I have a very simple project that works like a transpiler, taking a kind of metalanguage and converting it into Rust code. (Yes, it's a bit like reinventing gRPC.)
Currently, I have a file with numerous types, like:
#[derive(FromWire, ToWire)]
pub struct Foo {
pub a: String
}
FromWire
and ToWire
are traits that provide from_wire
and to_wire
methods.
The idea is to expose these types to Python without implementing code generation for Python. This choice is both for performance reasons in encoding and decoding and, admittedly, because I want to avoid extra coding work.
I haven't worked extensively with FFI bindings before, so I'm looking for advice on the best approach to create a Python package that can be deployed to PyPI.
I'm particularly interested in opinions because these files contain just types, like the ones shown earlier, performing similar functions. I'm hoping there’s existing automation for generating bindings.