https://github.com/codemix/bencha

Mocha-esque UI for the excellent benchmarkjs benchmarking library
https://github.com/codemix/bencha

Last synced: 3 months ago
JSON representation

Mocha-esque UI for the excellent benchmarkjs benchmarking library

• Host: GitHub
• URL: https://github.com/codemix/bencha
• Owner: codemix
• Created: 2013-04-26T14:37:40.000Z (almost 13 years ago)
• Default Branch: master
• Last Pushed: 2013-04-26T14:48:09.000Z (almost 13 years ago)
• Last Synced: 2025-01-20T01:35:17.262Z (about 1 year ago)
• Language: CoffeeScript
• Size: 125 KB
• Stars: 1
• Watchers: 2
• Forks: 0
• Open Issues: 1
• Metadata Files:
  • Readme: README.md

Awesome Lists containing this project

README

          
# Bencha

Mocha-esque UI for the excellent `benchmark.js` JavaScript benchmarking library.

Records benchmark history and flags performance improvements and regressions.

# Installation

    npm install bencha -g

# Writing Benchmarks

Create a `bench` directory in your project. In the `bench` directory, create

an `index.js` (or `index.coffee`), e.g.

    compare("RegExp vs String::indexOf", function(){

      var input = "demo.string"

      benchmark("RegExp", function(){

        /(\.)/.test(input);

      });

      benchmark("String::indexOf", function(){

        input.indexOf(".") > -1;

      });

    });

    suite("My Feature", function(){

      benchmark("foo() no arguments", function(){

        foo();

      });

      benchmark("foo() with arguments", function(){

        foo(true, false);

      });

      benchmark("bar()", function(){

        bar()

      });

    });

The above defines two benchmark suites, the first is a comparison suite, comparison suites compare

the results of each benchmark within the suite, declaring one the winner. The second declaration defines

a suite containing three separate benchmarks that are related but not explicitly compared with each other.

To split your benchmarks into seperate files, simply `require()` them from the main `index.js` file.

# Running Benchmarks

From your project root (or the directory containing the `bench` folder), run:

    bencha

This will run the benchmarks with the default `spec` reporter, recording the results in `bench/history.json`

Run `bencha --help` for more options.

# Licence

MIT