schala/schala-lang/language/src/tokenizing.rs

461 lines
13 KiB
Rust
Raw Normal View History

2021-10-19 22:24:27 -07:00
#![allow(clippy::upper_case_acronyms)]
2021-10-29 19:03:42 -07:00
use std::{
convert::{TryFrom, TryInto},
fmt,
iter::{Iterator, Peekable},
rc::Rc,
};
use itertools::Itertools;
2021-10-19 16:42:48 -07:00
/// A location in a particular source file. Note that the
/// sizes of the internal unsigned integer types limit
/// the size of a source file to 2^32 lines of
/// at most 2^16 characters, which should be plenty big.
#[derive(Debug, Clone, Copy, PartialEq, Default)]
pub struct Location {
2021-10-29 19:03:42 -07:00
pub(crate) line_num: u32,
pub(crate) char_num: u16,
}
impl fmt::Display for Location {
2021-10-29 19:03:42 -07:00
fn fmt(&self, f: &mut fmt::Formatter) -> fmt::Result {
write!(f, "{}:{}", self.line_num, self.char_num)
}
}
#[derive(Debug, PartialEq, Clone)]
2018-11-16 23:17:34 -08:00
pub enum TokenKind {
2021-10-29 19:03:42 -07:00
Newline,
Semicolon,
LParen,
RParen,
LSquareBracket,
RSquareBracket,
LAngleBracket,
RAngleBracket,
LCurlyBrace,
RCurlyBrace,
Pipe,
Backslash,
AtSign,
Comma,
Period,
Colon,
Underscore,
Slash,
Equals,
Operator(Rc<String>),
DigitGroup(Rc<String>),
HexLiteral(Rc<String>),
BinNumberSigil,
StrLiteral { s: Rc<String>, prefix: Option<Rc<String>> },
Identifier(Rc<String>),
Keyword(Kw),
EOF,
Error(String),
}
2018-11-16 23:17:34 -08:00
use self::TokenKind::*;
2018-11-16 23:17:34 -08:00
impl fmt::Display for TokenKind {
2021-10-29 19:03:42 -07:00
fn fmt(&self, f: &mut fmt::Formatter) -> fmt::Result {
match self {
&Operator(ref s) => write!(f, "Operator({})", **s),
&DigitGroup(ref s) => write!(f, "DigitGroup({})", s),
&HexLiteral(ref s) => write!(f, "HexLiteral({})", s),
&StrLiteral { ref s, .. } => write!(f, "StrLiteral({})", s),
&Identifier(ref s) => write!(f, "Identifier({})", s),
&Error(ref s) => write!(f, "Error({})", s),
other => write!(f, "{:?}", other),
}
2018-03-02 22:11:25 -08:00
}
}
#[derive(Debug, Clone, Copy, PartialEq)]
pub enum Kw {
2021-10-29 19:03:42 -07:00
If,
Then,
Else,
Is,
Func,
For,
While,
Let,
In,
Mut,
Return,
2021-11-01 00:14:15 -07:00
Continue,
Break,
2021-10-29 19:03:42 -07:00
Alias,
Type,
SelfType,
SelfIdent,
Interface,
Impl,
True,
False,
Module,
Import,
}
2021-10-14 02:47:19 -07:00
impl TryFrom<&str> for Kw {
2021-10-29 19:03:42 -07:00
type Error = ();
fn try_from(value: &str) -> Result<Self, Self::Error> {
Ok(match value {
"if" => Kw::If,
"then" => Kw::Then,
"else" => Kw::Else,
"is" => Kw::Is,
"fn" => Kw::Func,
"for" => Kw::For,
"while" => Kw::While,
"let" => Kw::Let,
"in" => Kw::In,
"mut" => Kw::Mut,
"return" => Kw::Return,
2021-11-01 00:14:15 -07:00
"break" => Kw::Break,
"continue" => Kw::Continue,
2021-10-29 19:03:42 -07:00
"alias" => Kw::Alias,
"type" => Kw::Type,
"Self" => Kw::SelfType,
"self" => Kw::SelfIdent,
"interface" => Kw::Interface,
"impl" => Kw::Impl,
"true" => Kw::True,
"false" => Kw::False,
"module" => Kw::Module,
"import" => Kw::Import,
_ => return Err(()),
})
}
}
2019-10-22 02:11:49 -07:00
#[derive(Debug, Clone, PartialEq)]
pub struct Token {
2021-10-29 19:03:42 -07:00
pub kind: TokenKind,
pub(crate) location: Location,
}
impl Token {
2021-10-29 19:03:42 -07:00
pub fn to_string_with_metadata(&self) -> String {
format!("{}({})", self.kind, self.location)
}
2021-10-29 19:03:42 -07:00
pub fn get_kind(&self) -> TokenKind {
self.kind.clone()
}
}
2021-10-29 19:03:42 -07:00
const OPERATOR_CHARS: [char; 17] =
['!', '$', '%', '&', '*', '+', '-', '.', ':', '<', '>', '=', '?', '^', '|', '~', '`'];
