gr0uch / tapdance Goto Github PK

It should expect the number of assertions for async testing.

t.run(() => {
  t.expect(2)
  t.ok(true)
  // should generate failing message
})

A feature I've grown accustomed to in test frameworks is a .only() method. The use case is this:

You're working on a library with 1000 tests. You need to add a new feature. It breaks a few tests, but you don't want to add logging to the method that's throwing, because it'll be called a thousand times. So you add a .only() to isolate just the failing tests, so that you can add debug logs.

One idea that works is commenting out all of the other tests, but that doesn't seem as efficient.

Would you open to adding a .only() to this lib? If not, how do you go about debugging tests when just a few fail among many?

Ensure that results are printed in the correct order, even if that means faster results are withheld for some time.