Hi,
Imagine a use case : listing some items.
Imagine another use case : create an item.
I want to start with acceptance tests against the interactors part of the clean architecture (as recommended by Uncle Bob).
I started with the first use case, since it provides a good business value to customer.
Since the (second) use case for creating items didn't exist yet, I provide a fixture that populate the repository with some items; in order to be listed; so that I fulfill the acceptance test of the use case.
I passed the first test for the first use case : listing some items.
Then, I pass the test for the second use case : creating an item.
Question : Should I now replace the fixture (pre-populated items repository) within the first acceptance test by a call to the "create an item" use case ?
Meaning, should a use case depend on another use case in acceptance tests on interactor parts ?