fn is_operator(c: &char) -> bool {
2021-10-29 19:03:42 -07:00
OPERATOR_CHARS.iter().any(|x| x == c)
}
type CharData = (usize, usize, char);
2018-03-02 02:57:04 -08:00
pub fn tokenize(input: &str) -> Vec<Token> {
2021-10-29 19:03:42 -07:00
let mut tokens: Vec<Token> = Vec::new();
let mut input = Iterator::intersperse(input.lines().enumerate(), (0, "\n"))
.flat_map(|(line_idx, line)| line.chars().enumerate().map(move |(ch_idx, ch)| (line_idx, ch_idx, ch)))
.peekable();
while let Some((line_num, char_num, c)) = input.next() {
let cur_tok_kind = match c {
'/' => match input.peek().map(|t| t.2) {
Some('/') => {
for (_, _, c) in input.by_ref() {
if c == '\n' {
break;
}
}
continue;
2021-10-19 21:27:05 -07:00
}
2021-10-29 19:03:42 -07:00
Some('*') => {
input.next();
let mut comment_level = 1;
while let Some((_, _, c)) = input.next() {
if c == '*' && input.peek().map(|t| t.2) == Some('/') {
input.next();
comment_level -= 1;
} else if c == '/' && input.peek().map(|t| t.2) == Some('*') {
input.next();
comment_level += 1;
}
if comment_level == 0 {
break;
}
}
if comment_level != 0 {
Error("Unclosed comment".to_string())
} else {
continue;
}
}
_ => Slash,
},
c if c.is_whitespace() && c != '\n' => continue,
'\n' => Newline,
';' => Semicolon,
':' => Colon,
',' => Comma,
'(' => LParen,
')' => RParen,
'{' => LCurlyBrace,
'}' => RCurlyBrace,
'[' => LSquareBracket,
']' => RSquareBracket,
'"' => handle_quote(&mut input, None),
'\\' => Backslash,
'@' => AtSign,
c if c.is_digit(10) => handle_digit(c, &mut input),
c if c.is_alphabetic() || c == '_' => handle_alphabetic(c, &mut input),
c if is_operator(&c) => handle_operator(c, &mut input),
unknown => Error(format!("Unexpected character: {}", unknown)),
};
let location =
Location { line_num: line_num.try_into().unwrap(), char_num: char_num.try_into().unwrap() };
tokens.push(Token { kind: cur_tok_kind, location });
}
tokens
}
2021-10-29 19:03:42 -07:00
fn handle_digit(c: char, input: &mut Peekable<impl Iterator<Item = CharData>>) -> TokenKind {
2021-10-19 21:27:05 -07:00
let next_ch = input.peek().map(|&(_, _, c)| c);
if c == '0' && next_ch == Some('x') {
input.next();
2021-10-29 19:03:42 -07:00
let rest: String = input
.peeking_take_while(|&(_, _, ref c)| c.is_digit(16) || *c == '_')
.map(|(_, _, c)| c)
.collect();
2021-10-19 21:27:05 -07:00
HexLiteral(Rc::new(rest))
} else if c == '0' && next_ch == Some('b') {
input.next();
BinNumberSigil
} else {
let mut buf = c.to_string();
2021-10-29 19:03:42 -07:00
buf.extend(input.peeking_take_while(|&(_, _, ref c)| c.is_digit(10)).map(|(_, _, c)| c));
2021-10-19 21:27:05 -07:00
DigitGroup(Rc::new(buf))
}
}
2021-10-29 19:03:42 -07:00
fn handle_quote(
input: &mut Peekable<impl Iterator<Item = CharData>>,
quote_prefix: Option<&str>,
) -> TokenKind {
let mut buf = String::new();
loop {
match input.next().map(|(_, _, c)| c) {
Some('"') => break,
Some('\\') => {
let next = input.peek().map(|&(_, _, c)| c);
if next == Some('n') {
input.next();
buf.push('\n')
} else if next == Some('"') {
input.next();
buf.push('"');
} else if next == Some('t') {
input.next();
buf.push('\t');
}
}
Some(c) => buf.push(c),
None => return TokenKind::Error("Unclosed string".to_string()),
}
}
2021-10-29 19:03:42 -07:00
TokenKind::StrLiteral { s: Rc::new(buf), prefix: quote_prefix.map(|s| Rc::new(s.to_string())) }
}
2021-10-29 19:03:42 -07:00
fn handle_alphabetic(c: char, input: &mut Peekable<impl Iterator<Item = CharData>>) -> TokenKind {
let mut buf = String::new();
buf.push(c);
let next_is_alphabetic = input.peek().map(|&(_, _, c)| !c.is_alphabetic()).unwrap_or(true);
