moved to workspaces

crates/akailang-parser/Cargo.toml

@@ -0,0 +1,8 @@
+[package]
+name = "akailang-parser"
+version.workspace = true
+edition.workspace = true
+
+[dependencies]
+logos = "0.15.0"
+chumsky = "0.10.1"

crates/akailang-parser/src/abstract_syntax_tree/ast.rs

@@ -0,0 +1,49 @@
+use crate::abstract_syntax_tree::definitions::*;
+
+// Option and Result -> define in Std
+#[derive(Clone, Debug, PartialEq)]
+pub enum Expr {
+    // var/fun_name
+    Ident(Identifier),
+
+    // 4
+    IntLiteral(i64),
+
+    // 12.5
+    FloatLiteral(f64),
+
+    // "eepy"
+    StringLiteral(String),
+
+    // true
+    BoolLiteral(bool),
+
+    // 'c'
+    CharLiteral(char),
+
+    // var x = intPlusOne(12)
+    CallExpr(Call),
+
+    // !x
+    UnaryExpr(Unary),
+
+    // 1 + 11
+    BinaryExpr(Binary),
+
+    // x = 64
+    AssignmentExpr(Assignment),
+
+    // var foo = "bar"
+    VarExpr(Var),
+
+    // if(...) { ... } else { ... }
+    Condition(Condition),
+
+    // fun helloWorld() { ... }
+    FunctionExpr(Function),
+
+    // ->
+    ReturnExpr,
+
+    Error,
+}

crates/akailang-parser/src/abstract_syntax_tree/definitions.rs

@@ -0,0 +1,161 @@
+use std::{ops::Range, rc::Rc};
+
+use crate::abstract_syntax_tree::ast::Expr;
+
+// Abstract Syntax Tree
+
+pub type BlockExpression = Vec<Expr>;
+
+pub type Span = Range<usize>;
+
+#[derive(Clone, Debug, PartialEq)]
+pub enum BinaryOp {
+    // Arithmetic
+    Multiply,
+    Divide,
+    Add,
+    Substract,
+
+    // Comparision
+    Equals,
+    NotEquals,
+    Less,
+    LessEquals,
+    Greater,
+    GreaterEquals,
+
+    // Logical
+    And,
+    Or,
+    // todo: bitwise
+}
+
+#[derive(Clone, Debug, PartialEq)]
+pub enum UnaryOp {
+    Not,
+    Negate,
+}
+
+#[derive(Clone, Debug, PartialEq)]
+pub enum Literal {
+    UnsignedInteger(u64),
+    Bool(bool),
+    Char(char),
+    String(Rc<str>),
+    Int(i64),
+    Float(f64),
+}
+
+#[derive(Clone, Debug, PartialEq)]
+pub enum Type {
+    Integer,
+    Float,
+    Bool,
+    Char,
+    String,
+}
+
+#[derive(Clone, Debug, PartialEq)]
+pub enum Value {
+    Integer(i64),
+    Float(f64),
+    Bool(bool),
+    Char(char),
+    String(String),
+}
+
+impl Value {
+    pub fn is_type(&self, ty: &Type) -> bool {
+        match (ty, self) {
+            (Type::Bool, Value::Bool(_)) => true,
+            (Type::Char, Value::Char(_)) => true,
+            (Type::Integer, Value::Integer(_)) => true,
+            (Type::Float, Value::Float(_)) => true,
+            (Type::String, Value::String(_)) => true,
+            _ => false,
+        }
+    }
+}
+
+//---------------------------------------
+//      Structs
+//---------------------------------------
+#[derive(Clone, Debug, PartialEq)]
+pub struct Identifier(pub String);
+
+#[derive(Clone, Debug, PartialEq)]
+pub struct While {
+    pub condition: Box<Expr>,
+    pub body: BlockExpression,
+}
+
+/// Example: `if (a > 12) { ... }`
+#[derive(Clone, Debug, PartialEq)]
+pub struct Condition {
+    pub condition: Box<Expr>,
+    pub if_body: BlockExpression,
+    pub else_body: Option<BlockExpression>,
+}
+
+/// Example: `x = y`
+#[derive(Clone, Debug, PartialEq)]
+pub struct Assignment {
+    pub target: String,
+    pub value: Box<Expr>,
+}
+
+/// Example: `var String: foo = "Test"`
+#[derive(Clone, Debug, PartialEq)]
+pub struct Var {
+    pub ty: Option<Type>,
+    pub ident: Identifier,
+    pub value: Box<Expr>,
+}
+
+/// Example: `x++`
+#[derive(Clone, Debug, PartialEq)]
+pub struct Unary {
+    pub operator: UnaryOp,
+    pub operand: Box<Expr>,
+}
+
+/// Example: `1 + 1`
+#[derive(Clone, Debug, PartialEq)]
+pub struct Binary {
+    pub lhs: Box<Expr>,
+    pub operator: BinaryOp,
+    pub rhs: Box<Expr>,
+}
+
+/// Represents the Structure of a `Function` in AkaiLang
+///
+/// Examples:
+///```AkaiLang
+///fun helloWorld() {
+///    print("Hello World")
+///}
+///```
+/// <br>
+///
+///```AkaiLang
+///fun returnsIntPlusOne(i: int): int {
+///    -> i + 1
+///}
+///```
+#[derive(Clone, Debug, PartialEq)]
+pub struct Function {
+    pub name: String,
+    pub params: Option<Vec<(String, Type)>>,
+    pub return_type: Option<Type>,
+    pub body: Option<BlockExpression>,
+    pub return_expr: Box<Option<Expr>>, // ' -> (return)'
+}
+
+#[derive(Clone, Debug, PartialEq)]
+pub struct Call {
+    /// name of the function being called
+    pub name: Box<Identifier>,
+
+    /// arguments supplied
+    pub arguments: Vec<Expr>,
+}

