Kima

A Programming Language with static types and (currently WIP) algebraic effects.

Getting started

Installing

git clone https://gitlab.com/michalis_pardalos/Kima.git
cd Kima
stack install

Usage

You can run a file with

kima run example.k

or start the REPL with

kima repl

any other options should be documented in

kima --help

Examples

Let\’s get the basic Hello World out of the way

fun main() -> Unit {
    print("Hello World!");
}

We can define our string as a local variable too:

fun main() -> Unit {
    let output: String = "Hello World!";
    print(output);
}

But there\’s no need to say the string is a String. Type inference takes care of that!

fun main() -> Unit {
    let output = "Hello World!";
    print(output);
}

What if we changed out mind about the string we want to print?

fun main() -> Unit {
    let output = "Hello World!";
    output = "Hello Everyone!"
    print(output);
}

Oops! That\’s an error. let creates immutable variables. Use var for mutable variables.

fun main() -> Unit {
    var output = "Hello World!";
    output = "Hello Everyone!"
    print(output);
}

That\’s better! We can also make a function to return a custom greeting

fun greet(name: String) -> String {
    "Hello " + name;
}

fun main() -> Unit {
    let output = greet("everyone!");
    print(output);
}

As you can see, functions return the value of the last statement in their body.

Functions can also take functions as arguments or return them:

fun adder(x: Int) -> (Int) -> Int {
    fun(y: Int) -> Int {
        x + y;
    }
}

fun main() -> {
    let add_3 = adder(3);
    print("3 + 5 = ");
    print(add_3(5));
}

You can define data types like so:

data Email(name: String, server: String, tld: String)

fun to_string(email: Email) -> String {
    email.name + "@" + email.server + "." + email.tld;
}

fun main() -> Unit {
    print(to_string(Email("john.smith", "example", "com")));
}

And more features are coming soon!

Features

Feature	Status
Builtin types	DONE
Higher-order functions	DONE
Product types (records)	DONE
Type checking	DONE
Type inference (bidirectional typing takes care of both)	DONE
Control flow (if, while)	DONE
Pattern matching	Planned
Effect system	Work in Progress
Sum types	Planned
Polymorphic types	Planned
Source position annotations (better error messages)	Planned