Clock: add and subtract expect a string, not a Clock

[IsaacG]

The add and subtract methods in the Clock test data have expected results which are strings. In Python, this translates to allowing solutions to return a string and not a Clock. I'm not sure how this manifests in other languages.

How burdensome would it be to change the expected results to be a clock instead? Would that be valid canonically (ie across tracks)?

[SaschaMann]

On the Julia track we expect objects but also have tests for the string representation of those objects.

How burdensome would it be to change the expected results to be a clock instead? Would that be valid canonically (ie across tracks)?

One would either have to duplicate or reimplement every single test case.

---

I don't think this change would be worth it. The canonical data isn't student-facing, it exists for the maintainers. They can choose how to interpret it and can convert between strings and objects however they like. Even a generator could trivially deal with this translation.

[ErikSchierboom]

I don't think this change would be worth it. The canonical data isn't student-facing, it exists for the maintainers. They can choose how to interpret it and can convert between strings and objects however they like. Even a generator could trivially deal with this translation.

Agreed. A track is free to have the implementation compare objects instead of strings.