crates/akailang-parser/src/abstract_syntax_tree/mod.rs

@@ -0,0 +1,52 @@
+pub mod ast;
+pub mod definitions;
+pub mod parser;
+
+#[macro_export]
+macro_rules! token {
+    (var) => {
+        just(Token::Var)
+    };
+    (=) => {
+        just(Token::Assign)
+    };
+    (if) => {
+        just(Token::If)
+    };
+    (else) => {
+        just(Token::Else)
+    };
+    (->) => {
+        just(Token::Return)
+    };
+    (:) => {
+        just(Token::Colon)
+    };
+    (,) => {
+        just(Token::Comma)
+    };
+    (*) => {
+        just(Token::Multiply)
+    };
+    (/) => {
+        just(Token::Divide)
+    };
+    (+) => {
+        just(Token::Add)
+    };
+    (-) => {
+        just(Token::Substract)
+    };
+    (lp) => {
+        just(Token::LParen)
+    };
+    (rp) => {
+        just(Token::RParen)
+    };
+    (lbrace) => {
+        just(Token::LBrace)
+    };
+    (rbrace) => {
+        just(Token::RBrace)
+    };
+}

crates/akailang-parser/src/abstract_syntax_tree/parser.rs

@@ -0,0 +1,324 @@
+use chumsky::{
+    Boxed, ConfigIterParser, IterParser, ParseResult, Parser,
+    combinator::Or,
+    container::Seq,
+    error::{Rich, Simple},
+    extra,
+    input::{Input, Stream, ValueInput},
+    prelude::{choice, end, just, nested_delimiters, recursive, skip_then_retry_until, via_parser},
+    primitive::select,
+    recursive, select, select_ref,
+    span::{self, SimpleSpan},
+    text::{
+        self,
+        ascii::{ident, keyword},
+        newline, whitespace,
+    },
+};
+use logos::{Logos, source};
+
+use crate::{
+    token,
+    {
+        abstract_syntax_tree::{ast::Expr, definitions::*},
+        lexer::tokens::Token,
+    },
+};
+
+// goal of parsing is to construct an abstract syntax tree
+
+pub fn parse(source: &str) -> Result<Vec<Expr>, Vec<Rich<'_, Token>>> {
+    let token_iter = Token::lexer(source)
+        .spanned()
+        .map(|(token, span)| (token.unwrap_or(Token::Error), span.into()));
+    let end_of_input: SimpleSpan = (0..source.len()).into();
+    let token_stream = Stream::from_iter(token_iter).map(
+        (0..end_of_input.into_iter().len()).into(),
+        |(t, s): (_, _)| (t, s),
+    );
+
+    parser().parse(token_stream).into_result()
+}
+
+fn parser<'src, I>() -> impl Parser<'src, I, Vec<Expr>, extra::Err<Rich<'src, Token>>>
+where
+    I: ValueInput<'src, Token = Token, Span = SimpleSpan>,
+{
+    let ident = select! { Token::Identifier(s) => s, }.labelled("identifier");
+
+    let expr = recursive(|expr| {
+        let atom = select! {
+            Token::FloatLiteral(x) => Expr::FloatLiteral(x),
+            Token::IntLiteral(x) => Expr::IntLiteral(x),
+            Token::BoolLiteral(x) => Expr::BoolLiteral(x),
+            Token::StringLiteral(s) => Expr::StringLiteral(s),
+        }
+        .labelled("value")
+        .or(expr.clone().delimited_by(token!(lp), token!(rp)));
+
+        // Product
+        let mul_div = atom.clone().foldl(
+            choice((
+                token!(*).to(BinaryOp::Multiply),
+                token!(/).to(BinaryOp::Divide),
+            ))
+            .then(atom)
+            .then_ignore(just(Token::NewLine).or_not())
+            .repeated(),
+            |lhs, (op, rhs)| {
+                Expr::BinaryExpr(Binary {
+                    lhs: Box::new(lhs),
+                    operator: op,
+                    rhs: Box::new(rhs),
+                })
+            },
+        );
+
+        // Sum
+        let add_sub = mul_div.clone().foldl(
+            choice((
+                token!(+).to(BinaryOp::Add),
+                token!(-).to(BinaryOp::Substract),
+            ))
+            .then(mul_div)
+            .then_ignore(just(Token::NewLine).or_not())
+            .repeated(),
+            |lhs, (op, rhs)| {
+                Expr::BinaryExpr(Binary {
