1
  2
  3
  4
  5
  6
  7
  8
  9
 10
 11
 12
 13
 14
 15
 16
 17
 18
 19
 20
 21
 22
 23
 24
 25
 26
 27
 28
 29
 30
 31
 32
 33
 34
 35
 36
 37
 38
 39
 40
 41
 42
 43
 44
 45
 46
 47
 48
 49
 50
 51
 52
 53
 54
 55
 56
 57
 58
 59
 60
 61
 62
 63
 64
 65
 66
 67
 68
 69
 70
 71
 72
 73
 74
 75
 76
 77
 78
 79
 80
 81
 82
 83
 84
 85
 86
 87
 88
 89
 90
 91
 92
 93
 94
 95
 96
 97
 98
 99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116

use std::iter;
use super::{Regex, Region, OPTION_NONE};

impl Regex {
    /// Returns the capture groups corresponding to the leftmost-first match
    /// in text. Capture group `0` always corresponds to the entire match.
    /// If no match is found, then `None` is returned.
    ///
    /// # Panics
    ///
    /// This method may panic in the case of memory overflow during execution or
    /// other internal errors of Oniguruma engine.
    pub fn captures<'t>(&self, text: &'t str) -> Option<Captures<'t>> {
        let mut region = Region::new();
        self.search_with_region(text, &mut region, OPTION_NONE)
            .unwrap()
            .map(|_| Captures { text: text, region: region })
    }
}

/// Captures represents a group of captured strings for a single match.
///
/// The 0th capture always corresponds to the entire match. Each subsequent
/// index corresponds to the next capture group in the regex. Positions
/// returned from a capture group are always byte indices.
///
/// `'t` is the lifetime of the matched text.
#[derive(Debug)]
pub struct Captures<'t> {
    text: &'t str,
    region: Region
}

impl<'t> Captures<'t> {
    /// Returns the start and end positions of the Nth capture group. Returns
    /// `None` if i is not a valid capture group or if the capture group did
    /// not match anything. The positions returned are always byte indices with
    /// respect to the original string matched.
    pub fn pos(&self, pos: usize) -> Option<(usize, usize)> {
        self.region.pos(pos)
    }

    /// Returns the matched string for the capture group `i`. If `i` isn't
    /// a valid capture group or didn't match anything, then `None` is returned.
    pub fn at(&self, pos: usize) -> Option<&'t str> {
        self.pos(pos).map(|(beg, end)| &self.text[beg..end])
    }

    /// Returns the number of captured groups.
    pub fn len(&self) -> usize {
        self.region.len()
    }

    /// Returns true if and only if there are no captured groups.
    pub fn is_empty(&self) -> bool {
        self.len() == 0
    }

    /// Creates an iterator of all the capture groups in order of appearance in
    // the regular expression.
    pub fn iter(&'t self) -> SubCaptures<'t> {
        SubCaptures { idx: 0, caps: self }
    }

    /// Creates an iterator of all the capture group positions in order of
    /// appearance in the regular expression. Positions are byte indices in
    /// terms of the original string matched.
    pub fn iter_pos(&'t self) -> SubCapturesPos<'t> {
        SubCapturesPos { idx: 0, caps: self }
    }
}

/// An iterator over capture groups for a particular match of a regular
/// expression.
///
///`'t` is the lifetime of the matched text.
pub struct SubCaptures<'t> {
    idx: usize,
    caps: &'t Captures<'t>
}

impl<'t> iter::Iterator for SubCaptures<'t> {
    type Item = Option<&'t str>;

    fn next(&mut self) -> Option<Option<&'t str>> {
        if self.idx < self.caps.len() {
            self.idx += 1;
            Some(self.caps.at(self.idx - 1))
        } else {
            None
        }
    }
}

/// An iterator over capture group positions for a particular match of
/// a regular expression.
///
/// Positions are byte indices in terms of the original
/// string matched. `'t` is the lifetime of the matched text.
pub struct SubCapturesPos<'t> {
    idx: usize,
    caps: &'t Captures<'t>
}

impl<'t> iter::Iterator for SubCapturesPos<'t> {
    type Item = Option<(usize, usize)>;

    fn next(&mut self) -> Option<Option<(usize, usize)>> {
        if self.idx < self.caps.len() {
            self.idx += 1;
            Some(self.caps.pos(self.idx - 1))
        } else {
            None
        }
    }
}