Simple, sequential, exception based test runner for node js.
- usual
test
,skip
,only
,todo
api - usual
before
,beforeEach
,afterEach
,after
hooks - async support
- speed
- timeout checks
- parallel testing
- assertions (tip)
- CLI (read more below)
npm i -D @marianmeres/test-runner
const suite = new TestRunner('My suite');
suite.test('My test', () => {
if (false) {
throw new Error('...')
}
});
suite.run();
See examples for more.
npx test-runner [-d some-dir] [-d another-dir]
Or there is a TestRunner.runAll
api, so something like this should work fine:
// tests/index.js
const args = process.argv.slice(2);
const verbose = args.includes('-v');
const whitelist = args.filter((v) => v !== '-v');
TestRunner.runAll([__dirname, '../src'], { verbose, whitelist });
runnable as
$ node tests [-v] [whitelist]
or also watchable as
$ nodemon -q tests -- [-v] [whitelist]
The TestRunner.runAll
looks by default for [<path>/]<file>.test.[tj]s
. Each test file must
have the suite instance "default exported":
// src/some.test.js
const suite = new TestRunner('My suite');
// tests definitions here...
export default suite;
// or depending on your env:
// module.exports = suite;
See examples for more.
Screenshots are taken from examples.
This runner does not spawn the actual tests executions into separate child processes.
This is an intentional choice, but it comes with the price of not beeing able to truly
isolate/kill/cancel the test execution and its context (such as pending setTimeout
s)
which can sometimes lead to unexpected results.
Check the definition files.