Versioned project notes for @stackline/deepmerge 1.1.2. Original attribution is retained in the linked repository.
Benchmark Methodology
The checked-in benchmark compares @stackline/deepmerge with
[email protected] in the same Node.js process and with the same nested
configuration objects.
npm ci
npm run benchmark
Set BENCHMARK_ITERATIONS to control the sample size:
BENCHMARK_ITERATIONS=500000 npm run benchmark
The benchmark warms both implementations before measurement and reports operations per second. It is a regression tool, not a universal performance claim. CPU, Node.js version, object shape, callbacks, and enabled options materially affect results.
Recorded results
| Release | Node.js | @stackline/deepmerge |
[email protected] |
|---|---|---|---|
| 1.0.4 | 24.20.0 | 86,405 ops/s | 143,443 ops/s |
| 1.1.0 | 24.20.0 | about 200,000 ops/s | about 150,000 ops/s |
Release 1.1.0 removed the per-node weak-map memoization, the per-object key set, and per-key path allocation, and writes most keys by assignment.
Compatibility and security take priority over winning a synthetic throughput number. Report the complete command, Node.js version, CPU, and raw output when sharing results.