Deprecate ChefSpec's coverage reporting.

[coderanger]

So people are aware they shouldn't use it.

[tas50]

YES!

[jason-riddle]

Ok, but what should be used instead?

[coderanger]

@jason-riddle Nothing, there is no replacement.

[iancward]

If the intent is to deprecate code coverage, you might want to update the description of the repo:

[jamietanna]

Out of interest, would there be any appetite for re-maintaining it? Or is it less used, so should be pushed out to a separate tool?

[tas50]

It's not a matter of deprecating it so we don't have to maintain the code. When using the coverage reporting functionality people tend to right bad specs. You should not be chasing 100% spec coverage in Chef cookbooks. It doesn't make sense and it gives you a false sense of security. You should be testing the portion of your cookbook with logic. If that's 10% of the total cookbook then you only need/want 10% coverage.

[daxgames]

But what about people that write GOOD specs that evaluate logic like guards and differing attribute values that cause resource collections to have different contents? In this case I think the coverage reports are very valuable and the fact there is no replacement is not good for those that do use it and do it right.

[blackknight36]

IMO this is an essential feature and should not be removed until there is an alternative available. Yes, writing unit tests is annoying but it is also essential in ensuring that your recipes work as expected.