pattern_lexer/

pattern.rs

use regex::Regex;

/// Returned by [Pattern] on match.
#[derive(Debug, Clone, PartialEq)]
pub struct Match<'a> {
    /// The string that was searched in
    pub base: &'a str,
    /// Start of the match
    pub start: usize,
    /// End of the match
    pub end: usize,
}

impl<'a> Match<'a> {
    /// Create a match from a base `&str` and start..end range.
    ///
    /// # Panics
    /// When ```rust start >= end``` or ```rust base.len() < end```.
    ///
    /// # Example
    /// ```should_panic
    /// # use pattern_lexer::pattern::Match;
    /// #
    /// let mat = Match::new("don't go to far...", 0, 100000);
    /// ```
    pub fn new(base: &'a str, start: usize, end: usize) -> Self {
        assert!(start < end);
        assert!(base.len() >= end);
        Self { base, start, end }
    }

    /// Convert to to `&str`.
    ///
    /// # Example
    /// ```rust
    /// # use pattern_lexer::pattern::Match;
    /// #
    /// let mat = Match::new("it's here not here", 5, 9);
    ///
    /// assert_eq!(mat.as_str(), "here");
    /// ```
    pub fn as_str(&self) -> &'a str {
        &self.base[self.start..self.end]
    }
}

/// A string Pattern.
///
/// The type implementing it can be used as a pattern for `&str`,
/// by default it is implemented for the following types:
///
/// | Pattern type              | Match condition                         |
/// |---------------------------|-----------------------------------------|
/// | ```char```                | is contained in string                  |
/// | ```&str```                | is substring                            |
/// | ```String```              | is substring                            |
/// | ```&[char]```             | any `char` is contained in string         |
/// | ```&[&str]```             | any `&str` is substring                   |
/// | ```F: Fn(&str) -> bool``` | `F` returns `true` for substring |
/// | ```Regex```               | `Regex` match substring        |
pub trait Pattern<'a> {
    /// Find all occurences of the pattern in the given `&str`.
    ///
    /// # Examples
    /// ```rust
    /// # use pattern_lexer::pattern::{Match, Pattern};
    /// #
    /// assert!("ab".find_in("cd").is_empty());
    /// assert_eq!("ab".find_in("cabd"), vec![Match::new("cabd", 1, 3)]);
    /// ```
    fn find_in(&self, value: &'a str) -> Vec<Match<'a>>;

    /// Find all occurences of the pattern in the given `&str` that are prefixes.
    ///
    /// # Examples
    /// ```rust
    /// # use pattern_lexer::pattern::{Match, Pattern};
    /// #
    /// assert!("ab".find_prefix_in("cdab").is_empty());
    /// assert_eq!("ab".find_prefix_in("abcd"), vec![Match::new("abcd", 0, 2)]);
    /// ```
    fn find_prefix_in(&self, value: &'a str) -> Vec<Match<'a>> {
        self.find_in(value)
            .into_iter()
            .filter(|mat| mat.start == 0)
            .collect()
    }

    /// Find all occurences of the pattern in the given `&str` that are suffixes.
    ///
    /// # Examples
    /// ```rust
    /// # use pattern_lexer::pattern::{Match, Pattern};
    /// #
    /// assert!("ab".find_suffix_in("abcd").is_empty());
    /// assert_eq!("ab".find_suffix_in("cdab"), vec![Match::new("cdab", 2, 4)]);
    /// ```
    fn find_suffix_in(&self, value: &'a str) -> Vec<Match<'a>> {
        let len = value.len();
        self.find_in(value)
            .into_iter()
            .filter(|mat| mat.end == len)
            .collect()
    }

    /// Find one occurrence of the pattern in the given `&str`.
    ///
    /// # Examples
    /// ```rust
    /// # use pattern_lexer::pattern::{Match, Pattern};
    /// #
    /// assert!("ab".find_one_in("cd").is_none());
    /// assert_eq!("ab".find_one_in("cdab"), Some(Match::new("cdab", 2, 4)));
    /// ```
    fn find_one_in(&self, value: &'a str) -> Option<Match<'a>> {
        self.find_in(value).into_iter().next()
    }

