@varkor @yodal What is your notion of "must use" wrt. const parameters? e.g. is
struct A<const N: usize>; valid despite
N being unused?
(given that there's no variance to speak of...)
right now, no, but I saw @boats's comment about it
that can be fixed in a separate PR
(that is, it's not related to
exactly what @varkor said
I was planning on throwing together a small PR, but I had been pulled into work and linker madness lately and it isn't a pressing issue
ugh... how do you move this to a separate topic?
@varkor do you have the link to that comment?
I feel it's reasonable to open up issues about const generics features that are already implemented, at least just for ourselves
@varkor definitely do so; it is helpful for T-Lang as well to review the history
@yodal thanks; I agree @boats's note makes sense (and I agree about type parameters as well, but that's a separate issue)
(I can't remember how to move existing conversations to new threads on Zulip)
but I can fix that issue once I've added some tests
Just want to say a big "thank you" for all your great work on const generics. Really eager to see all this moving towards the finishing line (and eventually landing it on stable) :)