Basic work on syntax framework

2023-09-02 00:46:48 +02:00 · 2023-09-02 00:46:48 +02:00 · d588c6b230
parent 74b06d60c5
commit d588c6b230
9 changed files with 251 additions and 170 deletions
--- a/Cargo.lock
+++ b/Cargo.lock
@ -13,24 +13,43 @@ dependencies = [
 "version_check",
 ]

-[[package]]
-name = "allocator-api2"
-version = "0.2.16"
-source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "0942ffc6dcaadf03badf6e6a2d0228460359d5e34b57ccdc720b7382dfbd5ec5"
-
 [[package]]
 name = "beef"
 version = "0.5.2"
 source = "registry+https://github.com/rust-lang/crates.io-index"
 checksum = "3a8241f3ebb85c056b509d4327ad0358fbbba6ffb340bf388f26350aeda225b1"

+[[package]]
+name = "bumpalo"
+version = "3.13.0"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "a3e2c3daef883ecc1b5d58c15adae93470a91d425f3532ba1695849656af3fc1"
+
+[[package]]
+name = "cc"
+version = "1.0.83"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "f1174fb0b6ec23863f8b971027804a42614e347eafb0a95bf0b12cdae21fc4d0"
+dependencies = [
+ "libc",
+]
+
 [[package]]
 name = "cfg-if"
 version = "1.0.0"
 source = "registry+https://github.com/rust-lang/crates.io-index"
 checksum = "baf1de4339761588bc0619e3cbc0120ee582ebb74b53b4efbf79117bd2da40fd"

+[[package]]
+name = "chumsky"
+version = "1.0.0-alpha.4"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "cc3172a80699de358070dd99f80ea8badc6cdf8ac2417cb5a96e6d81bf5fe06d"
+dependencies = [
+ "hashbrown",
+ "stacker",
+]
+
 [[package]]
 name = "fnv"
 version = "1.0.7"
@ -38,24 +57,21 @@ source = "registry+https://github.com/rust-lang/crates.io-index"
 checksum = "3f9eec918d3f24069decb9af1554cad7c880e2da24a9afd88aca000531ab82c1"

 [[package]]
-name = "getrandom"
-version = "0.2.10"
+name = "hashbrown"
+version = "0.13.2"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "be4136b2a15dd319360be1c07d9933517ccf0be8f16bf62a3bee4f0d618df427"
+checksum = "43a3c133739dddd0d2990f9a4bdf8eb4b21ef50e4851ca85ab661199821d510e"
 dependencies = [
- "cfg-if",
- "libc",
- "wasi",
+ "ahash",
 ]

 [[package]]
-name = "hashbrown"
-version = "0.14.0"
+name = "lasso"
+version = "0.7.2"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "2c6201b9ff9fd90a5a3bac2e56a830d0caa509576f0e503818ee82c181b3437a"
+checksum = "4644821e1c3d7a560fe13d842d13f587c07348a1a05d3a797152d41c90c56df2"
 dependencies = [
- "ahash",
- "allocator-api2",
+ "hashbrown",
 ]

 [[package]]
@ -102,12 +118,6 @@ version = "1.18.0"
 source = "registry+https://github.com/rust-lang/crates.io-index"
 checksum = "dd8b5dd2ae5ed71462c540258bedcb51965123ad7e7ccf4b9a8cafaa4a63576d"

-[[package]]
-name = "ppv-lite86"
-version = "0.2.17"
-source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "5b40af805b3121feab8a3c29f04d8ad262fa8e0561883e7653e024ae4479e6de"
-
 [[package]]
 name = "proc-macro2"
 version = "1.0.66"
@ -117,6 +127,15 @@ dependencies = [
 "unicode-ident",
 ]

+[[package]]
+name = "psm"
+version = "0.1.21"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "5787f7cda34e3033a72192c018bc5883100330f362ef279a8cbccfce8bb4e874"
+dependencies = [
+ "cc",
+]
+
 [[package]]
 name = "quote"
 version = "1.0.33"
@ -126,36 +145,6 @@ dependencies = [
 "proc-macro2",
 ]

