Mixed usage of `setTimeout` and `timers.setTimeout`

nock/lib/request_overrider.js uses both setTimeout and timers.setTimeout. Should the project use one or the other for the sake of consistency?

Using setTimeout is nice because if a someone is mocking the clock, then nock will use a mocked clock as well. They can still do it, but have to do additional work to mock timers.

[Chengxuan]

This could have been communicated better, our tests started failing after picked up 10.0.5 because most of them use fakeTimer which didn’t impact nock. Now they all timing out and it took me a while to figure out why. I agree to give more control is better, but I don’t think most of existing users know it uses a separate timer due to existing behaviour. They’ll need to go through the same debugging process as I did now.

Maybe update your readme with an example would help? This is a change of behaviour for all the existing tests uses fakeTimer. Those tests will start failing with no obvious error other than timing out and it would take a while to track down to the nock library.

[nockbot]

🎉 This issue has been resolved in version 11.0.0 🎉

The release is available on:

• npm package (@next dist-tag)
• GitHub release

Your semantic-release bot 📦🚀