+                    lhs: Box::new(lhs),
+                    operator: op,
+                    rhs: Box::new(rhs),
+                })
+            },
+        );
+
+        let assign_expr = ident
+            .then_ignore(token!(=))
+            .then(expr.clone())
+            .map(|(name, value)| {
+                Expr::AssignmentExpr(Assignment {
+                    target: name,
+                    value: Box::new(value),
+                })
+            });
+
+        assign_expr.or(add_sub)
+    });
+
+    let decl = recursive(|decl| {
+        let var = token!(var)
+            .ignore_then(ident)
+            .then_ignore(token!(=))
+            .then(expr.clone())
+            .then_ignore(just(Token::NewLine).or_not())
+            .map(|(name, rhs)| {
+                Expr::VarExpr(Var {
+                    ty: None,
+                    ident: Identifier(String::from(name)),
+                    value: Box::new(rhs),
+                })
+            });
+
+        let type_parser = choice((
+            just(Token::IntType).to(Type::Integer),
+            just(Token::FloatType).to(Type::Float),
+            just(Token::BoolType).to(Type::Bool),
+            just(Token::StringType).to(Type::String),
+        ));
+
+        let return_type_parser = token!(:).ignore_then(type_parser.clone()).or_not();
+
+        let return_expr = token!(->).ignore_then(expr.clone()).map(|expr| expr);
+
+        //---------------------------------------------------------------------------------------
+        // Function Parser
+        //---------------------------------------------------------------------------------------
+        let fun = just(Token::Fun)
+            .ignore_then(ident) // function name
+            .then(
+                // arguments
+                ident
+                    .then_ignore(token!(:))
+                    .then(type_parser.clone())
+                    .separated_by(token!(,))
+                    .allow_trailing()
+                    .collect::<Vec<_>>()
+                    .delimited_by(token!(lp), token!(rp))
+                    .or_not(),
+            )
+            .then(return_type_parser.clone()) // return type
+            .then_ignore(token!(lbrace)) // {
+            .then_ignore(just(Token::NewLine).repeated().or_not())
+            .then(
+                expr.clone()
+                    .then_ignore(just(Token::NewLine))
+                    .repeated()
+                    .collect::<Vec<Expr>>()
+                    .then(return_expr.or_not()),
+            )
+            .then_ignore(just(Token::NewLine).repeated().or_not())
+            .then_ignore(token!(rbrace)) // }
+            .map(|(((name, params), return_ty), (stmts, return_value))| {
+                Expr::FunctionExpr(Function {
+                    name: name,
+                    params,
+                    return_type: return_ty,
+                    body: Some(stmts),
+                    return_expr: Box::new(return_value),
+                })
+            });
+
+        var.or(fun).or(expr)
+    });
+
+    decl.clone()
+        .then_ignore(just(Token::NewLine).repeated())
+        .repeated()
+        .collect()
+}
+
+#[cfg(test)]
+mod tests {
+    use super::*;
+
+    #[test]
+    fn test_unary_expr() {
+        let negate_two = parse("!2");
+        assert!(negate_two.is_ok());
+        assert_eq!(
+            negate_two.clone().unwrap(),
+            vec![Expr::UnaryExpr(Unary {
+                operator: UnaryOp::Negate,
+                operand: Box::new(Expr::IntLiteral(2)),
+            })]
+        )
+    }
+
+    #[test]
+    fn test_bool() {
+        let var_bool = parse("var isUwU = true");
+        assert!(var_bool.is_ok());
+        assert_eq!(
+            var_bool.clone().unwrap(),
