intestine

Intestine provides the guts of a unit testing framework.

With intestine it's not hard at all to roll your own testing framework in whatever style you want with a pluggable runner so you can easily swap in code coverage or stackedy stack-traces.

example

Just whip up your own test handler...

examples/t/test.js:

var EventEmitter = require('events').EventEmitter;
 
module.exports = Test;
 
function Test (name) {
    this.name = name;
}
 
Test.prototype = new EventEmitter;
 
Test.prototype.ok = function (value, name) {
    this.emit('assert', {
        type : 'ok',
        ok : Boolean(value),
        name : name,
        found : value,
        wanted : true
    });
};
 
Test.prototype.equal = function (found, wanted, name) {
    this.emit('assert', {
        type : 'equal',
        ok : found == wanted,
        name : name,
        found : found,
        wanted : wanted
    });
};
 
Test.prototype.end = function () {
    this.emit('end');
};

Then whip up a few test files:

example/t/first.js:

test('foo', function (t) {
    t.equal(1 + 2, 3);
    t.end();
});
 
test('bar', function (t) {
    t.ok('', 'this is going to fail');
    t.end();
});

example/t/second.js:

test('baz', function (t) {
    t.ok('beep boop', 'non-falsy');
    t.end();
});

And now load the files with intestine(), injecting test() into each test's context. You can inject whatever you please into the runner, like a module.exports for nodeunit/expresso style tests, or a require() that intercepts custom test names.

example/t.js

var intestine = require('intestine')
var Test = require('./t/test');
 
var guts = intestine(function (runner) {
    runner.context.test = function (name, cb) {
        var t = new Test(name);
        runner.start(t);
        cb(t);
    };
});
 
var fs = require('fs');
guts.append(fs.readFileSync(__dirname + '/t/first.js'));
guts.append(fs.readFileSync(__dirname + '/t/second.js'));
 
var counts = { pass : 0, fail : 0, total : 0 };
 
guts.on('assert', function (res) {
    if (res.ok) {
        console.log('PASS: ' + res.name);
        counts.pass ++;
    }
    else {
        console.log('FAIL: ' + res.name);
        counts.fail ++;
    }
    counts.total ++;
});
 
guts.on('error', function (err) {
    console.error(err.message);
});
 
guts.on('end', function () {
    var percent = Math.floor(counts.pass / counts.total * 100);
    console.log('\n' + percent + '% OF TESTS PASSED');
});
 
guts.run();

output:

PASS: undefined
FAIL: this is going to fail
PASS: non-falsy

66% OF TESTS PASSED

Hooray, it works!

methods

var guts = intestine(wrapper)

Create a new intestine object with wrapper, which will be called with the runner for each file loaded to set up a custom context or what-have-you.

The wrapper should call runner.start(t) with the your custom test object t. Your test object should emit 'assert' and 'end' events.

guts.append(src, opts)

Load some source src to pass along to the runner.

You can set a custom runner with opts.runner. If not specified, the runner defaults to stackedy(src, opts).

Any custom runner should have a .run(context) method.

guts.run(context)

Run each test file with the runner under a context.

events

'start', test

Fired when a test begins.

'assert', result

Fired when an assert happens in a test.

result.test will be a reference to the test object that fired the event.

Otherwise, the format of result is complete up to the custom test handler.

'error', err

Error events are emitted when using the default stackedy runner.

'finish', test

Emitted when a single test is completed.

'end'

Emitted when all the tests are complete.