-[[package]]
-name = "rand"
-version = "0.8.5"
-source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "34af8d1a0e25924bc5b7c43c079c942339d8f0a8b57c39049bef581b46327404"
-dependencies = [
- "libc",
- "rand_chacha",
- "rand_core",
-]
-
-[[package]]
-name = "rand_chacha"
-version = "0.3.1"
-source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "e6c10a63a0fa32252be49d21e7709d4d4baf8d231c2dbce1eaa8141b9b127d88"
-dependencies = [
- "ppv-lite86",
- "rand_core",
-]
-
-[[package]]
-name = "rand_core"
-version = "0.6.4"
-source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "ec0be4795e2f6a28069bec0b5ff3e2ac9bafc99e6a9a7dc3547996c5c816922c"
-dependencies = [
- "getrandom",
-]
-
 [[package]]
 name = "regex-syntax"
 version = "0.6.29"
@ -166,9 +155,23 @@ checksum = "f162c6dd7b008981e4d40210aca20b4bd0f9b60ca9271061b07f78537722f2e1"
 name = "rhea"
 version = "0.1.0"
 dependencies = [
- "hashbrown",
+ "bumpalo",
+ "chumsky",
+ "lasso",
 "logos",
- "rand",
+]
+
+[[package]]
+name = "stacker"
+version = "0.1.15"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "c886bd4480155fd3ef527d45e9ac8dd7118a898a46530b7b94c3e21866259fce"
+dependencies = [
+ "cc",
+ "cfg-if",
+ "libc",
+ "psm",
+ "winapi",
 ]

 [[package]]
@ -195,7 +198,23 @@ source = "registry+https://github.com/rust-lang/crates.io-index"
 checksum = "49874b5167b65d7193b8aba1567f5c7d93d001cafc34600cee003eda787e483f"

 [[package]]
-name = "wasi"
-version = "0.11.0+wasi-snapshot-preview1"
+name = "winapi"
+version = "0.3.9"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "9c8d87e72b64a3b4db28d11ce29237c246188f4f51057d65a7eab63b7987e423"
+checksum = "5c839a674fcd7a98952e593242ea400abe93992746761e38641405d28b00f419"
+dependencies = [
+ "winapi-i686-pc-windows-gnu",
+ "winapi-x86_64-pc-windows-gnu",
+]
+
+[[package]]
+name = "winapi-i686-pc-windows-gnu"
+version = "0.4.0"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "ac3b87c63620426dd9b991e5ce0329eff545bccbbb34f3be09ff6fb6ab51b7b6"
+
+[[package]]
+name = "winapi-x86_64-pc-windows-gnu"
+version = "0.4.0"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "712e227841d057c1ee1cd2fb22fa7e5a5461ae8e48fa2ca79ec42cfc1931183f"
--- a/Cargo.toml
+++ b/Cargo.toml
@ -1,11 +1,10 @@
 [package]
-name = "rhea"
+name    = "rhea"
 version = "0.1.0"
 edition = "2021"

-# See more keys and their definitions at https://doc.rust-lang.org/cargo/reference/manifest.html
-
 [dependencies]
-logos = "*"
-hashbrown = "*"
-rand = "*"
+bumpalo = "3"
+chumsky = "1.0.0-alpha"
+lasso   = "0.7"
+logos   = "0.13"
--- a/src/main.rs
+++ b/src/main.rs
@ -1,61 +1,25 @@
 // Rhea

+use bumpalo::Bump;
 use logos::Logos;
-use memory::{Variable, VariableHashmap, VariableType};
+use std::io::{stdin, Read};
+use utils::default;

-mod memory;
+mod syntax;
+mod utils;

-#[derive(Logos, Debug, PartialEq)]
-#[logos(skip r"[ \t\n\f]+")] // Ignore this regex pattern between tokens
-enum Token {
-    #[token(".")]
-    Period,
+fn main() -> Result<(), Box<dyn std::error::Error>> {
+    let mut buf = default();
+    stdin().read_to_string(&mut buf)?;

