lindera_filter/token.rs
1use serde::Serialize;
2
3use lindera_core::word_entry::WordId;
4
5#[derive(Serialize, Clone)]
6pub struct Token {
7 /// Text content of the token.
8 pub text: String,
9
10 /// Starting position of the token in bytes.
11 pub byte_start: usize,
12
13 /// Ending position of the token in bytes.
14 pub byte_end: usize,
15
16 /// Position, expressed in number of tokens.
17 pub position: usize,
18
19 /// The length expressed in terms of number of original tokens.
20 pub position_length: usize,
21
22 /// The ID of the word and a flag to indicate whether the word is registered in the dictionary.
23 pub word_id: WordId,
24
25 /// Detailes about the token.
26 /// It contains metadata for tokens, such as part-of-speech information.
27 pub details: Vec<String>,
28}