+            vec![Expr::VarExpr(Var {
+                ty: None,
+                ident: Identifier(String::from("isUwU")),
+                value: Box::new(Expr::BoolLiteral(true))
+            })]
+        )
+    }
+
+    #[test]
+    fn test_binary_expr() {
+        let sum = parse("1 + 2");
+        assert!(sum.is_ok());
+        assert_eq!(
+            sum.clone().unwrap(),
+            vec![Expr::BinaryExpr(Binary {
+                lhs: Box::new(Expr::IntLiteral(1)),
+                operator: BinaryOp::Add,
+                rhs: Box::new(Expr::IntLiteral(2))
+            })]
+        )
+    }
+
+    #[test]
+    fn test_variable_decl() {
+        let var_without_expl_type = parse("var x = 12");
+        assert!(var_without_expl_type.is_ok());
+        assert_eq!(
+            var_without_expl_type.clone().unwrap(),
+            vec![Expr::VarExpr(Var {
+                ty: None,
+                ident: Identifier(String::from("x")),
+                value: Box::new(Expr::IntLiteral(12))
+            })]
+        )
+    }
+
+    #[test]
+    fn test_assignment() {
+        let assign = parse("x = 12");
+        assert!(assign.is_ok());
+        assert_eq!(
+            assign.clone().unwrap(),
+            vec![Expr::AssignmentExpr(Assignment {
+                target: String::from("x"),
+                value: Box::new(Expr::IntLiteral(12))
+            })]
+        )
+    }
+
+    #[test]
+    fn test_function_decl() {
+        // test without a body or args
+        let empty_fun = parse("fun helloWorld() { }");
+        assert!(empty_fun.is_ok());
+        assert_eq!(
+            empty_fun.clone().unwrap(),
+            vec![Expr::FunctionExpr(Function {
+                name: String::from("helloWorld"),
+                params: Some([].to_vec()),
+                return_type: None,
+                body: Some([].to_vec()),
+                return_expr: Box::new(None),
+            })]
+        );
+
+        // tests if empty new lines within a function works
+        let empty_fun_with_new_lines = parse(
+            r"fun emptyMulLines() {
+
+            }
+        ",
+        );
+        assert_eq!(
+            empty_fun_with_new_lines.clone().unwrap(),
+            vec![Expr::FunctionExpr(Function {
+                name: String::from("emptyMulLines"),
+                params: Some([].to_vec()),
+                return_type: None,
+                body: Some([].to_vec()),
+                return_expr: Box::new(None),
+            })]
+        );
+
+        // tests for return expr in functions
+        let fun_that_returns_int = parse(
+            r"fun returnsInt(): int {
+                -> 12 
+            }
+        ",
+        );
+        assert_eq!(
+            fun_that_returns_int.clone().unwrap(),
+            vec![Expr::FunctionExpr(Function {
+                name: String::from("returnsInt"),
+                params: Some([].to_vec()),
+                return_type: Some(Type::Integer),
+                body: Some([].to_vec()),
+                return_expr: Box::new(Some(Expr::IntLiteral(12))),
+            })]
+        );
+
+        // tests for return expr with previous statements in functions
+        let fun_multi_line_return = parse(
+            r"fun returnsInt(): int {
+                var x = 12
+                x = x + 1
+                -> x 
+            }
+        ",
+        );
+        assert_eq!(
+            fun_multi_line_return.clone().unwrap(),
+            vec![Expr::FunctionExpr(Function {
+                name: String::from("returnsInt"),
+                params: Some([].to_vec()),
+                return_type: Some(Type::Integer),
+                body: Some(vec![Expr::IntLiteral(12)]),
+                return_expr: Box::new(Some(Expr::ReturnExpr)),
+            })]
+        )
+    }
+}

