Compare commits
4 Commits
fc9cfcdf7c
...
d4fb4680a5
Author | SHA1 | Date | |
---|---|---|---|
![]() |
d4fb4680a5 | ||
![]() |
db83cb7403 | ||
![]() |
1a3e48fddf | ||
![]() |
2967ceb2fc |
@ -3,10 +3,6 @@ name = "deimos"
|
||||
version = "0.1.0"
|
||||
edition = "2021"
|
||||
|
||||
[[bin]]
|
||||
name = "dmc"
|
||||
path = "src/bin/compiler/main.rs"
|
||||
|
||||
[[bin]]
|
||||
name = "dm"
|
||||
path = "src/bin/dvm/main.rs"
|
||||
|
5
dm_std_lib/std/core/array.dm
Normal file
5
dm_std_lib/std/core/array.dm
Normal file
@ -0,0 +1,5 @@
|
||||
ns std::core
|
||||
|
||||
pub int Array<T>
|
||||
|
||||
impl ConstantByteArray(fld raw_address: USize, fld raw_size: USize) : Array<Byte>
|
5
dm_std_lib/std/core/string.dm
Normal file
5
dm_std_lib/std/core/string.dm
Normal file
@ -0,0 +1,5 @@
|
||||
ns std::core
|
||||
|
||||
pub int String
|
||||
|
||||
impl StringImpl(fld bytes: Array<Byte>) : String
|
1
sketching/hello_world.dm
Normal file
1
sketching/hello_world.dm
Normal file
@ -0,0 +1 @@
|
||||
println "Hello, World!"
|
@ -1,22 +0,0 @@
|
||||
use deimos::lexer::tokenize;
|
||||
use deimos::parser::parse;
|
||||
use std::process::exit;
|
||||
|
||||
fn main() {
|
||||
let src = String::from("print 42");
|
||||
let tokenize_result = tokenize(&src);
|
||||
if let Err(e) = tokenize_result {
|
||||
eprintln!("{}", e);
|
||||
exit(1);
|
||||
}
|
||||
let tokens = tokenize_result.unwrap();
|
||||
println!("{:?}", tokens);
|
||||
let parse_result = parse(&tokens);
|
||||
if let Err(e) = parse_result {
|
||||
eprintln!("{}", e);
|
||||
exit(1);
|
||||
}
|
||||
let compilation_unit = parse_result.unwrap();
|
||||
println!("{:?}", compilation_unit);
|
||||
// TODO: compilation_unit to DmModule
|
||||
}
|
8
src/compiler/mod.rs
Normal file
8
src/compiler/mod.rs
Normal file
@ -0,0 +1,8 @@
|
||||
use crate::parser::{DeimosParser, Rule};
|
||||
use crate::vm::lib::DmLib;
|
||||
use pest::Parser;
|
||||
|
||||
pub fn compile(src_unit: &str) -> DmLib {
|
||||
let p = DeimosParser::parse(Rule::compilation_unit, src_unit).unwrap();
|
||||
todo!()
|
||||
}
|
254
src/lexer/mod.rs
254
src/lexer/mod.rs
@ -1,254 +0,0 @@
|
||||
use std::iter::Peekable;
|
||||
use std::str::Chars;
|
||||
|
||||
#[derive(Debug, PartialEq, Eq)]
|
||||
pub enum Token {
|
||||
Namespace,
|
||||
Identifier(String),
|
||||
Public,
|
||||
Module,
|
||||
CurlyOpen,
|
||||
CurlyClose,
|
||||
Interface,
|
||||
Colon,
|
||||
Function,
|
||||
ParenOpen,
|
||||
ParenClose,
|
||||
Enum,
|
||||
LessThan,
|
||||
GreaterThan,
|
||||
Intersection,
|
||||
Union,
|
||||
And,
|
||||
Or,
|
||||
Equals,
|
||||
BigArrow,
|
||||
LittleArrow,
|
||||
Plus,
|
||||
Minus,
|
||||
Dot,
|
||||
Ellipsis,
|
||||
Abstract,
|
||||
NumberLiteral(String),
|
||||
}
|
||||
|
||||
pub fn tokenize(input: &String) -> Result<Vec<Token>, String> {
|
||||
let mut tokens: Vec<Token> = Vec::new();
|
||||
let mut peekable = input.chars().peekable();
