Skip to content
Open
Changes from 5 commits
Commits
File filter

Filter by extension

Filter by extension

Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
195 changes: 195 additions & 0 deletions text/3955-named-fn-trait-parameters.md
Original file line number Diff line number Diff line change
@@ -0,0 +1,195 @@
- Feature Name: `named_fn_trait_parameters`
- Start Date: 2026-04-24
- RFC PR: [rust-lang/rfcs#3955](https://github.com/rust-lang/rfcs/pull/3955)
- Rust Issue: [rust-lang/rust#0000](https://github.com/rust-lang/rust/issues/0000)

## Summary
[summary]: #summary

Allow (optional) named function parameters in `Fn`, `FnMut`, and `FnOnce`.
For example:
```rust
fn parse_my_data(
data: &str,
log: impl Fn(msg: String, priority: usize)
Comment thread
JonathanBrouwer marked this conversation as resolved.
) { }
```
Similar to named function pointer parameters, these names don't affect rust's semantics.

## Motivation
[motivation]: #motivation

### Benefit: Better documentation
Comment thread
JonathanBrouwer marked this conversation as resolved.

This allows users to better document the meaning of parameters in signatures. This is the primary benefit of this RFC.

For example, it is not immediately clear what the `String` and `usize` refer to in the type of `log`, providing names like in the example above is much clearer.

```rust
fn parse_my_data(
data: &str,
log: impl Fn(String, usize)
) { }
```

### Benefit: Better LSP hints
Copy link
Copy Markdown

@ChayimFriedman2 ChayimFriedman2 Apr 24, 2026

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

A major reason for this RFC is better documentation and better LSP hints. As a member of T-rust-analyzer, I unfortunately have to inform you that rust-analyzer cannot support this feature, at least not without significant changes (it does not show hints for named fn pointers parameters, either).

It might be possible, though, to support a few special cases.

View changes since the review

Copy link
Copy Markdown
Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Ah, sucks :c

I still believe this RFC is valuable even without this benefit though, as mentioned the primary benefit is "Better documentation"

Copy link
Copy Markdown
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Yea I think the main issue is that even if the type IR encodes this info, we might lose this information after inference depending on how the types are unified, trait bounds are proven etc

Copy link
Copy Markdown
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

That said, even though R-A won't necessarily be able to provide inlay hints, the hover-rustdoc could show this information, meaning that it technically still could have some usefulness in the LSP implementation, if only indirectly.

Copy link
Copy Markdown
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Yea I think the main issue is that even if the type IR encodes this info, we might lose this information after inference depending on how the types are unified, trait bounds are proven etc

when unifying two types, couldn't rust-analyzer try to propagate argument names if one type has them and the other doesn't?

Copy link
Copy Markdown

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

@programmerjake We probably could, but that depends on the solver allowing us to encode this info (which is a rustc crate and perf-sensitive, and also a major change).


When calling `log` in the body of `parse_my_data`, the LSP can provide the function parameter names as "inlay parameter name hints":
log(`data: `"Message".to_string(), `priority: `1);

This is a concrete advantage of this approach over using comments to do the same thing, such as in:
```rust
fn parse_my_data(
data: &str,
log: impl Fn(/* msg */ String, /* priority */ usize)
) { }
```

### Benefit: Better consistency with `fn` pointers

Imagine if `parse_my_data` looked like this:
```rust
fn parse_my_data(
data: &str,
log: fn(msg: String, priority: usize)
) { }
```

If due to new requirements the user decides that `impl Fn` suits the usecase better, having to remove the parameter names is unintuive.
This RFC removes this problem.

Note that the syntax for this feature does not exactly match that of `fn` pointers, see the [reference level explanation](#reference-level-explanation).

## Guide-level explanation
[guide-level-explanation]: #guide-level-explanation

You can give names to parameters in the `Fn`, `FnMut` and `FnOnce` traits to better document the meaning of these parameters, to help people who call your function.
These names are optional and don't have any semantic meaning. Named and unnamed parameters can be mixed, for example:

```rust
fn parse_my_data(
data: &str,
log: impl Fn(String, priority: usize)
) { }
```

This same syntax also applies to `Fn`, `FnMut` and `FnOnce` trait bounds, for example:
```rust
fn parse_my_data<
L: Fn(msg: String, priority: usize)
>(
data: &str,
log: L
) { }
```

## Reference-level explanation
[reference-level-explanation]: #reference-level-explanation

The syntax `Fn`, `FnMut` and `FnOnce` traits is currently not documented in the reference.
Copy link
Copy Markdown
Contributor

@clarfonthey clarfonthey Apr 24, 2026

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Mostly just adding this comment here from Zulip to keep track of some of the discussion.