    /// Find one occurrence of the pattern in the given `&str` that is prefix.
    ///
    /// # Examples
    /// ```rust
    /// # use pattern_lexer::pattern::{Match, Pattern};
    /// #
    /// assert!("ab".find_one_prefix_in("cdab").is_none());
    /// assert_eq!("ab".find_one_prefix_in("abcd"), Some(Match::new("abcd", 0, 2)));
    /// ```
    fn find_one_prefix_in(&self, value: &'a str) -> Option<Match<'a>> {
        self.find_prefix_in(value).into_iter().next()
    }

    /// Find one occurrence of the pattern in the given `&str` that is suffix.
    ///
    /// # Examples
    /// ```rust
    /// # use pattern_lexer::pattern::{Match, Pattern};
    /// #
    /// assert!("ab".find_one_suffix_in("abcd").is_none());
    /// assert_eq!("ab".find_one_suffix_in("cdab"), Some(Match::new("cdab", 2, 4)));
    /// ```
    fn find_one_suffix_in(&self, value: &'a str) -> Option<Match<'a>> {
        self.find_suffix_in(value).into_iter().next()
    }
}

impl<'a> Pattern<'a> for char {
    fn find_in(&self, value: &'a str) -> Vec<Match<'a>> {
        value
            .match_indices(&self.to_string())
            .map(|(i, mat)| Match::new(value, i, i + mat.len()))
            .collect()
    }
}

impl<'a> Pattern<'a> for [char] {
    fn find_in(&self, value: &'a str) -> Vec<Match<'a>> {
        self.iter().flat_map(|ch| ch.find_in(value)).collect()
    }
}

impl<'a, const N: usize> Pattern<'a> for [char; N] {
    fn find_in(&self, value: &'a str) -> Vec<Match<'a>> {
        self.as_slice().find_in(value)
    }
}

impl<'a, const N: usize> Pattern<'a> for &[char; N] {
    fn find_in(&self, value: &'a str) -> Vec<Match<'a>> {
        self.as_slice().find_in(value)
    }
}

impl<'a> Pattern<'a> for String {
    fn find_in(&self, value: &'a str) -> Vec<Match<'a>> {
        value
            .match_indices(self)
            .map(|(i, mat)| Match::new(value, i, i + mat.len()))
            .collect()
    }
}

impl<'a> Pattern<'a> for &str {
    fn find_in(&self, value: &'a str) -> Vec<Match<'a>> {
        self.to_string().find_in(value)
    }
}

impl<'a> Pattern<'a> for [&str] {
    fn find_in(&self, value: &'a str) -> Vec<Match<'a>> {
        self.iter().flat_map(|ch| ch.find_in(value)).collect()
    }
}

impl<'a, const N: usize> Pattern<'a> for [&str; N] {
    fn find_in(&self, value: &'a str) -> Vec<Match<'a>> {
        self.as_slice().find_in(value)
    }
}

impl<'a, const N: usize> Pattern<'a> for &[&str; N] {
    fn find_in(&self, value: &'a str) -> Vec<Match<'a>> {
        self.as_slice().find_in(value)
    }
}

impl<'a: 'b, 'b, F> Pattern<'a> for F
where
    F: Fn(&'b str) -> bool,
{
    fn find_in(&self, value: &'a str) -> Vec<Match<'a>> {
        let mut matches = Vec::new();
        let mut cur_1 = 0;
        // The goal is to check from left to right and to take the largest match
        while cur_1 < value.len() {
            let mut cur_2 = value.len();
            while cur_2 > cur_1 {
                let sub = &value[cur_1..cur_2];
                if (self)(sub) {
                    matches.push(Match::new(value, cur_1, cur_2));
                    cur_1 = cur_2;
                }
                cur_2 -= 1
            }
            cur_1 += 1;
        }
        matches
    }
}

impl<'a> Pattern<'a> for Regex {
    fn find_in(&self, value: &'a str) -> Vec<Match<'a>> {
        self.find_iter(value)
            .map(|mat| Match::new(value, mat.start(), mat.end()))
            .collect()
    }
}