|
||||
while let Some(c) = peekable.next() {
|
||||
match c {
|
||||
' ' | '\n' | '\r' | '\t' => { /* ignore */ }
|
||||
'{' => tokens.push(Token::CurlyOpen),
|
||||
'}' => tokens.push(Token::CurlyClose),
|
||||
':' => tokens.push(Token::Colon),
|
||||
'(' => tokens.push(Token::ParenOpen),
|
||||
')' => tokens.push(Token::ParenClose),
|
||||
'<' => tokens.push(Token::LessThan),
|
||||
'>' => tokens.push(Token::GreaterThan),
|
||||
'&' => match peekable.peek() {
|
||||
Some('&') => {
|
||||
let _ = peekable.next();
|
||||
tokens.push(Token::And);
|
||||
}
|
||||
Some(_) | None => tokens.push(Token::Intersection),
|
||||
},
|
||||
'|' => match peekable.next_if_eq(&'|') {
|
||||
Some(_) => tokens.push(Token::Or),
|
||||
None => tokens.push(Token::Union),
|
||||
},
|
||||
'=' => match peekable.next_if_eq(&'>') {
|
||||
Some(_) => tokens.push(Token::BigArrow),
|
||||
None => tokens.push(Token::Equals),
|
||||
},
|
||||
'+' => tokens.push(Token::Plus),
|
||||
'-' => match peekable.next_if_eq(&'>') {
|
||||
Some(_) => tokens.push(Token::LittleArrow),
|
||||
None => tokens.push(Token::Minus),
|
||||
},
|
||||
'.' => {
|
||||
let mut count = 1;
|
||||
while let Some(_) = peekable.next_if_eq(&'.') {
|
||||
count += 1;
|
||||
}
|
||||
match count {
|
||||
1 => tokens.push(Token::Dot),
|
||||
3 => tokens.push(Token::Ellipsis),
|
||||
_ => return Err(String::from("Unexpected number of tokens after '.'")),
|
||||
}
|
||||
}
|
||||
'0'..='9' => {
|
||||
let mut buffer = String::new();
|
||||
buffer.push(c);
|
||||
while let Some(num_char) = peekable.next_if(|c| {
|
||||
c.is_digit(10)
|
||||
|| match c {
|
||||
'_' | 'x' | 'L' | 'd' => true,
|
||||
_ => false,
|
||||
}
|
||||
}) {
|
||||
buffer.push(num_char);
|
||||
}
|
||||
tokens.push(Token::NumberLiteral(buffer));
|
||||
}
|
||||
_ => {
|
||||
if let Some(token) = match_identifier_or_keyword(c, &mut peekable) {
|
||||
tokens.push(token);
|
||||
} else {
|
||||
return Err(String::from(format!("Unexpected token: {}", c)));
|
||||
}
|
||||
}
|
||||
}
|
||||
}
|
||||
Ok(tokens)
|
||||
}
|
||||
|
||||
fn match_identifier_or_keyword(start_char: char, peekable: &mut Peekable<Chars>) -> Option<Token> {
|
||||
if !is_valid_identifier_start_char(start_char) {
|
||||
return None;
|
||||
}
|
||||
|
||||
// append start char
|
||||
let mut buffer = String::new();
|
||||
buffer.push(start_char);
|
||||
|
||||
// munch while we have valid identifier chars
|
||||
while let Some(c) = peekable.next_if(|next_char| is_valid_identifier_char(*next_char)) {
|
||||
buffer.push(c);
|
||||
}
|
||||
|
||||
// match to a keyword if possible, else identifier
|
||||
match buffer.as_str() {
|
||||
"abs" => Some(Token::Abstract),
|
||||
"enum" => Some(Token::Enum),
|
||||
"fn" => Some(Token::Function),
|
||||
"int" => Some(Token::Interface),
|
||||
"mod" => Some(Token::Module),
|
||||
"ns" => Some(Token::Namespace),
|
||||
"pub" => Some(Token::Public),
