Cannot expect(setTimeout) without legacy mode of useFakeTimers()

[kenlyon]

💥 Regression Report

I upgraded to jest 27.0 and found my tests failed due to the change in useFakeTimers().

The error is:

Matcher error: received value must be a mock or spy function

Received has type:  function
Received has value: [Function setTimeout]

It does work when I use the "legacy" option, however my usage is the recommended approach. I don't think it constitutes the "subtle implementation differences" described in your blog.

Last working version

Worked up to version: 26.6.3

Stopped working in version: 27.0.4 (Probably 27.0.0)

To Reproduce

Steps to reproduce the behavior:

1. In a test file, call jest.useFakeTimers("modern"); (or jest.useFakeTimers() in v27+)
2. Attempt to use setTimeout as a mock, such as expect(setTimeout).toHaveBeenCalledTimes(0);

The problem might only happen on Windows, as I couldn't reproduce it in repl and everything else was the same.

Expected behavior

It should still work, as this is the technique described in your docs.

Link to repl or repo (highly encouraged)

repl.it demo

NOTE: The repl runs without error, but fails locally on my Windows 10 workstation. I confirmed I have an identical package-lock.json and have cleaned my npm cache etc. Might it be only an issue on Windows?

Run npx envinfo --preset jest

Paste the results here:

  System:
    OS: Windows 10 10.0.19041
    CPU: (8) x64 Intel(R) Core(TM) i7-9700 CPU @ 3.00GHz
  Binaries:
    Node: 14.16.0 - C:\Program Files\nodejs\node.EXE    
    Yarn: 1.22.10 - C:\Program Files\nodejs\yarn.CMD    
    npm: 6.14.11 - C:\Program Files\nodejs\npm.CMD      
  npmPackages:
    jest: ^27.0.4 => 27.0.4

[Smrtnyk]

seems like a duplicate of #11500

[kenlyon]

@Smrtnyk They might be related, but I'm describing a different symptom. I don't know if setTimeout is working or not. I don't get past expect(setTimeout) because the fake version of setTimeout does not appear to be a valid mock.

[kenlyon]

@Smrtnyk I should have read through the other thread before responding. The discussion does seem to be going in the direction I thought this one might, so there is some duplication in that sense. It's the same underlying cause, but reported via a different use case. I see pros and cons to "duplicate" vs "related" but I don't care to split hair over it. :) I'm just glad that others are seeing the same thing.

It's weird that it works in my repl.it though.

[SimenB]

Duplicate of #11447

[github-actions]

This issue has been automatically locked since there has not been any recent activity after it was closed. Please open a new issue for related bugs.
Please note this issue tracker is not a help forum. We recommend using StackOverflow or our discord channel for questions.