automaton 0.0.2

Parse regular languages
Documentation 
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
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143

use {Token};
use std::ops;
use std::collections::HashSet;
use std::hash::{Hash, Hasher};

#[derive(Clone, Debug)]
pub struct Alphabet {
    tokens: HashSet<Letter>,
}

impl Alphabet {
    pub fn empty() -> Alphabet {
        Alphabet {
            tokens: set![],
        }
    }

    pub fn single(letter: Letter) -> Alphabet {
        Alphabet {
            tokens: set![letter],
        }
    }

    pub fn extend(&mut self, other: &Alphabet) {
        for token in &other.tokens {
            self.insert(token.clone());
        }
    }

    pub fn refine(&mut self) {
        let mut tokens: Vec<Option<Letter>> = self.tokens.iter()
            .cloned()
            .map(Some)
            .collect();

        let mut i = 0;

        while i < tokens.len() {
            if tokens[i].is_none() {
                i += 1;
                continue;
            }

            for j in i+1..tokens.len() {
                if tokens[i].is_none() || tokens[j].is_none() {
                    continue;
                }

                let new = Letter::disjoint(
                    tokens[i].as_ref().expect("token[i] is none"),
                    tokens[j].as_ref().expect("token[j] is none"));

                if let Some(new) = new {
                    tokens[i].take();
                    tokens[j].take();

                    tokens.extend(new.into_iter().map(Some));
                }
            }

            i += 1;
        }

        self.tokens.clear();
        self.tokens.extend(tokens.into_iter().filter_map(|t| t));
    }
}

impl ops::Deref for Alphabet {
    type Target = HashSet<Letter>;

    fn deref(&self) -> &HashSet<Letter> {
        &self.tokens
    }
}

impl ops::DerefMut for Alphabet {
    fn deref_mut(&mut self) -> &mut HashSet<Letter> {
        &mut self.tokens
    }
}

#[derive(Clone, Debug, Eq, PartialEq)]
pub struct Letter(pub ops::Range<u32>);

impl Letter {
    pub fn disjoint(a: &Letter, b: &Letter) -> Option<Vec<Letter>> {
        // If the tokens don't overlap, nothing more to do
        if a.0.end <= b.0.start || b.0.end <= a.0.start {
            return None;
        }

        let mut points = [
            a.0.start,
            a.0.end,
            b.0.start,
            b.0.end,
        ];

        points.sort();

        fn push(dst: &mut Vec<Letter>, range: ops::Range<u32>) {
            if range.end > range.start {
                dst.push(Letter(range));
            }
        }

        let mut ret = Vec::with_capacity(3);

        push(&mut ret, ops::Range { start: points[0], end: points[1] });
        push(&mut ret, ops::Range { start: points[1], end: points[2] });
        push(&mut ret, ops::Range { start: points[2], end: points[3] });

        Some(ret)
    }

    pub fn contains(&self, other: &Letter) -> bool {
        self.0.start <= other.0.start && self.0.end >= other.0.end
    }

    pub fn contains_val(&self, val: u32) -> bool {
        self.start <= val && self.end > val
    }

    pub fn to_token<T: Token>(&self) -> T {
        <T as Token>::from_range(&self.0)
    }
}

impl ops::Deref for Letter {
    type Target = ops::Range<u32>;

    fn deref(&self) -> &ops::Range<u32> {
        &self.0
    }
}

impl Hash for Letter {
    fn hash<H>(&self, state: &mut H) where H: Hasher {
        self.start.hash(state);
        self.end.hash(state);
    }
}