cool-runner

Cool Runner

An intentionally minimal, slightly opinionated test runner for NodeJS (does not work in the browser!)

Installation

Local installation:

yarn add cool-runner or npm install cool-runner --save

You may also install globally:

yarn add -g cool-runner or npm install -g cool-runner

Writing Tests

All tests must extend from the TestCase class.

const {TestCase} = require('cool-runner');
const assert = require('assert');

class MyTestCase extends TestCase {

    constructor() {
        super();

        // How long to wait for async operations
        // Default is 2000
        this.setTimeout(2000);
    }

    /*
     * Tests should start or end with the word "test".
     * camelCase, PascalCase, snake_case, etc. are all fine
     */
    testSomeSynchronousFunctionality(done) {
        assert.strictEqual(obj.runLogic(), true);
        done();
    }

    /**
     * `done` must be called for both synchronous and async tests
     * unless using promises ...
     */
    some_async_functionality_test(done) {
        runSomeAsyncLogic.then(response => {
            assert.deepStricEqual(response, mock);
            done();
        });
    }

    /**
     * Alternatively, return a promise that resolves after 
     * everything is finished
     */
    some_async_functionality_test_with_promise() {
        return new Promise((resolve, reject) => {
            runSomeAsyncLogic.then(response => {
                assert.deepStricEqual(response, mock);
                resolve();
            });
        });
    }

    /**
     * not all methods need to be tests
     */
    squareUtilityMethod(param) {
        return param * 2;
    }

    /**
     * Call the utility method from this test case
     */
    test_some_logic_with_utility_method(done) {
        assert.equal(this.squareUtilityMethod(2), 4);
        done();
    }

    /* other helpful methods to know about */

    beforeEach(done) {
        // do something before every test is run
        done();
    }

    /**
     * Same as above, you can call the callback when finished
     * or return a promise that resolves.
     */
    beforeAll() {
        // called once before the entire suite is run.
        return new Promise((resolve, reject) => {
            // do something important
            resolve();
        });
    }

    afterEach(done) {
        // do some cleanup after each test
        done();
    }

    afterAll() {
        // called once after suite is finished
    }
}

module.exports = MyTestCase;

Running Tests

CoolRunner exposes two commands, which recursively search your test directory and run all of the valid test cases.

cool-runner test: run tests once

• --testDir=tests: Specify the relative path to some other directory besides tests where tests are stored

cool-runner watch: run tests on file change to either src or tests

• --testDir=tests: Specify the relative path to some other directory besides tests where tests are stored
• --srcDir=src: Specify the relative path to some other directory besides src where source code is stored

Directory Structure

By default, CoolRunner assumes you're directory structure looks like this:

- node_modules/
- src/     // all source code
- tests/   // all tests
- etc...

If you interleave source code and tests, run the following to recursively search your source directory, running only valid test cases

cool-runner test --testDir=src

(Note, that there is a slight performance benefit to keeping them in separate directories. This benefit could be more important as the size of the tests and source directory grow.)

Why CoolRunner?

There are lots of great test runners available in the JavaScript ecosystem (which I use and enjoy!). However, CoolRunner is designed with a few things in mind:

• Speed: Tests should run in <1s, preferrably faster.
• Ease of use: Straightfoward API. Sensible defaults.
• Easy to wire-up to debuggers for interactive testing + debugging.
• Avoid any global state.

Contributing

Contributions welcome on Gitbub.

Issues

Submit issues on Github.

Authors

• Spencer Jones (sjones6)