Serde like rename for Display trait on enum

[ModProg]

i.e. I would expect

#[derive(Display)]
#[display(rename="snake_case")]
enum Test{
   First
}

println!("{}", Test::First) // -> prints "first"

The serde behaviour is documented here:
https://serde.rs/container-attrs.html#rename_all

#[serde(rename_all = "...")]

Rename all [...] variants [...] according to the given case convention. The possible values are "lowercase", "UPPERCASE", "PascalCase", "camelCase", "snake_case", "SCREAMING_SNAKE_CASE", "kebab-case", "SCREAMING-KEBAB-CASE".

[JelteF]

This sounds useful. It fits well with the FromStr for unit enum support that was introduced in #176.

bikeshed: I think I prefer rename over rename_all. That way you could also put rename on a single field. The top level rename would then simply set the default for enum items.

[JelteF]

A good question raised in the thread in #225 would be what should be the default? My summary of that discussion is that there are two different usages for which a different default makes sense:

1. Error types: you probably want a custom error message. By requiring to set this attribute you would not be able to forget to set a custom message for one of your variants.
2. Any other enum type: You probably want nice display by default

I'd say go for strict by default for now, i.e. error by default. Then if that is deemed too cumbersome we can always relax the default. However, unrelaxing the default would be a breaking change.

[ilslv]

@JelteF

bikeshed: I think I prefer rename over rename_all. That way you could also put rename on a single field. The top level rename would then simply set the default for enum items.

Actually #[display(rename("...")) is an alias for direct #[display("...")], so I would vote for rename_all as seen in serde and already widespread across the ecosystem.

I'd say go for strict by default for now, i.e. error by default. Then if that is deemed too cumbersome we can always relax the default. However, unrelaxing the default would be a breaking change.

I think that going explicitly over implicitly is a viable option, but only when it's consistent. We already have 2 examples for implicit display:

#[derive(Display)]
struct Unit;

#[derive(Display)]
struct SingleField(u8); // works for named `{ field: u8 }` too

I see implicit Display impl on enums with unit fields as understandable and discoverable continuation for that:

#[derive(Display)]
enum Enum {
    A,
    B,
}

Moreover this behaviour is already usable on master (and released as well):

// derive_more::From fits nicely into this pattern as well
#[derive(Debug, Display, Error, From)]
enum CompoundError {
    Simple,
    WithSource {
        source: Simple,
    },
    #[from(ignore)]
    WithBacktraceFromSource {
        #[error(backtrace)]
        source: Simple,
    },
    #[display(fmt = "{source}")]
    WithDifferentBacktrace {
        source: Simple,
        backtrace: Backtrace,
    },
    WithExplicitSource {
        #[error(source)]
        explicit_source: WithSource,
    },
    #[from(ignore)]
    WithBacktraceFromExplicitSource {
        #[error(backtrace, source)]
        explicit_source: WithSource,
    },
    Tuple(WithExplicitSource),
    WithoutSource(#[error(not(source))] Tuple),
}

[marcocondrache]

Any update on this issue? Applying a pre processing function on every variant name is very useful in different situations.

[JelteF]

Okay this dropped off my radar.

@marcocondrache at the moment we're focussing on tasks that are blockers of 1.0.0, this is not one of them, because it can be implemented in a backwards compatible way. Feel free to contribute a PR that implements this.

@ilslv Sorry I didn't respond. Your response make total sense. Let's go with rename_all and implicitely doing it (because we already do that and it doesn't seem worth breaking backwards compatibility for this).