CharliePoole

The feature seems somewhat desirable. It could be done on an adhoc basis - i.e. allow `RetryAttribute` at higher levels than the method and apply it to all methods. I'd...

Your problem is likely to be lost if it's appended on a different issue, so you should file it separately. Before doing that, you should ascertain if the problem is...

Two problems here... 1. If your attribute worked, it would run the entire assembly multiple times on failure. That's what retry attribute does to the test it is placed on....

@nunit/framework-team This issue has been here for two years now. It has neither been accepted as a potential feature nor rejected, at least formally. There is some indication in the...

Explicit has always been a bit confusing. Originally designed as s property of a category, it quickly changed to an independent attribute with the meaning "only run on positive selection."...

@daflame76 In my own work, I use Explicit for things that should never run in CI - it doesn't matter whether they are unit-, integration- or acceptance-tests. That's an easy...

@daflame76 You are right, in designing Explicit, the intention was: * The lhs expression should not select any explicit tests because it's negative (an exclude). * The rhs expression should...

I agree with your idea myself, but I also think the old behavior without the new switch has to be fixed. It has to do with the way the binary...

What I meant in terms of design is exactly __how__ the syntax of the where clause should be changed to incorporate explicit. Personally, I'm less enthusiastic at a switch, which...

@daflame76 The current behavior is not precisely what I would call "legacy behavior" - I think the actual term is "bug". 😄