NyxCode

> So, sorry for this, that got way to long. All of it is just an idea, and none of it is anywhere near fleshed out. Please, let me know...

would love to take a loot at this once i find some time 👍

Hey! If we add an `impl` for it behind a `bnum-impl` feature gate, then that `impl` should be 1:1 compatible with their serde serialization. I think that's the only way...

Our impls for large integer types (e.g `u128`) *do* currently use `bigint`, but it was a somewhat painfull tradeoff with them. Serializing them as `number` would be serde-compatible by default,...

Looking at bnum's serde feature, they just use plain derives under the hood. They have two families of types, `BIntXX` and `BUintXX`. `BIntXX` is serialized to `{"bits":{"digits":[42,0,0,0,0,0,0,0]}}`, `BUintXX` is serialized...

I explored a simple approach to implementing this: Example: ```rs #[derive(TS)] #[ts(export, split, export_to = "tests-out/split_enum/")] enum X { A(i32), B { x: i32, b: i32 }, } ``` In...

Hey! While I share @gustavo-shigueo's sentiment, i still think supporting this would be pretty cool. Maybe we could add a `#[ts(repr)]`, which would try to be compatible with `serde_repr`? Anyways,...

Hey, thanks for opening this PR! Is there a real use-case behind this feature, or is this more of a "it'd be really cool if this worked" kind of thing?...

Before too much effort is put into this, i think we should answer the question if we really want this feature. I am yet to be convinced of the benefit,...

With #397, I think we finally found a motivation for this - external codegen tools. I have implemented something similar for `#[ts(rename)]` in #398. There, I've just allowed full `Expr`,...