I suspect that Clara does enough code generation and evaluation behind the scenes to confuse tooling like Cloverage, at least without some modification to specifically support Clara, but it isn’t something I’ve looked into. I don’t think the rules network could be effectively statically analyzed but perhaps something could be created that looked at what rules were interacted with; that information is available via Clara’s tracing

I’m not even sure how Cloverage performs its analysis. Is it via tools.analyzer ?

I'm not exactly sure how it works; but a search in the repo for tools.analyzer returns empty

I'm mostly just interested in seeing what rules are firing during tests, to make sure I have sufficient tests.

@U0E2EQFME you could do some reading of the inspected session

Or maybe attach a tracing listener to track the rules fired

That information should be available via either method

Yeah, I think I looked at it before and it didn’t use analyzer. Not sure how it can check coverage. Especially via macros.

perhaps magic