charabia 0.9.9

A simple library to detect the language, tokenize the text and normalize the tokens
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

use std::collections::HashSet;
use std::sync::LazyLock;

use super::CharNormalizer;
use crate::detection::Script;
use crate::normalizer::CharOrStr;
use crate::Token;

static NONSPACING_MARKS: LazyLock<HashSet<u32>> = LazyLock::new(|| {
    let bytes = include_bytes!("../../dictionaries/bin/nonspacing_mark/marks.bin");

    HashSet::from_iter(
        bytes.chunks_exact(4).map(|chunk| u32::from_ne_bytes(chunk.try_into().unwrap())),
    )
});

/// A global [`Normalizer`] removing nonspacing marks.
///
/// This normalizer uses built-in `HashSet` internally to check over the marks set
pub struct NonspacingMarkNormalizer;

impl CharNormalizer for NonspacingMarkNormalizer {
    fn normalize_char(&self, c: char) -> Option<CharOrStr> {
        (!is_nonspacing_mark(c)).then(|| c.into())
    }

    fn should_normalize(&self, token: &Token) -> bool {
        matches!(
            token.script,
            Script::Hebrew | Script::Thai | Script::Arabic | Script::Latin | Script::Greek
        ) && token.lemma().chars().any(is_nonspacing_mark)
    }
}

/// Returns true if the character is a nonspacing mark
fn is_nonspacing_mark(c: char) -> bool {
    NONSPACING_MARKS.contains(&(c as u32))
}

#[cfg(test)]
mod test {
    use std::borrow::Cow::Owned;

    use crate::normalizer::test::test_normalizer;
    use crate::normalizer::{Normalizer, NormalizerOption};
    use crate::token::TokenKind;

    // base tokens to normalize.
    fn tokens() -> Vec<Token<'static>> {
        vec![
            Token {
                lemma: Owned("ง่าย".to_string()),
                char_end: "ง่าย".chars().count(),
                byte_end: "ง่าย".len(),
                script: Script::Thai,
                ..Default::default()
            },
            Token {
                lemma: Owned("أَب".to_string()),
                char_end: "أَب".chars().count(),
                byte_end: "أَب".len(),
                script: Script::Arabic,
                ..Default::default()
            },
            Token {
                lemma: Owned("כָּבוֹד".to_string()),
                char_end: "כָּבוֹד".chars().count(),
                byte_end: "כָּבוֹד".len(),
                script: Script::Hebrew,
                ..Default::default()
            },
        ]
    }

    // expected result of the current Normalizer.
    fn normalizer_result() -> Vec<Token<'static>> {
        vec![
            Token {
                lemma: Owned("งาย".to_string()),
                char_end: 4,
                byte_end: 12,
                char_map: Some(vec![(3, 3), (3, 0), (3, 3), (3, 3)]),
                script: Script::Thai,
                ..Default::default()
            },
            Token {
                lemma: Owned("أب".to_string()),
                char_end: "أَب".chars().count(),
                byte_end: "أَب".len(),
                char_map: Some(vec![(2, 2), (2, 0), (2, 2)]),
                script: Script::Arabic,
                ..Default::default()
            },
            Token {
                lemma: Owned("כבוד".to_string()),
                char_end: "כָּבוֹד".chars().count(),
                byte_end: "כָּבוֹד".len(),
                script: Script::Hebrew,
                char_map: Some(vec![(2, 2), (2, 0), (2, 0), (2, 2), (2, 2), (2, 0), (2, 2)]),
                ..Default::default()
            },
        ]
    }

    // expected result of the complete Normalizer pipeline.
    fn normalized_tokens() -> Vec<Token<'static>> {
        vec![
            Token {
                lemma: Owned("งาย".to_string()),
                char_end: 4,
                byte_end: 12,
                char_map: Some(vec![(3, 3), (3, 0), (3, 3), (3, 3)]),
                script: Script::Thai,
                kind: TokenKind::Word,
                ..Default::default()
            },
            Token {
                lemma: Owned("اب".to_string()),
                char_end: "أَب".chars().count(),
                byte_end: "أَب".len(),
                char_map: Some(vec![(2, 2), (2, 0), (2, 2)]),
                script: Script::Arabic,
                kind: TokenKind::Word,
                ..Default::default()
            },
            Token {
                lemma: Owned("כבוד".to_string()),
                char_end: "כָּבוֹד".chars().count(),
                byte_end: "כָּבוֹד".len(),
                script: Script::Hebrew,
                char_map: Some(vec![(2, 2), (2, 0), (2, 0), (2, 2), (2, 2), (2, 0), (2, 2)]),
                kind: TokenKind::Word,
                ..Default::default()
            },
        ]
    }

    test_normalizer!(NonspacingMarkNormalizer, tokens(), normalizer_result(), normalized_tokens());
}