I'm attempting to publish a high quality building block with as much automated testing as I can muster. However, I'm stymied by attempts to properly set up my Blackboard Learn server to support this. Having a working server that reacts to changes is vital for end to end tests (driven by Selenium or some other similar solution) where I can't mock/stub methods.
Ideally, I would be able to create a standard baseline set of data and quickly read changes to that data that occur as part of the tests, as well as roll back to the baseline between each test.
I'm curious as to what strategies the rest of the community uses when it comes these sorts of automated tests.