@jsperf.dev/core

API

Class: Benchmark

Extends: EventEmitter

new Benchmark<Context>(options)

Arguments:

• options - object - optional - Default: {}
  • warmup - boolean - optional - Default: true
  • samples - number - optional - Default: 10
  • meta - Meta - optional - Default: {}

Instance Properties

Benchmark.context

• Context

A property that will be passed as the first argument to each lifecyle method and run script.

Benchmark.meta

• Meta

A property that represents general information about the benchmark instance. The title is defaulted to process.argv[1] (generally the path of the benchmark script).

Benchmark.results

• Map<string, Array<PerformanceEntry>>

This property is updated after the benchmark suite executes.

Benchmark.samples

• number

The number of samples the benchmark suite will run. This property can be modified at any point prior to calling benchmark.start().

Benchmark.warmup

• boolean

When set to true, the warmup step will run. This property can be modified at any point prior to calling benchmark.start().

Instance Methods

Benchmark.afterAll(func)

Arguments:

• func - FunctionWithContext<Context> - required

Lifecycle method for adding a function that executes after all other scripts are executed.

Benchmark.afterEach(func)

Arguments:

• func - FunctionWithContext<Context> - required

Lifecycle method for adding a function that executes after each run script is executed.

Benchmark.beforeAll(func)

Arguments:

• func - FunctionWithContext<Context> - required

Lifecycle method for adding a function that executes before all scripts are executed.

Benchmark.beforeEach(func)

Arguments:

• func - FunctionWithContext<Context> - required

Lifecycle method for adding a function that executes before each run script is executed.

Benchmark.run(id, file)

Arguments:

• id - string - required
• file - string - required

Add a run to the benchmark instance. The id must be unique and file must be the absolute path to the script.

Benchmark.start()

Returns: Promise<void>

Executes the benchmark suite. Resolves once it completes execution. Does not rethrow errors thrown during execution. Errors are emitted through the error event.

Instance Events

start

Emitted at the beginning of the microtask queued during the constructor. It will not be emitted if no runs have been added.

end

Emitted at the end of the microtask after all runs have been executed.

error

Emitted whenever an error is thrown. Will return the thrown error.

const [error] = await once(benchmark, "error");

Type: FunctionWithContext<Context>

• (context: Context, ...extraArgs: unknown[]) => void | Promise<void>

Interface: Meta

• title - string - optional
• description - string - optional

Testing

Execute tests using pnpm test