Expressive Lambda Expressions

[wiseaidev]

• This RFC proposes the introduction of lambda expressions (anonymous functions) in Rust using (arg1, arg2, ..., argn) -> { // code block } notation.
• Lambda expressions provide a concise syntax for defining small, inline functions that can capture surrounding scope variables.
• The feature aims to improve code readability and expressiveness, especially for short, one-time-use functions.
• It aligns with the trend of incorporating functional programming features into modern languages.
• This RFC will not only introduce lambda expressions but also enable developers to implement custom closures from lambda expressions:

// Define a Closure struct
struct MyClosure<F> {
    func: F,
}

impl<F> MyClosure<F> {
    // Create a new Closure instance
    fn new(func: F) -> Self {
        MyClosure { func }
    }

    // Call the closure with arguments
    fn call(&self, x: i32, y: i32) -> i32
    where
        F: Fn(i32, i32) -> i32,
    {
        (self.func)(x, y)
    }
}

fn main() {
    // Create a Closure instance using a lambda expression
    let addition = MyClosure::new((x, y) -> { x + y });
    let result_add = addition.call(6, 3);
    println!("Addition Result: {}", result_add); // Outputs "Addition Result: 9"
}

Rendered

[compiler-errors]

This RFC simultaneously seems to acknowledge the existence of closures (also focusing very narrowly on RFC 1558) while suggesting an addition to the language that is essentially just closures with a new syntax. What is the difference here?

[compiler-errors]

There's not really an argument being made here about why closures are less intuitive or why this new syntax is more intuitive.

[wiseaidev]

It is more like a complement to closures.

[compiler-errors]

Also, honest question -- was this RFC written by a LLM? Sorry if I'm overstepping here, but the way the text of the RFC reads suggests that it may have been.

[wiseaidev]

Also, honest question -- was this RFC written by a LLM?

Off-topic/grey question. Well, I can't reveal all my secrets, but let's just say I'm here to make Rust even more "LLMazing"! 😄 On a serious note, lambda expressions can be a powerful addition to the language, and I'm here to help answer any questions or concerns you might have about this RFC.

[CryZe]

Well the LLM failed to acknowledge (partly) that Rust already has closures and thus there's no section even comparing the two and how the new syntax would be better. Half of this reads like Rust doesn't have any "lambdas" at all.

[wiseaidev]

Y'all pressing L + Ratio for no objective reason. It's making me feel really sad 😢. I'mma heading out 🚪🚶.

[BoxyUwU]

dont waste peoples time by opening PRs generated with ai

[wiseaidev]

Fine. I'mma try harder next time. Besides the point, why does it matter if a PR was generated with AI? Does it suggest that human-generated ones are inherently superior? 🤔

P.S. I see you

P.P.S. On President Joe Biden's life, this PR it is not generated with AI.

[crabdancing]

I... wha? What is happening!? Why, @wiseaidev? Why are you doing this to us? What did we do to deserve this?

[wiseaidev]

I... wha? What is happening!? Why, @wiseaidev?

Nothing worth mentioning, ma fren. I'm simply trying my best to enhance the language, nothing more, nothing less. You're welcome to either accept or decline this PR, and that's perfectly okay. It seems like there's a bit of unnecessary drama surrounding it. And why criticize the author of this PR instead of objectively evaluating the content?

[workingjubilee]

Off-topic/grey question. Well, I can't reveal all my secrets, but let's just say I'm here to make Rust even more "LLMazing"! 😄

Fine. I'mma try harder next time. Besides the point, why does it matter if a PR was generated with AI? Does it suggest that human-generated ones are inherently superior? 🤔

Machines currently cannot claim copyright on their works, and that does not mean that you can then assert copyright by copying their output and saying "I am a human, therefore it is my work." We rely on contributors having title over what they write, or if they reference outside material, declaring the reference.

P.P.S. On President Joe Biden's life, this PR it is not generated with AI.

The question to answer would have been, "did you use any machine-learning-model-driven tools to generate even one byte of output which you then copied into this pull request?"

And why criticize the author of this PR instead of objectively evaluating the content?

Because the content is meaningless if the PR cannot be accepted due to legal considerations. Do you want to be part of a lawsuit for the next 10 years of your life ongoing for every 3 months because you did something as pointless as fudge the truth about whether you used a model to emit text or not?

[wiseaidev]

Machines currently cannot claim copyright on their works

The key question here is: How can you tell if it's written by machines or humans? What GPTs and other language models do is quite similar to what we do naturally – they read documents, learn from them, and predict/write/generate relatively new outputs/content, but at a higher scale. So, you could level the same accusation against humans: learning/using copyrighted material, and you cannot claim anything at all.

[NicoleKai]

@wiseaidev
Thanks for the PR, but please let it go. We already have closures.