I mentioned that we should probably try to unify the syntax for Fn traits, fn pointers, and functions without bodies. It appears that similarly, the reference is not fully clear on these either.

Particularly, the example you gave that did not compile:

trait Test {
    fn thing((x, y): (usize, usize));
}

seems to directly contradict what the reference says: https://doc.rust-lang.org/stable/reference/items/functions.html

In particular, it appears that only name: Type is supported, not arbitrary patterns.

View changes since the review

Copy link
Copy Markdown
Member

@fmease fmease Apr 24, 2026

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

We've already briefly discussed this on Zulip but just to get others up to speed: The Reference is correct in stating that the "parameter names" of even bodiless functions are patterns (PatNoTopAlt to be clear).

The grammar rules describe what's syntactically allowed and doesn't dictate what's semantically allowed or forbidden.

Patterns except CommonIdent | "_" in bodiless functions are semantically ill-formed but syntactically allowed.

Copy link
Copy Markdown
Member

@fmease fmease Apr 24, 2026

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

To further clarify this (since the Reference doesn't document this nuance which I'll rectify in a moment):

The parameters of function pointer types basically share the same grammar as parameters of trait associated functions in Rust 2015 where the "parameter name" is optional and where the parameter pattern is restricted. As I've mentioned on Zulip it obeys the following rules (excerpt):

RestrictedPat = RestrictedIdent | "&" RestrictedIdent | "&&" RestrictedIdent | "mut" CommonIdent;
RestrictedIdent = CommonIdent | "_" | "false" | "true";

In Rust >2015, however, the grammar of trait associated functions is the same as all other kinds of functions (free, foreign, impl associated).

Copy link
Copy Markdown
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

In the past RestrictedPat was semantically accepted as well.
I thought we eliminated it many years ago through a deprecation lint, but apparently the lint only worked at semantic level and wasn't reported at parsing time. I suspect we just didn't have the parse-time linting infra at the time (lint buffering, etc).

Copy link
Copy Markdown
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

I'm currently working on yeeting that. I've seen the historical RFCs, future-incompat issues and relevant PRs.

Fun fact, the & and the && are the remnants of argument modes (2012/2013-ish era?). Only by happenstance do they get parsed as borrow patterns nowadays.

Back in the day there were 5 argument modes, - (by move), & (by mut ref), && (by ref), + (by copy), ++ (by value). That used to look something like fn f(-a: A, &b: B, &&c: C, +d: D, ++e: E) {}

That's literally the reason why, it's beyond insane. Crater run coming soon btw :)

Copy link
Copy Markdown
Member

@fmease fmease Apr 24, 2026

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

The syntax Fn, FnMut and FnOnce traits is currently not documented in the reference.

It is documented. See e.g., https://doc.rust-lang.org/nightly/reference/paths.html#grammar-TypePathFnInputs. More specifically TypePathFn.

View changes since the review

Copy link
Copy Markdown
Author

@JonathanBrouwer JonathanBrouwer Apr 24, 2026

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Ahhh that is where it is, I will change the reference level explanation in terms of those rules

Copy link
Copy Markdown
Member

@fmease fmease Apr 24, 2026

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Addendum: Type paths are not the only paths featuring parenthesized generic argument lists. Expression and pattern paths have the same behavior. For example the following is legal:

#[cfg(false)]
fn scope() {
    let F::() -> () = F::() -> ()::Y;
}

Copy link
Copy Markdown
Author

@JonathanBrouwer JonathanBrouwer Apr 24, 2026

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

I've rewritten the reference-level explanation in terms of TypePathFn. I will process your addendums later

Copy link
Copy Markdown
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Final addendum: This syntax is obviously not tied to Fn, FnMut and FnOnce; the path can be anything. Moreover, from a semantic standpoint, not only are Fn{,Mut,Once} legal but also AsyncFn, AsyncFnMut and AsyncFnOnce.

Copy link
Copy Markdown
Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Type paths are not the only paths featuring parenthesized generic argument lists. Expression and pattern paths have the same behavior. For example the following is legal:

Is this behaviour documented in the reference anywhere?

