Clarification on use of interactive needed

[kennedymwavu]

Background Info

I was going through the source code and stumbled upon this:

Link: https://github.com/JohnCoene/firebase/blob/b72e8847191d7826f0bfd204409682581cd8251e/R/class-auth.R#L280C25-L280C25

if (interactive) is used instead of the expected if (interactive()). I can't find any explicit definition for interactive as a variable.
This pattern is repeated multiple times so I'd assume it's not a bug.

Question

Is this a unique R feature or am I overlooking something?

Expectation

My expectation of if (interactive) is that it should throw an error since interactive is a function, not a logical. And, indeed, it is what happens in my reprex below:

check_interactive_mode <- \() {
  if (interactive) {
    print("This is an interactive session")
  } else {
    print("This is a non-interactive session")
  }
}

check_interactive_mode()
#> Error in if (interactive) {: argument is not interpretable as logical

^{Created on 2023-12-13 with reprex v2.0.2}

[JohnCoene]

Thank you for this.

It's clearly a bug from @munoztd0 :p

But, because this runs in a shiny server interactive() will never be true.
This has to be reworked

[munoztd0]

yeah my bad I fixed the typo and now I understand why it worked when it should have not..

[munoztd0]

Removed my mistake in the commit fad6e21.

To re explain the issue is that Windows internal clock didn't align with the API so that would block people developing on the windows machine.

I just added a way to bypass this error for development purposes but it should be fine once in prod on a linux server. I don't know about a Windows server though don't really have anything to really test it and I don't think it woud come up very often ..

[munoztd0]

I will close this one and reopen the main one #64