|
||||
_ => Some(Token::Identifier(buffer)),
|
||||
}
|
||||
}
|
||||
|
||||
fn is_valid_identifier_start_char(c: char) -> bool {
|
||||
match c {
|
||||
'a'..='z' | 'A'..='Z' | '_' => true,
|
||||
_ => false,
|
||||
}
|
||||
}
|
||||
|
||||
fn is_valid_identifier_char(c: char) -> bool {
|
||||
match c {
|
||||
'a'..='z' | 'A'..='Z' | '0'..='9' | '_' => true,
|
||||
_ => false,
|
||||
}
|
||||
}
|
||||
|
||||
#[cfg(test)]
|
||||
mod tests {
|
||||
use super::*;
|
||||
use std::fs::File;
|
||||
use std::io::Read;
|
||||
use std::path::Path;
|
||||
|
||||
#[test]
|
||||
fn simple_ns() {
|
||||
let result = tokenize(&String::from("ns simple")).unwrap();
|
||||
assert_eq!(Token::Namespace, result[0]);
|
||||
assert_eq!(Token::Identifier(String::from("simple")), result[1]);
|
||||
}
|
||||
|
||||
#[test]
|
||||
#[cfg_attr(miri, ignore)]
|
||||
fn simple_ns_file() {
|
||||
let mut src_file = File::open(Path::new("test-data/lexer/simple_ns.dm")).unwrap();
|
||||
let mut src = String::new();
|
||||
let _ = src_file.read_to_string(&mut src);
|
||||
let result = tokenize(&src).unwrap();
|
||||
assert_eq!(Token::Namespace, result[0]);
|
||||
assert_eq!(Token::Identifier(String::from("simple")), result[1]);
|
||||
}
|
||||
|
||||
#[test]
|
||||
fn pub_mod_simple() {
|
||||
let result = tokenize(&String::from("pub mod simple")).unwrap();
|
||||
assert_eq!(Token::Public, result[0]);
|
||||
assert_eq!(Token::Module, result[1]);
|
||||
assert_eq!(Token::Identifier(String::from("simple")), result[2]);
|
||||
}
|
||||
|
||||
#[test]
|
||||
fn curly_open_and_close() {
|
||||
let result = tokenize(&String::from("{ }")).unwrap();
|
||||
assert_eq!(Token::CurlyOpen, result[0]);
|
||||
assert_eq!(Token::CurlyClose, result[1]);
|
||||
}
|
||||
|
||||
#[test]
|
||||
fn simple_int() {
|
||||
let result = tokenize(&String::from("int simple")).unwrap();
|
||||
assert_eq!(Token::Interface, result[0]);
|
||||
assert_eq!(Token::Identifier(String::from("simple")), result[1]);
|
||||
}
|
||||
|
||||
#[test]
|
||||
fn ns_pub_mod_simple() {
|
||||
let result = tokenize(&String::from("ns simple_ns\npub mod simple { }")).unwrap();
|
||||
assert_eq!(Token::Namespace, result[0]);
|
||||
assert_eq!(Token::Identifier(String::from("simple_ns")), result[1]);
|
||||
assert_eq!(Token::Public, result[2]);
|
||||
assert_eq!(Token::Module, result[3]);
|
||||
assert_eq!(Token::Identifier(String::from("simple")), result[4]);
|
||||
assert_eq!(Token::CurlyOpen, result[5]);
|
||||
assert_eq!(Token::CurlyClose, result[6]);
|
||||
}
|
||||
|
||||
#[test]
|
||||
fn curly_open_and_close_no_space() {
|
||||
let result = tokenize(&String::from("{}")).unwrap();
|
||||
assert_eq!(Token::CurlyOpen, result[0]);
|
||||
assert_eq!(Token::CurlyClose, result[1]);
|
||||
}
|
||||
|
||||
#[test]
|
||||
fn interface_function() {
|
||||
let result = tokenize(&String::from("fn test(): Test")).unwrap();