-    #[token("(")]
-    LeftParen,
+    let lexer = syntax::token::Token::lexer_with_extras(&buf, default());
+    let arena = Bump::new();
+    syntax::parser::parse(
+        lexer
+            .spanned()
+            .map(|(item, span)| (item.unwrap_or(syntax::token::Token::Invalid), span.into())),
+        &arena,
+    );

-    #[token("{")]
-    LeftBrace,
-
-    #[token(")")]
-    RightParen,
-
-    #[token("}")]
-    RightBrace,
-
-    #[token("include")]
-    Include,
-
-    #[token("=")]
-    Equals,
-
-    #[token(";")]
-    Semicolon,
-
-    #[token("\"")]
-    Quote,
-
-    // Or regular expressions.
-    #[regex("[a-zA-Z]+")]
-    Text,
-}
-
-fn main() {
-    let mut memmap = Vec::new();
-    let mut varmap: VariableHashmap = hashbrown::HashMap::new();
-
-    let variable_name = "abcd".to_string();
-    let variable_type = Some(memory::VariableType::Signed);
-    let constant = false;
-    let length = 8;
-    let abc = Variable::new(&mut memmap, variable_type, constant, length);
-
-    varmap.insert(variable_name, abc);
-
-    let mut lex = Token::lexer(include_str!("../assets/examples/library.rhea"));
-    for x in lex {
-        println!("{:?}", x);
-    }
-
-    //todo compile to hb
+    Ok(())
 }
--- a/src/memory.rs
+++ b/src/memory.rs
@ -1,53 +0,0 @@
-use std::ops::Range;
-
-// I do not like std
-use hashbrown::HashMap;
-
-pub type MemoryMap = Vec<(u64, u64)>;
-
-pub type VariableHashmap = HashMap<String, Variable>;
-
-pub enum VariableType {
-    Unsigned,
-    Signed,
-    String,
-}
-
-pub struct Variable {
-    vtype: VariableType,
-    constant: bool,
-    memory_addr: u64,
-    length: u64,
-}
-impl Variable {
-    pub fn new(
-        memmap: &mut MemoryMap,
-        variable_type: Option<VariableType>,
-        constant: bool,
-        length: u64,
-    ) -> Self {
-        let addr = get_random_addr_and_validate(memmap, length);
-        //todo type guessing
-        let vtype = VariableType::String;
-        Self {
-            vtype,
-            constant,
-            memory_addr: addr,
-            length,
-        }
-    }
-}
-
-use rand::prelude::*;
-
-pub fn get_random_addr_and_validate(memmap: &mut MemoryMap, length: u64) -> u64 {
-    let raddr: u64 = rand::random();
-    // TODO: validate fr later
-    // for (x, _) in memmap.iter() {
-    //     if *x == raddr {
-    //     }
-    // }
-
-    memmap.push((raddr, raddr + length));
-    raddr
-}
--- a/src/syntax/ast.rs
+++ b/src/syntax/ast.rs
@ -0,0 +1,39 @@
+use lasso::Spur;
+
+use super::token::IntLit;
+
+#[derive(Clone, Copy, Debug, PartialEq, Eq)]
+pub enum Expr<'a> {
+    Binary(BinaryOperator, &'a Self, &'a Self),
+    Unary(UnaryOperator, &'a Self),
+    Literal(Literal),
+}
+
+#[derive(Clone, Copy, Debug, PartialEq, Eq)]
+pub enum Literal {
+    String(Spur),
+    Integer(IntLit),
+}
+
+#[derive(Clone, Copy, Debug, PartialEq, Eq)]
+pub enum BinaryOperator {
+    Plus,
+    Minus,
+    Star,
+    Slash,
+    And,
+    Or,
+    Lt,
+    Gt,
+    Equ,
+    Nequ,
+    LtEqu,
+    GtEqu,
+}
+