if c == '_' && next_is_alphabetic {
return TokenKind::Underscore;
}
loop {
match input.peek().map(|&(_, _, c)| c) {
Some(c) if c == '"' => {
input.next();
return handle_quote(input, Some(&buf));
}
Some(c) if c.is_alphanumeric() || c == '_' => {
input.next();
buf.push(c);
}
_ => break,
}
}
2021-10-29 19:03:42 -07:00
match Kw::try_from(buf.as_str()) {
Ok(kw) => TokenKind::Keyword(kw),
Err(()) => TokenKind::Identifier(Rc::new(buf)),
}
}
2021-10-29 19:03:42 -07:00
fn handle_operator(c: char, input: &mut Peekable<impl Iterator<Item = CharData>>) -> TokenKind {
match c {
'<' | '>' | '|' | '.' | '=' => {
let next = &input.peek().map(|&(_, _, c)| c);
let next_is_op = next.map(|n| is_operator(&n)).unwrap_or(false);
if !next_is_op {
return match c {
'<' => LAngleBracket,
'>' => RAngleBracket,
'|' => Pipe,
'.' => Period,
'=' => Equals,
_ => unreachable!(),
};
}
}
2021-10-29 19:03:42 -07:00
_ => (),
};
2021-10-29 19:03:42 -07:00
let mut buf = String::new();
2021-10-29 19:03:42 -07:00
if c == '`' {
loop {
match input.peek().map(|&(_, _, c)| c) {
Some(c) if c.is_alphabetic() || c == '_' => {
input.next();
buf.push(c);
}
Some('`') => {
input.next();
break;
}
_ => break,
}
}
} else {
buf.push(c);
loop {
match input.peek().map(|&(_, _, c)| c) {
Some(c) if is_operator(&c) => {
input.next();
buf.push(c);
}
_ => break,
}
}
}
2021-10-29 19:03:42 -07:00
TokenKind::Operator(Rc::new(buf))
}
#[cfg(test)]
mod schala_tokenizer_tests {
2021-10-29 19:03:42 -07:00
use super::{Kw::*, *};
macro_rules! digit {
($ident:expr) => {
DigitGroup(Rc::new($ident.to_string()))
};
}
macro_rules! ident {
($ident:expr) => {
Identifier(Rc::new($ident.to_string()))
};
}
macro_rules! op {
($ident:expr) => {
Operator(Rc::new($ident.to_string()))
};
}
fn token_kinds(input: &str) -> Vec<TokenKind> {
tokenize(input).into_iter().map(move |tok| tok.kind).collect()
}
#[test]
fn tokens() {
let output = token_kinds("let a: A<B> = c ++ d");
assert_eq!(
output,
vec![
Keyword(Let),
ident!("a"),
Colon,
ident!("A"),
LAngleBracket,
ident!("B"),
RAngleBracket,
Equals,
ident!("c"),
op!("++"),
ident!("d")
]
);
}
#[test]
fn underscores() {
let output = token_kinds("4_8");
assert_eq!(output, vec![digit!("4"), Underscore, digit!("8")]);
let output = token_kinds("aba_yo");
assert_eq!(output, vec![ident!("aba_yo")]);
}
#[test]
fn comments() {
let output = token_kinds("1 + /* hella /* bro */ */ 2");
assert_eq!(output, vec![digit!("1"), op!("+"), digit!("2")]);
let output = token_kinds("1 + /* hella /* bro */ 2");
assert_eq!(output, vec![digit!("1"), op!("+"), Error("Unclosed comment".to_string())]);
//TODO not sure if I want this behavior
let output = token_kinds("1 + /* hella */ bro */ 2");
assert_eq!(
output,
vec![
digit!("1"),
op!("+"),
Identifier(Rc::new("bro".to_string())),
Operator(Rc::new("*".to_string())),
Slash,
DigitGroup(Rc::new("2".to_string()))
]
);
}
#[test]
fn backtick_operators() {
let output = token_kinds("1 `plus` 2");
assert_eq!(output, vec![digit!("1"), op!("plus"), digit!("2")]);
}
#[test]
fn string_literals() {
let output = token_kinds(r#""some string""#);
assert_eq!(output, vec![StrLiteral { s: Rc::new("some string".to_string()), prefix: None }]);
let output = token_kinds(r#"b"some bytestring""#);
assert_eq!(
output,
vec![StrLiteral {
s: Rc::new("some bytestring".to_string()),
prefix: Some(Rc::new("b".to_string()))
}]
);
let output = token_kinds(r#""Do \n \" escapes work\t""#);
assert_eq!(
output,
vec![StrLiteral { s: Rc::new("Do \n \" escapes work\t".to_string()), prefix: None }]
);
}
}