rust-lib/src/nom/combinators.rs

use nom::IResult;
use nom::bytes::complete::take_while_m_n;
use nom::character::complete::char;
use nom::character::complete::multispace0;
use nom::combinator::eof;
use nom::error::ParseError;
use nom::sequence::delimited;
use nom::sequence::terminated;
use nom::{Input, Parser};

/// Trim leading and trailing whitespace from the input Parser
/// - Parameters
///    - `inner`: The parser to trim
/// - Returns: A parser that trims leading and trailing whitespace from the input and then runs the value from the inner parser
pub fn trim<I, O, F, E: ParseError<I>>(inner: F) -> impl Parser<I, Output = O, Error = E>
where
    I: Input,
    F: Parser<I, Output = O, Error = E>,
    <I as Input>::Item: nom::AsChar,
{
    delimited(multispace0, inner, multispace0)
}

/// Parse a parenthesized expression. This parser will parse an expression that is surrounded by parentheses
/// and will trim the whitespace surrounding the expression.
/// - Parameters
///     - `inner`: The parser to run inside the parentheses
/// - Returns: A parser that parses a parenthesized expression
pub fn parenthesized<I, O, F, E: ParseError<I>>(inner: F) -> impl Parser<I, Output = O, Error = E>
where
    I: Input,
    F: Parser<I, Output = O, Error = E>,
    <I as Input>::Item: nom::AsChar,
{
    delimited(char('('), inner, char(')'))
}

/// Take where the predicate is true and the length is exactly `n`
/// - Parameters
///   - `n`: The length of the string to take
///   - `predicate`: The predicate to call to validate the input
/// - Returns: A parser that takes `n` characters from the input
pub fn take_where<F, I>(n: usize, predicate: F) -> impl FnMut(I) -> IResult<I, I>
where
    I: Input,
    F: Fn(<I as Input>::Item) -> bool,
{
    take_while_m_n(n, n, predicate)
}

/// Parse the inner parser and then the end of the input.
/// Very useful for ensuring that the entire input is consumed.
/// - Parameters
///     - `inner`: The parser to run
/// - Returns: A parser that runs the inner parser and then the end of the input
/// # Example
/// ```
/// use lib::nom::combinators::exhausted;
/// use nom::bytes::complete::{tag};
/// use nom::Parser;
///
/// let input = "test";
/// let (remaining, result) = exhausted(tag::<&str, &str, nom::error::Error<&str>>("test")).parse(input).unwrap();
/// assert_eq!(remaining, "");
/// assert_eq!(result, "test");
/// ```
/// - Fails if the input is not exhausted
/// ```
/// use lib::nom::combinators::exhausted;
/// use nom::bytes::complete::{tag};
/// use nom::Parser;
///
/// let input = "test";
/// assert!(exhausted(tag::<&str, &str, nom::error::Error<&str>>("tes")).parse(input).is_err());
/// ```
pub fn exhausted<F, I, O, E: ParseError<I>>(inner: F) -> impl Parser<I, Output = O, Error = E>
where
    I: Input,
    F: Parser<I, Output = O, Error = E>,
{
    terminated(inner, eof)
}

#[cfg(test)]
mod tests {
    use super::*;
    use nom::bytes::complete::take_while;

    #[test]
    fn test_trim_both_sides() {
        let input = " test ";
        let (remaining, result) = trim(take_where(4, |c: char| c.is_ascii_alphabetic()))
            .parse(input)
            .unwrap();
        assert_eq!(remaining, "");
        assert_eq!(result, "test");
    }

    #[test]
    fn test_trim_leading() {
        let input = " test";
        let (remaining, result) = trim(take_where(4, |c: char| c.is_ascii_alphabetic()))
            .parse(input)
            .unwrap();
        assert_eq!(remaining, "");
        assert_eq!(result, "test");
    }

    #[test]
    fn test_trim_trailing() {
        let input = "test ";
        let (remaining, result) = trim(take_where(4, |c: char| c.is_ascii_alphabetic()))
            .parse(input)
            .unwrap();
        assert_eq!(remaining, "");
        assert_eq!(result, "test");
    }

    #[test]
    fn test_trim_no_trim() {
        let input = "test";
        let (remaining, result) = trim(take_where(4, |c: char| c.is_ascii_alphabetic()))
            .parse(input)
            .unwrap();
        assert_eq!(remaining, "");
        assert_eq!(result, "test");
    }

    #[test]
    fn test_parenthesized() {
        let input = "(test)";
        let (remaining, result) = parenthesized(take_where(4, |c: char| c.is_ascii_alphabetic()))
            .parse(input)
            .unwrap();
        assert_eq!(remaining, "");
        assert_eq!(result, "test");
    }

    #[test]
    fn test_parenthesized_parse_until_end() {
        let input = "(test)";
        assert!(
            parenthesized::<&str, &str, _, nom::error::Error<&str>>(take_while(|_| true))
                .parse(input)
                .is_err()
        );
    }

    #[test]
    fn test_take_where() {
        let input = "test";
        let (remaining, result) = take_where(4, |c: char| c.is_ascii_alphabetic())(input).unwrap();
        assert_eq!(remaining, "");
        assert_eq!(result, "test");
    }

    #[test]
    fn test_take_where_not_enough() {
        let input = "tes";
        assert!(take_where(4, |c: char| c.is_ascii_alphabetic())(input).is_err());
    }

    #[test]
    fn test_take_where_too_much() {
        let input = "testing";
        assert_eq!(
            take_where(4, |c: char| c.is_ascii_alphabetic()).parse(input),
            Ok(("ing", "test"))
        );
    }

    #[test]
    fn test_take_where_predicate_false() {
        let input = "test";
        assert!(
            take_where(4, |c: char| c.is_ascii_digit())
                .parse(input)
                .is_err()
        );
    }

    #[test]
    fn test_exhausted() {
        let input = "test";
        let (remaining, result) = exhausted(take_where(4, |c: char| c.is_ascii_alphabetic()))
            .parse(input)
            .unwrap();
        assert_eq!(remaining, "");
        assert_eq!(result, "test");
    }

    #[test]
    fn test_exhausted_not_exhausted() {
        let input = "test ";
        assert!(
            exhausted(take_where(4, |c: char| c.is_ascii_alphabetic()))
                .parse(input)
                .is_err()
        );
    }

    #[test]
    fn test_exhausted_tuple() {
        let input = "test";
        let (remaining, result) = exhausted((
            take_where(3, |c: char| c.is_ascii_alphabetic()),
            take_while(|c: char| c.is_ascii_alphabetic()),
        ))
        .parse(input)
        .unwrap();
        assert_eq!(remaining, "");
        assert_eq!(result, ("tes", "t"));
    }
}