token.rs

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
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173

use crate::error::Error;


#[derive(Debug, Clone, PartialEq)]
pub enum Token {
    Character(char),
    Command(String),
    Text(String),
    LeftBrace,
    RightBrace,
    Caret,
    Underscore,
    Ampersand,
    RowBreak,
}


#[derive(Debug, Clone, PartialEq)]
pub struct Located {
    pub token: Token,
    pub position: usize,
}


pub fn tokenize(source: &str) -> Result<Vec<Located>, Error> {
    let characters: Vec<char> = source.chars().collect();
    let mut tokens = Vec::new();
    let mut index = 0;
    while index < characters.len() {
        let start = index;
        let character = characters[index];
        index += 1;
        let token = match character {
            c if c.is_whitespace() => continue,
            '{' => Token::LeftBrace,
            '}' => Token::RightBrace,
            '^' => Token::Caret,
            '_' => Token::Underscore,
            '&' => Token::Ampersand,
            '\\' => {
                let (token, next) = read_command(&characters, index, start)?;
                index = next;
                token
            }
            c => Token::Character(c),
        };
        tokens.push(Located {token, position: start});
    }
    Ok(tokens)
}


fn read_command(characters: &[char], index: usize, start: usize) -> Result<(Token, usize), Error> {
    match characters.get(index) {
        None => Err(Error::new("no command name after the backslash", start)),
        Some('\\') => Ok((Token::RowBreak, index + 1)),
        Some(first) if first.is_ascii_alphabetic() => {
            let mut name = String::new();
            let mut next = index;
            while let Some(character) = characters.get(next) {
                if !character.is_ascii_alphabetic() {
                    break;
                }
                name.push(*character);
                next += 1;
            }
            if name == "text" {
                let (body, after) = read_text(characters, next, start)?;
                return Ok((Token::Text(body), after));
            }
            Ok((Token::Command(name), next))
        }
        Some(character) => Ok((Token::Command(character.to_string()), index + 1)),
    }
}


fn read_text(characters: &[char], index: usize, start: usize) -> Result<(String, usize), Error> {
    let mut next = index;
    while matches!(characters.get(next), Some(character) if character.is_whitespace()) {
        next += 1;
    }
    if characters.get(next) != Some(&'{') {
        return Err(Error::new("\\text must be followed by {text}", start));
    }
    next += 1;
    let mut body = String::new();
    let mut depth = 1;
    while let Some(&character) = characters.get(next) {
        next += 1;
        match character {
            '{' => {
                depth += 1;
                body.push(character);
            }
            '}' => {
                depth -= 1;
                if depth == 0 {
                    return Ok((body, next));
                }
                body.push(character);
            }
            _ => body.push(character),
        }
    }
    Err(Error::new("\\text has an unclosed brace", start))
}


#[cfg(test)]
mod tests {
    use super::*;

    fn kinds(source: &str) -> Vec<Token> {
        tokenize(source).unwrap().into_iter().map(|located| located.token).collect()
    }

    #[test]
    fn reads_commands_and_characters() {
        assert_eq!(
            kinds(r"\frac{a}{2}"),
            vec![
                Token::Command("frac".to_string()),
                Token::LeftBrace,
                Token::Character('a'),
                Token::RightBrace,
                Token::LeftBrace,
                Token::Character('2'),
                Token::RightBrace,
            ]
        );
    }

    #[test]
    fn reads_row_break_and_ampersand() {
        assert_eq!(kinds(r"a & b \\ c"), vec![
            Token::Character('a'),
            Token::Ampersand,
            Token::Character('b'),
            Token::RowBreak,
            Token::Character('c'),
        ]);
    }

    #[test]
    fn keeps_spaces_inside_text() {
        assert_eq!(kinds(r"\text{a b}"), vec![Token::Text("a b".to_string())]);
        assert_eq!(kinds(r"\text {x}"), vec![Token::Text("x".to_string())]);
    }

    #[test]
    fn single_character_commands() {
        assert_eq!(kinds(r"\, \; \{"), vec![
            Token::Command(",".to_string()),
            Token::Command(";".to_string()),
            Token::Command("{".to_string()),
        ]);
    }

    #[test]
    fn records_positions() {
        let tokens = tokenize(r"x + \alpha").unwrap();
        assert_eq!(tokens[0].position, 0);
        assert_eq!(tokens[1].position, 2);
        assert_eq!(tokens[2].position, 4);
    }

    #[test]
    fn reports_a_lonely_backslash() {
        assert!(tokenize("a\\").is_err());
        assert!(tokenize(r"\text x").is_err());
    }
}