Testing

Langoost has a built-in test runner and an assert module — no external framework needed.

Writing tests

Put tests in files named *_test.goost. Each top-level function whose name starts with test_ is a test. A test passes if it returns normally and fails if it throws — and the assert helpers throw on failure.

// math_test.goost
import assert

fn add(a, b) { return a + b }

fn test_add() {
    assert.eq(add(2, 2), 4)
}

fn test_add_negative() {
    assert.eq(add(-1, -1), -2, "negatives should add")
}

fn test_throws_on_bad_index() {
    assert.throws(fn() {
        let xs = [1, 2, 3]
        return xs[99]            // out of bounds → throws
    })
}

The assert module

import assert

Function	Signature	Description
`ok`	`assert.ok(value [, msg])`	Fail if `value` is falsy
`eq`	`assert.eq(actual, expected [, msg])`	Fail unless equal (uses the VM’s `==` semantics)
`ne`	`assert.ne(actual, expected [, msg])`	Fail if equal (the inverse of `eq`)
`throws`	`assert.throws(fn [, msg])`	Run `fn()` and fail unless it throws

The optional msg is included in the failure output.

Running tests

$ langoost test                      # discover *_test.goost under .
$ langoost test ./tests -v           # verbose: list every test
$ langoost test -run add             # only tests whose name contains "add"

The runner reports each pass/fail and exits non-zero if any test fails — so it drops straight into CI:

math_test.goost
  ok    test_add
  ok    test_add_negative
  FAIL  test_throws_on_bad_index

2 passed, 1 failed

See the CLI reference for all flags.

Reference · View as Markdown · llms-full.txt