Copy link
Copy Markdown
Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

This syntax is obviously not tied to Fn, FnMut and FnOnce; the path can be anything. Moreover, from a semantic standpoint, not only are Fn{,Mut,Once} legal but also AsyncFn, AsyncFnMut and AsyncFnOnce.

I've made this more explicit in the RFC


Before this RFC, the syntax rules are:
```grammar,types
ImplTraitParen -> `impl` TraitBoundParen

TraitBoundParen ->
( `?` | ForLifetimes )? TypePath TraitBoundParenArgs
| `(` ( `?` | ForLifetimes )? TypePath TraitBoundParenArgs `)`

TraitBoundParenArgs -> `(` FunctionParametersNoAttr? `)` BareFunctionReturnType?

FunctionParametersNoAttr ->
Type ( `,` Type )* `,`?

BareFunctionReturnType -> `->` TypeNoBounds
```

After this RFC, the following rules will change:

```grammar,types
TraitBoundParenArgs -> `(` MaybeNamedFunctionParametersNoAttr? `)` BareFunctionReturnType?

MaybeNamedFunctionParametersNoAttr ->
MaybeNamedParamNoAttr ( `,` MaybeNamedParamNoAttr )* `,`?

MaybeNamedParamNoAttr ->
( ( IDENTIFIER | `_` ) `:` )? Type
```

Note that this means that:

- Attributes are not allowed on parameters of these traits. This remains unchanged from the current situation. The following will not work:
```rust
fn test(x: impl Fn(#[allow(unused)] msg: String, priority: usize), y: usize) { }
```
Note that attributes are already allowed on `fn` pointers:
```rust
fn test(x: fn(#[allow(unused)] msg: String, priority: usize), y: usize) { }
```
The reason why attributes are not allowed is to keep this RFC and the implementation simple, and because I don't see a use for them.
Comment thread
JonathanBrouwer marked this conversation as resolved.
Outdated

- This syntax does not match that of `fn` pointers exactly.
For historic reasons, the following `fn` pointer type is allowed:
```rust
#[cfg(false)]
type T = fn(mut x: (), &x: (), &&x: (), false: (), &_: (), &true: ());
```
But this RFC proposes that the following `impl Fn` type is not allowed:

```rust
#[cfg(false)]
impl Fn(mut x: (), &x: (), &&x: (), false: (), &_: (), &true: ());
```

The names of function parameters are limited to ``IDENTIFIER | `_` ``.
The reason why we don't match the syntax of `fn` pointer types is because the syntax was a historic mistake and we should not repeat that.

## Drawbacks
[drawbacks]: #drawbacks

* This makes the syntax of `impl Fn` and friends slightly more complicated
* This keeps the syntax of `impl Fn` and friends inconsistent with that of `fn` pointers, as to avoid the historic mistakes mentioned in the [reference level explanation](#reference-level-explanation).

## Rationale and alternatives
[rationale-and-alternatives]: #rationale-and-alternatives

* An alternative would be to match the `fn` pointer syntax perfectly. This would make the implementation more complicated, without much benefit other than consistency.
* This needs to be implemented in the language, it cannot be provided by a macro or library as it
Comment thread
JonathanBrouwer marked this conversation as resolved.
Outdated
* This makes Rust code easier to read, as it adds better ways to document function signatures.

## Prior art
[prior-art]: #prior-art

In Rust, this is already allowed in `fn` pointers:
```rust
type LogFunction = fn(msg: String, priority: usize);
```

In TypeScript:
```ts
type LogFunction = (msg: string, priority: number) => void;
```

In Kotlin:
```kotlin
fun log(data: String, logFunction: (msg: String, priority: Int) -> Unit) { }
```

## Unresolved questions
[unresolved-questions]: #unresolved-questions

* Should duplicate parameter names be allowed in named fn trait arguments? This is currently allowed for `fn` pointers and other functions without an accompanying `Body`.
Copy link
Copy Markdown
Contributor

@clarfonthey clarfonthey Apr 24, 2026

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

This feels like something that, even if we allow, we should lint against. It feels reasonable to not propose as part of the RFC, but I feel like what makes the most sense is to replicate what traits/functions do but allow for a deny-by-default lint in the future.

View changes since the review

Copy link
Copy Markdown
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Old issue for this - rust-lang/rust#33995.

```rust
type T = fn(x: usize, x: usize);
```
```rust
trait Test {
fn thing(x: usize, x: usize);
}
```

## Future possibilities
[future-possibilities]: #future-possibilities

* We could allow attributes on `impl Fn` parameters in the future.