Remix.run Logo
▲ ai_ 2 hours ago

The function you seek already exists before you: https://doc.rust-lang.org/std/mem/fn.discriminant.html

Is there anything you can do with separate types for the tag and the union that std::mem::discriminant and pattern matching can't do?

▲Jtsummers 2 hours ago | parent | next [-]

As someone who hasn't used Rust enough, and not knowing Zig but it appears Zig does something similar to Ada which I do know, how would you use the same discriminant across multiple Rust enums and guarantee that you implemented variants for each discriminant value?

In Ada, if I use an enum (carrying the example from a link I used in a cousin comment):

  type Expr_Kind_Type is (Bin_Op_Plus,
                           Bin_Op_Minus,
                           Num);
https://learn.adacore.com/courses/intro-to-ada/chapters/more...

Variant records using this enum have to implement the variant for each element of the enum. If I, for some reason, have multiple records using the same discriminant I can know that each one implements it for each value of the discriminant. Does Rust have a way to synchronize your enums similar to that?

▲bombela 41 minutes ago | parent | next [-]

  pub trait Payloads {
      type Read;
      type Write;
  }

  pub enum FileOp<P: Payloads> {
      Read(P::Read),
      Write(P::Write),
  }

  pub struct Inputs;

  impl Payloads for Inputs {
      type Read = ();
      type Write = u8;
  }

  pub type Request = FileOp<Inputs>;

  pub struct Outputs;

  impl Payloads for Outputs {
      type Read = u8;
      type Write = ();
  }

  pub type Reply = FileOp<Outputs>;

  pub fn execute(request: Request) -> Reply {
      match request {
          Request::Read(()) => Reply::Read(42),
          Request::Write(_) => Reply::Write(()),
      }
  }
▲Jtsummers 28 minutes ago | parent [-]

This looks like it would handle synchronizing multiple variant records/enums, and if you just wanted the "plain" enum without any payload, you'd need to do:

  pub struct Bare;
  impl Payloads for Bare {
    type Read = ();
    type Write = ();
  }

  pub type BareFileOp = FileOp<Bare>
Correct? So you'd need to create it with BareFileOp::Read(()), but at least it wouldn't carry any overhead of actually containing data.
▲afdbcreid an hour ago | parent | prev [-]

Why do you need to do that? You can't be generic over "any enum using this tag" anyway (in Rust, at least). If you just want to make sure you make no mistake, you can use a dead match from the tag to the enum, or automate it with a macro, and the compiler will enforce exhaustiveness (though not on both sides, unless you use a macro).

▲Jtsummers an hour ago | parent [-]

If I'm building up a type checker and an interpreter, I very well could (continuing that example) end up with different data structures using the same discriminant (but containing different data about the system). In Rust I'd have to do something like:

  enum Parser {
    bin_add(...),
    bin_sub(...),
    ...
  }

  enum Typer {
     bin_add(...),
     bin_sub(...),
     ...
  }
And I'd have to hope they were both complete and match (at least according to what I know now, that's why I'm asking). If I go back in to the Parser enum and add a `un_minus`, then I have to also go into the Typer enum (apologies for the bad naming, I'm at work and commenting between my way too many meetings) and manually add it. I have no compile time guarantee that both are complete and synchronized. In Ada, I modify the existing enum and both will fail to compile until I add in the variant for `un_minus`. One edit, multiple failures distributed across the code base as a result telling me all the places where I have to add in a new code branch or new variant.

EDIT: Run on sentence broken up.

▲bombela 38 minutes ago | parent [-]

You can use one generic enum with a trait holding the payload type. I posted an example as response to your first comment.

▲tcfhgj an hour ago | parent | prev | next [-]

yeah, but fortunately strum exists.

I have found it quite useful to be able to generate a dedicated SomeEnumDiscriminants enum for SomeEnum, using the following derive:

https://docs.rs/strum/latest/strum/derive.EnumDiscriminants....

▲xeubie 2 hours ago | parent | prev [-]

Yes, there is something you can do with separate types. You can name a variant without having a value of it. `discriminant` only works on an existing value, so to refer to "cats" you'd have to build a dummy `Cat`, and the result is opaque. You can't match on it, iterate over it, or create one from a byte you parsed off the wire. This is so obviously worse, I can't believe you bothered to ask.

▲sagarm 22 minutes ago | parent | next [-]

You design the enum. If you need kinds separate from the union itself, you have that power.

The union type always implies a kind, after all, so it's a generalization of the concept.

▲afdbcreid an hour ago | parent | prev [-]

This is a minor advantage (but is an advantage) and when needed (rarely) you can usually workaround with a different fieldless enum and potentially a proc-macro to autogenerate it.

▲xeubie an hour ago | parent | next [-]

It's not at all minor. Sometimes the tag enum already exists. Rust can't build a data enum on top of it: a derive macro only sees its own enum, so it generates a new tag enum instead. Now you have two tag enums to keep in sync by hand. Proc macros are always used to resolve deficiencies in the rust language, and they always lead to greater complexity and longer compile times.

▲tcfhgj an hour ago | parent | prev [-]

> proc-macro to autogenerate it.

https://docs.rs/strum/latest/strum/derive.EnumDiscriminants....