Jasmine + PhantomJS

In this tutorial, we will be using PhantomJS, a headless scripted browser built on WebKit, to launch tests with the grunt-contrib-jasmine test-runner.

We first install PhantomJS.

				npm install phantomjs

Next, we install grunt-contrib-jasmine plugin locally to run Jasmine specs through PhantomJS

				npm install grunt-contrib-jasmine --save-dev

Consider a JavaScript file test.js, containing two functions, pi() and sin(). The function pi() returns the value of π (upto 3 decimal places) and the function sin() returns a numeric value between -1 and 1, for a given argument in radians 

				
				function pi() {
				  return 3.141;
				}

				function sin(x) {
				  return Math.sin(x);
				}

This file is saved inside the src directory.

We write the specs in Jasmine (below) to check that the value of π returned by pi() equals 3.141 and the value of the trigonometric ratio sine is always ≤ 1 and ≥ -1, for any random number generated between 1 and 100. We name this spec file test.spec.js and save it inside the spec directory

				
					describe("Value of pi", function() {
					  it("is equal to 3.141", function() {
					    expect(pi()).toEqual(3.141);
					  });
					});

					describe("Value of sine", function() {
					  it("is less than or equal to 1", function() {
					    expect(sin(Math.floor((Math.random() * 100) + 1)) 
					    	<= 1).toBeTruthy();
					  });

					  it("is greater than or equal to -1", function() {
					    expect(sin(Math.floor((Math.random() * 100) + 1)) 
					    	< -1).toBeFalsy();
					  });
					});

And finally, Gruntfile.js is created and configured as shown below

				
				module.exports = function(grunt) {
					grunt.initConfig({
						pkg: grunt.file.readJSON('package.json'),

				        jasmine: { 
				        	test: {           
				                src:'src/test.js',
				                options: {                   
				                    'specs': 'spec/test.spec.js'
				                }
				            }           
				        }
						
					});	

					grunt.loadNpmTasks('grunt-contrib-jasmine');
					grunt.registerTask('default', ['jasmine']);
				}

Jasmine specs can now be run typing either the command 

				grunt jasmine

or

				grunt