|
||||
assert_eq!(Token::Function, result[0]);
|
||||
assert_eq!(Token::Identifier(String::from("test")), result[1]);
|
||||
assert_eq!(Token::ParenOpen, result[2]);
|
||||
assert_eq!(Token::ParenClose, result[3]);
|
||||
assert_eq!(Token::Colon, result[4]);
|
||||
assert_eq!(Token::Identifier(String::from("Test")), result[5]);
|
||||
}
|
||||
|
||||
#[test]
|
||||
fn interface_prop() {
|
||||
let result = tokenize(&String::from("test: Test")).unwrap();
|
||||
assert_eq!(Token::Identifier(String::from("test")), result[0]);
|
||||
assert_eq!(Token::Colon, result[1]);
|
||||
assert_eq!(Token::Identifier(String::from("Test")), result[2]);
|
||||
}
|
||||
|
||||
#[test]
|
||||
fn enum_decl() {
|
||||
let result = tokenize(&String::from("enum Test {}")).unwrap();
|
||||
assert_eq!(Token::Enum, result[0]);
|
||||
assert_eq!(Token::Identifier(String::from("Test")), result[1]);
|
||||
assert_eq!(Token::CurlyOpen, result[2]);
|
||||
assert_eq!(Token::CurlyClose, result[3]);
|
||||
}
|
||||
|
||||
#[test]
|
||||
fn spread_operator() {
|
||||
let result = tokenize(&String::from("{ ...props }")).unwrap();
|
||||
assert_eq!(Token::CurlyOpen, result[0]);
|
||||
assert_eq!(Token::Ellipsis, result[1]);
|
||||
assert_eq!(Token::Identifier(String::from("props")), result[2]);
|
||||
assert_eq!(Token::CurlyClose, result[3]);
|
||||
}
|
||||
|
||||
#[test]
|
||||
fn simple_number() {
|
||||
let result = tokenize(&String::from("123456")).unwrap();
|
||||
assert_eq!(Token::NumberLiteral(String::from("123456")), result[0]);
|
||||
}
|
||||
}
|
@ -1,4 +1,4 @@
|
||||
pub mod lexer;
|
||||
mod compiler;
|
||||
pub mod parser;
|
||||
mod util;
|
||||
pub mod vm;
|
||||
|
@ -1,24 +1,31 @@
|
||||
compilation_unit = { SOI ~ namespace? ~ declaration* ~ EOI }
|
||||
namespace = { "ns" ~ fqn }
|
||||
|
||||
fqn = { identifier ~ ( "::" ~ fqn )* }
|
||||
fqn = { identifier ~ ( "::" ~ identifier )* }
|
||||
identifier = @{ identifier_start_char ~ identifier_char* }
|
||||
identifier_start_char = { 'a'..'z' | 'A'..'Z' | "_" }
|
||||
identifier_char = { 'a'..'z' | 'A'..'Z' | '0'..'9' | "_" }
|
||||
|
||||
declaration = { "decl"? ~ "pub"? ~ ( interface | module | property | function ) }
|
||||
declaration = { decl? ~ pub? ~ ( interface | implementation | module | function ) }
|
||||
decl = { "decl" }
|
||||
pub = { "pub" }
|
||||
|
||||
interface = { "int" ~ identifier ~ generics_declaration? ~ interface_extends_list? ~ "{" ~ declaration* ~ "}"}
|
||||
interface_extends_list = { ":" ~ type ~ ( "+" ~ type )* }
|
||||
interface = { "int" ~ identifier ~ generics_declaration? ~ extends_list? ~ ( "{" ~ declaration* ~ "}" )? }
|
||||
extends_list = { ":" ~ type ~ ( "+" ~ type )* }
|
||||
|
||||
implementation = { "impl" ~ identifier ~ generics_declaration? ~ impl_ctor? ~ extends_list? }
|
||||
impl_ctor = { "(" ~ impl_ctor_args? ~ ")" }
|
||||
impl_ctor_args = { impl_ctor_arg ~ ( "," ~ impl_ctor_arg )* }
