Negative coverage when integer overflows

[daveMueller]

closes #1266

[daveMueller]

I'm not so sure about the test. The loop to int.MaxValue takes quite some time and slows down the build job.

[petli]

I agree the test seems overkill. The interesting logic is in the collection of the hit results, can that be tested by a unit test that just reads negative hits, rather than a full integration test?

[petli]

I have deleted two review comments (and updated the third) as I realised I did my sums wrong and the situation I was thinking about couldn't happen.

[daveMueller]

Thanks @petli. I will try to somehow mock a InstrumenterResult instead but I'm always having a hard time when I want to do this.

[petli]

Good unit test!

[MarcoRossignoli]

Thanks to all!