+#[derive(Clone, Copy, Debug, PartialEq, Eq)]
+pub enum UnaryOperator {
+    Not,
+    Neg,
+    Deref,
+}
--- a/src/syntax/mod.rs
+++ b/src/syntax/mod.rs
@ -0,0 +1,3 @@
+pub mod ast;
+pub mod parser;
+pub mod token;
--- a/src/syntax/parser.rs
+++ b/src/syntax/parser.rs
@ -0,0 +1,18 @@
+use super::{ast::Expr, token::Token};
+use bumpalo::Bump;
+use chumsky::{extra::Full, input::Stream, prelude::*};
+
+type Item = (Token, SimpleSpan);
+type Extra<'s> = Full<Rich<'s, Item, SimpleSpan>, State<'s>, ()>;
+struct State<'a> {
+    arena: &'a Bump,
+}
+
+pub fn parse(input: impl Iterator<Item = Item>, arena: &Bump) {
+    expr().parse_with_state(Stream::from_iter(input), &mut State { arena });
+}
+
+fn expr<'s, 'a, I: Iterator<Item = (Token, SimpleSpan)> + 's>(
+) -> impl Parser<'s, Stream<I>, Expr<'a>, Extra<'s>> {
+    todo()
+}
--- a/src/syntax/token.rs
+++ b/src/syntax/token.rs
@ -0,0 +1,88 @@
+use lasso::Spur;
+
+use {lasso::Rodeo, logos::Logos};
+
+#[derive(Default)]
+pub struct Lextras {
+    pub interner: Rodeo,
+}
+
+#[derive(Copy, Clone, Debug, PartialEq, Eq)]
+pub enum IntLit {
+    Signed(i64),
+    Unsigned(u64),
+}
+
+#[derive(Logos, Copy, Clone, Debug, PartialEq, Eq)]
+#[logos(extras = Lextras)]
+#[logos(skip r"[ \t\n\f]+")]
+#[logos(skip r"-- .*")]
+#[rustfmt::skip]
+pub enum Token {
+    #[token("(")] LeftParen,
+    #[token(")")] RightParen,
+    #[token("{")] LeftCurly,
+    #[token("}")] RightCurly,
+    #[token(".")] Dot,
+    #[token(",")] Comma,
+    #[token(";")] Semicolon,
+    #[token(":=")] //__
+    #[token("≔") ] Bind,
+
+    #[token("←")]  //_
+    #[token("<-")] Set,
+
+    #[token(":>")] Pipe,
+
+    #[token("+")] Plus,
+    #[token("-")] Minus,
+    #[token("*")] Star,
+    #[token("/")] Slash,
+    #[token("&")] And,
+    #[token("|")] Or,
+    #[token("~")] Tilde,
+    
+    #[token("<")] Lt,
+    #[token(">")] Gt,
+    #[token("=")] Equ,
+    #[token("≠") ] //__
+    #[token("/=")] Nequ,
+    #[token("≤") ] //___
+    #[token("<=")] LtEqu,
+    #[token("≥") ] //___,
+    #[token(">=")] GtEqu,
+
+    #[token("match")] Match,
+    #[token("else")] Else,
+    #[token("loop")] Loop,
+    #[token("const")] Const,
+    #[token("var")] Var,
+    #[token("func")] Func,
+    #[token("module")] Module,
+
+    #[regex(
+        r"\p{XID_Start}\p{XID_Continue}*",
+        |l| l.extras.interner.get_or_intern(l.slice())
+    )] Ident(Spur),
+
+    #[regex(
+        "\"[^\"]*\"",
+        |l| {
+            let slice = l.slice();
+            l.extras.interner.get_or_intern(&slice[1..slice.len() - 1])
+        }
+    )] String(Spur),
+
+    #[regex(
+        "-?[0-9]+",
+        |l| {
+            Some(if let Some(slice) = l.slice().strip_prefix('-') {
+                IntLit::Signed(slice.parse::<i64>().ok()?)
+            } else {
+                IntLit::Unsigned(l.slice().parse::<u64>().ok()?)
+            })
+        }
+    )] Int(IntLit),
+
+    Invalid,
+}
--- a/src/utils.rs
+++ b/src/utils.rs
@ -0,0 +1,4 @@
+#[inline(always)]
+pub fn default<T: Default>() -> T {
+    Default::default()
+}