truth: [`benchmark_data/benchmark_results.json`](benchmark_data/benchmark_results.json) (regenerate this section with `make readme-benchmarks`). Narrative: [`PERF_2026-07-17.md`](PERF_2026-07-17.md).
42+
Numbers: **2026-07-17**. Source of truth: [`benchmark_data/benchmark_results.json`](benchmark_data/benchmark_results.json) (regenerate this section with `make readme-benchmarks`). Narrative: [`PERF_2026-07-17.md`](PERF_2026-07-17.md).
4343

4444
### Full-document dict — `xml_to_dict` vs `xmltodict.parse`
4545

‎REVIEW_2026-07-17.md‎

Lines changed: 1 addition & 1 deletion
Original file line numberDiff line numberDiff line change
@@ -1,7 +1,7 @@
11
# xml_iterator review — 2026-07-17
22

33
Reviewer: Claude Code (agent), model `claude-fable-5` (Fable 5).
4-
All bugs below were reproduced against the currently built extension; timings measured on this machine (`bleepblop`) using the cached `benchmark_data/SwissProt.xml` (110 MB).
4+
All bugs below were reproduced against the currently built extension; timings measured using the cached `benchmark_data/SwissProt.xml` (110 MB).
55

66
## TL;DR
77

‎bench_common.py‎

Lines changed: 1 addition & 1 deletion
Original file line numberDiff line numberDiff line change
@@ -392,7 +392,7 @@ def render_benchmarks_section(data: Optional[Dict[str, Any]] = None) -> str:
392392
lines: List[str] = [
393393
"Release builds only (`make develop` / `make build`). Debug extensions are ~9× slower.",
394394
"",
395-
f"Numbers: **{stamp.split()[0] if stamp else 'unknown'}**, machine `bleepblop`. "
395+
f"Numbers: **{stamp.split()[0] if stamp else 'unknown'}**. "
396396
f"Source of truth: [`benchmark_data/benchmark_results.json`](benchmark_data/benchmark_results.json) "
397397
f"(regenerate this section with `make readme-benchmarks`). "
398398
f"Narrative: [`PERF_2026-07-17.md`](PERF_2026-07-17.md).",

0 commit comments

Comments
 (0)