Skip to documentation

reference

Benchmark Methodology — @stackline/stable-stringify

Project benchmarks notes for @stackline/stable-stringify 1.0.6. Review behavior, requirements and limitations.

Versioned project notes for @stackline/stable-stringify 1.0.6. Original attribution is retained in the linked repository.

Benchmark Methodology

The benchmark runs native JSON.stringify, [email protected], stable mode, safe mode, and RFC 8785 mode against the same representative payload in one Node.js process.

npm ci
npm run benchmark

Control the sample size with BENCHMARK_ITERATIONS:

BENCHMARK_ITERATIONS=250000 npm run benchmark

Native JSON is a throughput baseline only. It does not recursively sort object keys and is not behaviorally equivalent. Stable, safe, and canonical modes also implement different contracts, so compare a mode only with alternatives that provide the same guarantees.

When publishing results, include the command, Node.js version, CPU, complete raw output, and payload shape. Do not present one machine's result as a universal ranking.

Package reference