Andy Arvanitis

I think using something like this as a memory management option is a great idea. I've also wondered about the use of effects for it too. The plan is definitely...

This should be pretty easy to do now: https://github.com/andyarvanitis/pure11/wiki/Memory#unmanaged-data

Yeah, only building the relevant FFI files has been on the todo list (even if implicitly). It would also reduce build times.

Unfortunately I don't really use nix myself, so I can't be of help. The nix support we have is from other contributors like @kayhide. However, I can say that the...

I haven't (yet?) tried any approach to a C++ REPL for purescript-native. In the past, for an unrelated project, I experimented with a couple ways to do a C/C++/ObjC REPL...

Neat! I'll try to have a look at it sometime soon.

When doing a type assertion (cast) in go, if you don't provide a (discarded) failure bool variable it will panic (with a description) if it fails. In other words, you...

The idea was to end up with the (allegedly faster) two-result (ignored failure) assertions in ffi implementations. In my readme I intended to convey that an author should use the...

Well, if I end up getting a definitive answer that the panics don't cause any overhead I would just get rid of the two-result versions everywhere. I'd like to avoid...

I don't think there's a way around arrays needing to be of type `[]Any`, short of using opaque types in places, as you mention. A helper function for boxing arrays...