Last active
May 13, 2025 14:03
-
-
Save emadb/aae4e95bcd34c68787389caad7cedc15 to your computer and use it in GitHub Desktop.
Parser Combinators in Rust
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
#[derive(Debug, Clone, PartialEq)] | |
struct ParseError(String); | |
#[derive(Debug, PartialEq)] | |
struct ParseOutput<T> { | |
result: T, | |
rest: String, | |
} | |
type ParseResult<T> = Result<ParseOutput<T>, ParseError>; | |
pub struct Parser<T> { | |
parse_fn: Box<dyn Fn(&str) -> ParseResult<T>>, | |
} | |
impl<T> Parser<T> { | |
fn new<F>(parse_fn: F) -> Self | |
where | |
F: Fn(&str) -> ParseResult<T> + 'static, | |
{ | |
Parser { | |
parse_fn: Box::new(parse_fn), | |
} | |
} | |
fn run(&self, input: &str) -> ParseResult<T> { | |
(self.parse_fn)(input) | |
} | |
} | |
fn parse_char(char_to_match: char) -> Parser<char> { | |
let parse_fn = move |input: &str| { | |
let chars: Vec<char> = input.chars().collect(); | |
let head = chars.first().unwrap(); | |
let tail: &[char] = &chars[1..]; | |
if *head == char_to_match { | |
Ok(ParseOutput::<char> { | |
result: *head, | |
rest: tail.iter().collect::<String>(), | |
}) | |
} else { | |
Err(ParseError("TODO: Error".to_string())) | |
} | |
}; | |
Parser::new(parse_fn) | |
} | |
fn bind<T1, T2>(p1: Parser<T1>, p2: Parser<T2>) -> impl Fn(String) -> ParseResult<(T1, T2)> { | |
move |input: String| -> ParseResult<(T1, T2)> { | |
match (p1.parse_fn)(&input) { | |
Ok(o1) => match (p2.parse_fn)(&o1.rest) { | |
Ok(o2) => Ok(ParseOutput::<(T1, T2)> { | |
result: (o1.result, o2.result), | |
rest: o2.rest, | |
}), | |
Err(e2) => Err(e2), | |
}, | |
Err(e1) => Err(e1), | |
} | |
} | |
} | |
fn between<T1, T2, T3>( | |
p1: Parser<T1>, | |
p2: Parser<T2>, | |
p3: Parser<T3>, | |
) -> impl Fn(String) -> ParseResult<T2> { | |
move |input: String| -> ParseResult<T2> { | |
match (p1.parse_fn)(&input) { | |
Ok(o1) => match (p2.parse_fn)(&o1.rest) { | |
Ok(o2) => match (p3.parse_fn)(&o2.rest) { | |
Ok(o3) => Ok(ParseOutput::<T2> { | |
result: o2.result, | |
rest: o3.rest, | |
}), | |
Err(e3) => Err(e3), | |
}, | |
Err(a2) => Err(a2), | |
}, | |
Err(e1) => Err(e1), | |
} | |
} | |
} | |
fn or_else<T1>(p1: Parser<T1>, p2: Parser<T1>) -> impl Fn(String) -> ParseResult<T1> { | |
move |input: String| -> ParseResult<T1> { | |
match (p1.parse_fn)(&input) { | |
Ok(o1) => Ok(ParseOutput::<T1> { | |
result: o1.result, | |
rest: o1.rest, | |
}), | |
Err(_) => (p2.parse_fn)(&input), | |
} | |
} | |
} | |
#[cfg(test)] | |
mod tests { | |
use super::*; | |
// <134>May 6 17:52:12 127.0.0.1 example of a nice log message\n | |
#[test] | |
fn parse_string() { | |
let input_string = "abcd"; | |
let char_parser = parse_char('a'); | |
let result = char_parser.run(input_string); | |
let expected_parse_output = ParseOutput { | |
result: 'a', | |
rest: String::from("bcd"), | |
}; | |
assert_eq!(result, Ok(expected_parse_output)); | |
} | |
#[test] | |
fn bind_test() { | |
let input_string = String::from("abcd"); | |
let pa: Parser<char> = parse_char('a'); | |
let pb: Parser<char> = parse_char('b'); | |
let result = bind(pa, pb)(input_string); | |
let expected_parse_output = ParseOutput { | |
result: ('a', 'b'), | |
rest: String::from("cd"), | |
}; | |
assert_eq!(result, Ok(expected_parse_output)); | |
} | |
#[test] | |
fn between_test() { | |
let input_string = String::from("<a>"); | |
let pa: Parser<char> = parse_char('<'); | |
let pb: Parser<char> = parse_char('a'); | |
let pc: Parser<char> = parse_char('>'); | |
let result = between(pa, pb, pc)(input_string); | |
let expected_parse_output = ParseOutput { | |
result: 'a', | |
rest: String::from(""), | |
}; | |
assert_eq!(result, Ok(expected_parse_output)); | |
} | |
#[test] | |
fn or_else_test() { | |
let input_string = String::from("abc"); | |
let pa: Parser<char> = parse_char('z'); | |
let pb: Parser<char> = parse_char('a'); | |
let result = or_else(pa, pb)(input_string); | |
let expected_parse_output = ParseOutput { | |
result: 'a', | |
rest: String::from("bc"), | |
}; | |
assert_eq!(result, Ok(expected_parse_output)); | |
} | |
} |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment