feat: configurable high-iteration benchmarks with env var overrides

[Copilot]

Benchmark iteration counts were hardcoded, making it impossible to run at higher fidelity without editing source. This adds runtime control via env vars and a dedicated high-iteration npm script.

Changes

• benchmarks/logger.bench.ts — replaces hardcoded BenchmarkConfig constants with makeConfig() that reads ZARIO_BENCH_ITERATIONS and ZARIO_BENCH_SAMPLES at runtime; warmup auto-scales to 20% of iterations
• package.json — adds bench and bench:high scripts; adds tsx to devDependencies
• benchmarks/RESULTS.md — captures results from the 500k-iteration run (8 samples)

Usage

# Standard defaults (50k–200k iter/sample, 6 samples)
npm run bench

# High-fidelity preset (500k iter/sample, 8 samples)
npm run bench:high

# Fully custom
ZARIO_BENCH_ITERATIONS=1000000 ZARIO_BENCH_SAMPLES=10 npm run bench

High-iteration results snapshot (500k/sample, 8 samples)

Benchmark	Median ops/sec	Median ns/op
Simple message (sync)	11,552,893	87
With metadata (sync)	10,572,753	95
Deep metadata (sync)	6,687,858	150
Filtered (early exit)	~1,007,000,000	1
Simple JSON log	11,568,692	86
Async enqueue	5,079,084	197
Child logger	10,453,676	96

[Copilot]

Pull request overview

Adds runtime configurability to the logger benchmark suite so benchmark fidelity can be increased without editing source, and provides a high-iteration preset script for repeatable runs.

Changes:

• Introduces env-var driven benchmark configs (ZARIO_BENCH_ITERATIONS, ZARIO_BENCH_SAMPLES) with a shared makeConfig() helper and warmup scaling.
• Adds bench and bench:high npm scripts and includes tsx to run the TypeScript benchmark file directly.
• Adds a benchmarks/RESULTS.md snapshot for the high-iteration preset run.

Reviewed changes

Copilot reviewed 3 out of 4 changed files in this pull request and generated 2 comments.

File	Description
package.json	Adds benchmark scripts and tsx dev dependency to run TS benchmarks via Node.
package-lock.json	Updates lockfile to include tsx and its dependency tree.
benchmarks/logger.bench.ts	Replaces hardcoded configs with env-var override support and warmup scaling logic.
benchmarks/RESULTS.md	Documents a high-iteration benchmark run output and how to run benchmarks.

---

💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.

[Copilot]

The benchmark config reads iteration/sample counts directly via Number(...) and only checks > 0. This allows non-integers (e.g. 0.5), Infinity (e.g. 1e309/Infinity), or very large values, which can lead to misleading metrics (division by non-integer) or effectively infinite loops/hangs. Consider parsing as an integer and validating with Number.isFinite + Number.isInteger, and either clamping or throwing a clear error when invalid/out-of-range values are provided.

Suggested change

	const BENCH_ITERATIONS = Number(process.env["ZARIO_BENCH_ITERATIONS"] ?? 0);
	const BENCH_SAMPLES = Number(process.env["ZARIO_BENCH_SAMPLES"] ?? 0);
	function makeConfig(baseIterations: number, baseWarmup: number): BenchmarkConfig {
	const iterations = BENCH_ITERATIONS > 0 ? BENCH_ITERATIONS : baseIterations;
	const warmupIterations = BENCH_ITERATIONS > 0 ? Math.floor(iterations * 0.2) : baseWarmup;
	const samples = BENCH_SAMPLES > 0 ? BENCH_SAMPLES : 6;
	function parsePositiveIntegerEnv(name: string): number | undefined {
	const rawValue = process.env[name];
	if (rawValue == null || rawValue.trim() === "") {
	return undefined;
	}
	const parsedValue = Number(rawValue);
	if (!Number.isFinite(parsedValue) || !Number.isInteger(parsedValue) || parsedValue <= 0) {
	throw new Error(
	`Invalid ${name}: expected a finite positive integer, received "${rawValue}".`,
	);
	}
	return parsedValue;
	}
	const BENCH_ITERATIONS = parsePositiveIntegerEnv("ZARIO_BENCH_ITERATIONS");
	const BENCH_SAMPLES = parsePositiveIntegerEnv("ZARIO_BENCH_SAMPLES");
	function makeConfig(baseIterations: number, baseWarmup: number): BenchmarkConfig {
	const iterations = BENCH_ITERATIONS ?? baseIterations;
	const warmupIterations = BENCH_ITERATIONS != null ? Math.floor(iterations * 0.2) : baseWarmup;
	const samples = BENCH_SAMPLES ?? 6;

[Copilot]

The bench:high script sets env vars using POSIX inline assignment (ZARIO_BENCH_ITERATIONS=...). This won’t work in Windows cmd.exe (and differs in PowerShell), so the benchmark scripts become shell-dependent. If cross-platform support is desired, use a cross-platform env setter (e.g., cross-env) or move the preset logic into the benchmark script itself (e.g., via CLI args).