crates/akailang-parser/src/lexer/mod.rs

@@ -0,0 +1 @@
+pub mod tokens;

crates/akailang-parser/src/lexer/tokens.rs

@@ -0,0 +1,205 @@
+use std::fmt;
+
+use logos::Logos;
+
+#[derive(Logos, Debug, Clone, PartialEq)]
+#[regex(r"[\t\f]+", logos::skip)]
+pub enum Token {
+    // Identifier
+    #[regex(r"[a-zA-Z_][a-zA-Z0-9_]*", |lex| lex.slice().to_owned())]
+    Identifier(String),
+
+    // Keywords
+    #[token("fun")]
+    Fun,
+
+    #[token("class")]
+    Class,
+
+    #[token("var")]
+    Var,
+
+    #[token("interface")]
+    Interface,
+
+    #[token("derive")]
+    Derive,
+
+    #[token("impl")]
+    Impl,
+
+    #[token("if")]
+    If,
+
+    #[token("else")]
+    Else,
+
+    #[token("->")]
+    Return,
+
+    #[token("enum")]
+    Enum,
+
+    // Types
+    #[token("int")]
+    IntType,
+
+    #[token("float")]
+    FloatType,
+
+    #[token("bool")]
+    BoolType,
+
+    #[token("String")]
+    StringType,
+
+    // Literals
+    #[regex(r#""([^"\\]|\\.)*""#, |lex| lex.slice().to_owned())]
+    StringLiteral(String),
+
+    #[regex(r"[0-9]+", |lex| lex.slice().parse::<i64>().ok())]
+    IntLiteral(i64),
+
+    #[regex(r"[0-9]+\.[0-9]+", |lex| lex.slice().parse::<f64>().ok())]
+    FloatLiteral(f64),
+
+    #[token("true", |_| true)]
+    #[token("false", |_| false)]
+    BoolLiteral(bool),
+
+    // Operators
+    #[token("=")]
+    Assign,
+
+    #[token("==")]
+    Equals,
+
+    #[token("!=")]
+    NotEquals,
+
+    #[token("<", priority = 2)]
+    Less,
+
+    #[token("<=")]
+    LessEquals,
+
+    #[token(">", priority = 2)]
+    Greater,
+
+    #[token(">=")]
+    GreaterEquals,
+
+    #[token("*")]
+    Multiply,
+
+    #[token("/")]
+    Divide,
+
+    #[token("+")]
+    Add,
+
+    #[token("-")]
+    Substract,
+
+    #[token("&&")]
+    And,
+
+    #[token("||")]
+    Or,
+
+    // Punctiuation
+    #[token("(")]
+    LParen,
+
+    #[token(")")]
+    RParen,
+
+    #[token("{")]
+    LBrace,
+
+    #[token("}")]
+    RBrace,
+
+    #[token("[")]
+    LBracket,
+
+    #[token("]")]
+    RBracket,
+
+    #[token(":")]
+    Colon,
+
+    #[token(",")]
+    Comma,
+
+    #[token(".")]
+    Dot,
+
+    // Special
+    #[regex(r"\n")]
+    NewLine,
+
+    #[regex(r"//[^\r]*", logos::skip)]
+    #[regex(r"/\*([^*]|\*[^/])*\*/", logos::skip)]
+    Comment,
+
+    #[regex(r"[ \t\f]+", logos::skip)]
+    Whitespace,
+
+    Eof,
+
+    Error,
+}
+
+impl fmt::Display for Token {
+    fn fmt(&self, f: &mut fmt::Formatter<'_>) -> fmt::Result {
+        match self {
+            Token::Identifier(ident) => write!(f, "{ident}"),
