Andrew Lee

@anujad95 There are a lot of failing tests in this PR which need to be fixed before we can review it. You should also update the branch as well.

@Robbybp The `_run_test` method and associated class is not intended to be universal - it is meant to be a helper class for common tests based off of existing IDAES...

@Robbybp The first thing to note is that Pyomo is actually collecting a lot of information in the background that is not immediately apparent. E.g., we get construction times for...

@Robbybp r.e. different types of tests; for now I would hope that we can make a general testing class that covers most cases (once the unit consistency issue is resolved,...

@Robbybp Ah - that was part of the intent behind the three methods (build, initialize and solve) - so that we could reuse the same code between multiple tests. I...

@Robbybp I've added some changes so that you only need to provide a single test class for each case which is expected to have the methods to build, initialize and...

@Robbybp So, a few comments in response (there is a lot there so I will probably miss something. 1. First, I think we are talking about different things here. The...

@Robbybp OK, I think I see the source of the problem here - the real issue is that we do not have a formal, easily discoverable definition of what constitutes...

@Robbybp I think I have the solution: the `PerformanceTest` base class should be merged with the existing `TestModel` class and have only two methods that it must define: `build_model` and...

@Robbybp I think I have something working that doesn't require the dict of test classes and just uses a common base class and `pytest` discovery. This does mean however that...