Is this issue about improving our code coverage score, or are there tests of functionality that you think should be included in ExtractGraphXTest.scala?

If this is about bringing up the code coverage score on ExtractGraphX.scala to 100%, I would contend that isn't a useful target. The codecov report says it's just the case class declarations that aren't being adequately tested. I think this is because case classes contain a number of generated methods, such as equals, copy, and toString, which we aren't testing.

To demonstrate, I added a test of EdgeData's equals method. See how the cobertura report improves. Without equals test:

With:

Adding a test of toString bumps up the branch coverage by one. Et cetera.

I think as long as we understand there's a valid reason why we are not hitting 100%, it's reasonable, and sensible, not to bother pursuing it.

If this is about bringing up the code coverage score on ExtractGraphX.scala to 100%, I would contend that isn't a useful target.

Agree!

Background; I went through every file, and just created an issue for it just to documented and be consistent with working one our test coverage. When we started this process, I believe we were around 50% coverage. So, we've improved a fair bit.This particular one isn't that big of a deal. I personally don't mind if we get 100% on it. If you want to add something in for the classes, I'm cool with it. If you don't we can just leave this issue open or close it and tag it won't fix.

I think as long as we understand there's a valid reason why we are not hitting 100%, it's reasonable, and sensible, not to bother pursuing it.

Agree!

Let's close and tag it won't-fix.