`enum_variant_names` too trigger-happy

[couchand]

Summary

I have an enum representing possible errors from deserialization:

pub enum PromotionError {
    NoPromotionId,
    InvalidPromotionId(String),
    InvalidProductId(String),
}

clippy produces this: warning: all variants have the same postfix: Id and "helpfully" suggests: help: remove the postfixes and use full paths to the variants instead of glob imports

Maybe it's just me but the suggestion here seems completely irrelevant. I'm not even sure what the lint is trying to catch: what's the stylistic concern here?

Lint Name

enum_variant_names

Reproducer

I tried this code:

pub enum PromotionError {
    NoPromotionId,
    InvalidPromotionId(String),
    InvalidProductId(String),
}

I saw this happen:

[<output>](warning: all variants have the same postfix: Id)

I expected to see this happen:

No error.

Version

rustc 1.68.1 (8460ca823 2023-03-20)
binary: rustc
commit-hash: 8460ca823e8367a30dda430efda790588b8c84d3
commit-date: 2023-03-20
host: x86_64-unknown-linux-gnu
release: 1.68.1
LLVM version: 15.0.6

Additional Labels

No response

[bonsairobo]

Yea I hit this as well and agree that the current implementation is annoying in some cases. The stated reason for the lint is to avoid post-fixing variants with the enum type name, so that's what the lint should check for.

[7sDream]

This code section does the check of all variant in a enum has a common prefix/postfix.

It makes sense that variants should not repeat enum names, but I don't see anything wrong with enums having a common prefix/postfix that is independent of the enum name.

At least there needs an configure option.

---

And, according to the code, if enum-variant-name-threshold is set to 0 or 1 and there is only one variant in the enum, then no matter what name it is, this lint will report an (un-fixable)warning.

[spearman]

It would be nice if it actually checked for variants that repeated the enum type name instead of just any repeated string, which I don't find useful.