|
||||
impl_ctor_arg = { fld? ~ identifier ~ ":" ~ type }
|
||||
fld = { "fld" }
|
||||
|
||||
module = { "mod" ~ identifier ~ "{" ~ declaration* ~ "}" }
|
||||
|
||||
property = { identifier ~ ":" ~ type }
|
||||
|
||||
function = { "fn" ~ generics_declaration? ~ identifier ~ "(" ~ args_list? ~ ")" ~ ( ":" ~ type )? ~ ( "{" ~ "}" )? }
|
||||
|
||||
function = { "fn" ~ generics_declaration? ~ identifier ~ "(" ~ args_list? ~ ")" ~ ( ":" ~ type )? ~ ( function_body | function_equals_body ) }
|
||||
function_equals_body = { "=" ~ expression }
|
||||
function_body = { "{" ~ "}" }
|
||||
function_body = { "{" ~ statement* ~ "}" }
|
||||
|
||||
type = { identifier ~ generics_declaration? }
|
||||
generics_declaration = { "<" ~ identifier ~ ( "," ~ identifier )* ~ ">"}
|
||||
@ -26,12 +33,29 @@ generics_declaration = { "<" ~ identifier ~ ( "," ~ identifier )* ~ ">"}
|
||||
args_list = { arg ~ ( "," ~ arg )* }
|
||||
arg = { identifier ~ ( ":" ~ "..."? ~ type )? }
|
||||
|
||||
expression = { literal }
|
||||
statement = { call_stmt }
|
||||
|
||||
literal = { number_literal }
|
||||
call_stmt = { call_expr }
|
||||
|
||||
call_expr = { call_expr_with_parens | call_expr_no_parens }
|
||||
call_expr_with_parens = { fqn ~ "(" ~ call_args? ~ ")" }
|
||||
call_expr_no_parens = { fqn ~ call_args }
|
||||
call_args = { call_arg ~ ( "," ~ call_arg )* }
|
||||
call_arg = { expression }
|
||||
|
||||
expression = { literal | identifier }
|
||||
|
||||
literal = { number_literal | string_literal }
|
||||
|
||||
number_literal = { int_literal }
|
||||
|
||||
int_literal = { '0'..'9'+ }
|
||||
|
||||
WHITESPACE = { " " | "\t" | "\n" | "\r" }
|
||||
string_literal = ${ "\"" ~ string_inner ~ "\"" }
|
||||
string_inner = @{ string_char* }
|
||||
string_char = {
|
||||
!( "\"" | "\\" ) ~ ANY
|
||||
| "\\" ~ ( "\"" | "\\" | "/" | "b" | "f" | "n" | "r" | "t" )
|
||||
| "\\" ~ ( "u" ~ ASCII_HEX_DIGIT{4} )
|
||||
}
|
||||
|
||||
WHITESPACE = _{ " " | "\t" | "\n" | "\r" }
|
@ -1,18 +1,5 @@
|
||||
mod types;
|
||||
|
||||
use crate::lexer::Token;
|
||||
use crate::parser::types::AstNode;
|
||||
use pest::Parser;
|
||||
use pest_derive::Parser;
|
||||
|
||||
#[derive(Parser)]
|
||||
#[grammar = "parser/deimos.pest"]
|
||||
struct DeimosParser;
|
||||
|
||||
pub fn parse(tokens: &Vec<Token>) -> Result<AstNode, String> {
|
||||
let p = DeimosParser::parse(Rule::compilation_unit, "ns std::core")
|
||||
.expect("unable to parse")
|
||||
.next()
|
||||
.unwrap();
|
||||
todo!()
|
||||
}
|
||||
pub struct DeimosParser;
|
||||
|
@ -1,11 +0,0 @@
|
||||
use std::fmt::Debug;
|
||||
|
||||
pub type NodeChildren = Vec<Box<AstNode>>;
|
||||
|
||||
#[derive(Debug)]
|
||||
pub enum AstNode {
|
||||
CompilationUnit(NodeChildren),
|
||||
BlockStatement(NodeChildren),
|
||||
Statement(NodeChildren),
|
||||
Expression(NodeChildren),
|
||||
}
|
Loading…
Reference in New Issue
Block a user