Kitlith

concrete codegen for the above example ```rust #[doc(hidden)] #[allow(dead_code, non_camel_case_types, missing_docs, clippy::panic)] impl HasGenericBuilder { pub fn build(self) -> HasGeneric where T: Default { let (x, y) = self.fields; let...

perhaps i need to take a closer look again, i think i completely missed the closures? but the whole idea was to do this in a way that wouldn't interfere...

> 2^N is not something one should bring up so casually. Learn to respect the exponential. Respect - and fear. Succumbing to it would be so easy - being able...

Okay, this is an issue. I was trying to figure out why the FFI was depending on dynamic libusb. Turns out, the current github release builds of the FFI predate...

would it be reasonable to do something like "two material slots can be merged if and only if they are animated in the same animation, together, for all animations they...

A sketch for my manual idea: User Interface: A list of materials sets. Each material set's UI shows a list of every material slot on the avatar, and allows the...

hmm, i think there's going to be a bit more involved here, actually. First off, `ClientMessage`, `DeviceMessage`, and `ButtplugFFIServerMessage` have unused fields, in the form of id (and index for...

mark this as something to consider for ffi core 3.0 i guess, if we don't decide to nix protobuf entirely by then.

I think this issue could be updated to be about the Java FFI (not that we'll have users of it for much longer) We should already have mac and linux...

This sounds like a macro evaluation order issue. Ret moving the `#[derive(...)]` underneath the binrw directives (which, when using both, can be replaced with `#[binrw]`) If I'm correct, what's happening...