On 13/06/2017 19:31, Andrzej Krzemienski wrote:
First of all. I haven't seen anyone give an example where they need a strong guarantee for variant's assignment. There was one attempt with a state machine, but even there it seamed only theoretical. I think the questions if we can implement it are secondary to if anyone really needs it.
Actually the state machine example was a thought exercise that (I think) proved a strong guarantee was *not* useful for that case.
If whoever has a real (as opposed to theoretical) need for a strong guarantee, please respond.
I think the first barrier is that someone needs to come up with a case where a strong guarantee on assignment for *any* type (not variant) is useful. Once some type T needs it, it would become easier to argue that variant<T, ...> should provide it as well (due to the desire that a variant containing a T should act as much like a T as possible). I'm still not entirely sure I'd agree with that argument, but at least it becomes possible to have that argument with something concrete in mind.