Talk:Z13381

duplicate?

This looks about the same as contains (Z12696), with the arguments switched. Do we need both? --99of9 (talk) 01:45, 23 February 2024 (UTC)

Oh, I see you needed the arguments switched to make cancel → cancel(l)ing (Z13137). --99of9 (talk) 02:08, 23 February 2024 (UTC)

There’s madness in my method 🤪 GrounderUK (talk) 02:15, 23 February 2024 (UTC)

Because it’s a constraint imposed by the structure of Z20/tester, but not by compositions more generally, we may want to take this to project chat. It does not say in the function model that the first function in a tester must be the function it tests but that has been my limited experience. Personally, I would relax the validation such that there is only a requirement that there be a call to the function in there somewhere. If we don’t, we should have some way to specify which argument in the Validator takes the evaluation of the call to the function being tested, rather than a proliferation of functions like this one. That said, I think this particular function provides a general way to handle an “any of” scenario, so perhaps the proliferation would turn out to be minimal.

I think I concluded that we can’t use an argument reference in the validator that refers to an argument in the test function call but I’m not 100% sure. That is how I assumed it would be possible to connect the evaluation result to the appropriate validator input. Is that something you’ve succeeded in doing? GrounderUK (talk) 11:00, 23 February 2024 (UTC)