+            Token::Fun => write!(f, "fun"),
+            Token::Class => write!(f, "class"),
+            Token::Var => write!(f, "var"),
+            Token::Interface => write!(f, "interface"),
+            Token::Derive => write!(f, "derive"),
+            Token::Impl => write!(f, "impl"),
+            Token::If => write!(f, "if"),
+            Token::Else => write!(f, "else"),
+            Token::Return => write!(f, "->"),
+            Token::Enum => write!(f, "enum"),
+            Token::IntType => write!(f, "int"),
+            Token::FloatType => write!(f, "float"),
+            Token::BoolType => write!(f, "bool"),
+            Token::StringType => write!(f, "String"),
+            Token::StringLiteral(s) => write!(f, "{s}"),
+            Token::IntLiteral(i) => write!(f, "{i}"),
+            Token::FloatLiteral(fl) => write!(f, "{fl}"),
+            Token::BoolLiteral(b) => write!(f, "{b}"),
+            Token::Assign => write!(f, "="),
+            Token::Equals => write!(f, "=="),
+            Token::NotEquals => write!(f, "!="),
+            Token::Less => write!(f, "<"),
+            Token::LessEquals => write!(f, "<="),
+            Token::Greater => write!(f, ">"),
+            Token::GreaterEquals => write!(f, ">="),
+            Token::Multiply => write!(f, "*"),
+            Token::Divide => write!(f, "/"),
+            Token::Add => write!(f, "+"),
+            Token::Substract => write!(f, "-"),
+            Token::And => write!(f, "&&"),
+            Token::Or => write!(f, "||"),
+            Token::LParen => write!(f, "("),
+            Token::RParen => write!(f, ")"),
+            Token::LBrace => write!(f, "{{"),
+            Token::RBrace => write!(f, "}}"),
+            Token::LBracket => write!(f, "["),
+            Token::RBracket => write!(f, "]"),
+            Token::Colon => write!(f, ":"),
+            Token::Comma => write!(f, ","),
+            Token::Dot => write!(f, "."),
+            Token::Comment => write!(f, ""),
+            Token::NewLine => write!(f, "\n"),
+            Token::Whitespace => write!(f, ""),
+            Token::Eof => write!(f, ""),
+            Token::Error => write!(f, "<error>"),
+        }
+    }
+}

crates/akailang-parser/src/lib.rs

@@ -0,0 +1,17 @@
+pub mod abstract_syntax_tree;
+pub mod lexer;
+
+pub fn add(left: u64, right: u64) -> u64 {
+    left + right
+}
+
+#[cfg(test)]
+mod tests {
+    use super::*;
+
+    #[test]
+    fn it_works() {
+        let result = add(2, 2);
+        assert_eq!(result, 4);
+    }
+}