charabia 0.10.0

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
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157

use std::borrow::Cow;
use std::sync::LazyLock;

use lindera::dictionary::{load_embedded_dictionary, DictionaryKind};
use lindera::mode::{Mode, Penalty};
use lindera::segmenter::Segmenter as LinderaSegmenter;

use crate::segmenter::Segmenter;

/// Japanese specialized [`Segmenter`].
///
/// This Segmenter uses lindera internally to segment the provided text.
pub struct JapaneseSegmenter;

static LINDERA: LazyLock<LinderaSegmenter> = LazyLock::new(|| {
    #[cfg(all(feature = "japanese-segmentation-ipadic", feature = "japanese-segmentation-unidic"))]
    compile_error!("Feature japanese-segmentation-ipadic and japanese-segmentation-unidic are mutually exclusive and cannot be enabled together");

    #[cfg(feature = "japanese-segmentation-ipadic")]
    let dictionary_kind = DictionaryKind::IPADIC;
    #[cfg(feature = "japanese-segmentation-unidic")]
    let dictionary_kind = DictionaryKind::UniDic;

    let dictionary = load_embedded_dictionary(dictionary_kind).unwrap();
    LinderaSegmenter::new(Mode::Decompose(Penalty::default()), dictionary, None)
});

impl Segmenter for JapaneseSegmenter {
    fn segment_str<'o>(&self, to_segment: &'o str) -> Box<dyn Iterator<Item = &'o str> + 'o> {
        let tokens = LINDERA.segment(Cow::Borrowed(to_segment)).unwrap();

        let result: Vec<&'o str> = tokens
            .into_iter()
            .map(|token| {
                let start = token.byte_start;
                let end = token.byte_end;
                &to_segment[start..end]
            })
            .collect();

        Box::new(result.into_iter())
    }
}

#[cfg(test)]
mod test {
    use crate::segmenter::test::test_segmenter;

    const TEXT: &str = "関西国際空港限定トートバッグ すもももももももものうち 123 456";

    const SEGMENTED: &[&str] = if cfg!(feature = "japanese-segmentation-ipadic") {
        &[
            "関西",
            "国際",
            "空港",
            "限定",
            "トートバッグ",
            " ",
            "すもも",
            "も",
            "もも",
            "も",
            "もも",
            "の",
            "うち",
            " ",
            "123",
            " ",
            "456",
        ]
    } else if cfg!(feature = "japanese-segmentation-unidic") {
        &[
            "関西",
            "国際",
            "空港",
            "限定",
            "トート",
            "バッグ",
            " ",
            "すもも",
            "も",
            "もも",
            "も",
            "もも",
            "の",
            "うち",
            " ",
            "123",
            " ",
            "456",
        ]
    } else {
        &[]
    };

    const TOKENIZED: &[&str] = if cfg!(feature = "japanese-segmentation-ipadic") {
        &[
            "関西",
            "国際",
            "空港",
            "限定",
            // Use "とうとばっぐ" instead when feature "japanese-transliteration" is enabled or become default
            #[cfg(feature = "japanese-transliteration")]
            "とうとは\u{3099}っく\u{3099}",
            #[cfg(not(feature = "japanese-transliteration"))]
            "トートハ\u{3099}ック\u{3099}",
            " ",
            "すもも",
            "も",
            "もも",
            "も",
            "もも",
            "の",
            "うち",
            " ",
            "123",
            " ",
            "456",
        ]
    } else if cfg!(feature = "japanese-segmentation-unidic") {
        &[
            "関西",
            "国際",
            "空港",
            "限定",
            // Use "とうとばっぐ" instead when feature "japanese-transliteration" is enabled or become default
            #[cfg(feature = "japanese-transliteration")]
            "とうと",
            #[cfg(not(feature = "japanese-transliteration"))]
            "トート",
            #[cfg(feature = "japanese-transliteration")]
            "は\u{3099}っく\u{3099}",
            #[cfg(not(feature = "japanese-transliteration"))]
            "ハ\u{3099}ック\u{3099}",
            " ",
            "すもも",
            "も",
            "もも",
            "も",
            "もも",
            "の",
            "うち",
            " ",
            "123",
            " ",
            "456",
        ]
    } else {
        &[]
    };

    #[cfg(all(feature = "japanese-segmentation-ipadic", feature = "japanese-segmentation-unidic"))]
    compile_error!("Feature japanese-segmentation-ipadic and japanese-segmentation-unidic are mutually exclusive and cannot be enabled together");

    // Macro that run several tests on the Segmenter.
    test_segmenter!(JapaneseSegmenter, TEXT, SEGMENTED, TOKENIZED, Script::Cj, Language::Jpn);
}