from-scratch/lang/tests/tokens.rs
2025-10-28 13:12:27 +01:00

117 lines
2.9 KiB
Rust

#[unsafe(no_mangle)]
extern "C" fn panic() -> ! {
panic!("Called panic from external code.");
}
#[derive(Debug, Clone, Copy, PartialEq, Eq)]
struct Lexeme(u8);
impl Lexeme {
fn lex(&self) -> &'static str {
// SAFETY: lens contains the correct length for each lexeme, and lexemes
// contains pointers to valid 'static UTF-8 data.
unsafe {
core::str::from_utf8_unchecked(
core::slice::from_raw_parts(
(&raw const LEXEMES).add((self.0) as usize).read(),
(&raw const LEXEME_LENS).add((self.0) as usize).read(),
)
)
}
}
}
trait AsLexeme {
fn as_lexeme(self) -> Option<Lexeme>;
}
impl AsLexeme for u8 {
fn as_lexeme(self) -> Option<Lexeme> {
match self {
1.. => Some(Lexeme(self)),
_ => None,
}
}
}
#[allow(dead_code)]
unsafe extern "C" {
unsafe fn tokeniser_init(path: *const i8) -> ();
unsafe fn tokeniser_print() -> ();
unsafe fn is_ident(len: usize) -> bool;
unsafe fn is_number(len: usize) -> bool;
unsafe fn skip_whitespace() -> ();
unsafe fn find_lexeme() -> u8;
static mut LEXEMES: *const u8;
static mut LEXEME_LENS: usize;
static mut NUM_LEXEMES: usize;
static mut TOKENS: u8;
static mut input_file: u32;
static mut buffer: *mut u8;
static mut cursor: usize;
static mut buffer_len: usize;
unsafe fn exit(code: i32) -> !;
}
fn collect_tokens() -> Vec<Lexeme> {
let mut lexemes = Vec::new();
unsafe {
while let Some(lexeme) = find_lexeme().as_lexeme() {
lexemes.push(lexeme);
}
}
lexemes
}
fn main() {
unsafe {
// assert initial state
assert_eq!((&raw const input_file).read(), 0);
assert_eq!((&raw const buffer_len).read(), 0);
assert_eq!((&raw const cursor).read(), 0);
assert_eq!((&raw const buffer).read(), core::ptr::null_mut());
eprint!("Initializing tokeniser.. ");
tokeniser_init(c"tests/tokens/keywords.l".as_ptr());
eprintln!("ok.");
assert_eq!(&collect_tokens()[..], &[
Lexeme(4),
Lexeme(1),
Lexeme(2),
Lexeme(3),
Lexeme(4),
Lexeme(8),
Lexeme(13),
Lexeme(11),
Lexeme(10),
Lexeme(9),
Lexeme(5),
][..]);
eprint!("Initializing tokeniser.. ");
tokeniser_init(c"tests/tokens/delimiters.l".as_ptr());
eprintln!("ok.");
assert_eq!(&collect_tokens()[..], &[
Lexeme(19),
Lexeme(18),
Lexeme(28),
Lexeme(29),
Lexeme(21),
Lexeme(20),
Lexeme(24),
Lexeme(12),
Lexeme(23),
Lexeme(22),
Lexeme(15),
][..]);
eprintln!("Finished tokenising.");
}
}