jayce

jayce 是一个分词器 🌌

示例

use jayce::{Duo, Tokenizer};
use std::sync::OnceLock;

const SOURCE: &str = "Excalibur = 5000$; // Your own language!";

fn duos() -> &'static Vec<Duo<&'static str>> {
    static DUOS: OnceLock<Vec<Duo<&'static str>>> = OnceLock::new();
    DUOS.get_or_init(|| {
        vec![
            Duo::new("whitespace", r"^[^\S\n]+", false),
            Duo::new("commentLine", r"^//(.*)", false),
            Duo::new("commentBlock", r"^/\*(.|\n)*?\*/", false),
            Duo::new("newline", r"^\n", false),
            Duo::new("price", r"^[0-9]+\$", true),
            Duo::new("semicolon", r"^;", true),
            Duo::new("operator", r"^=", true),
            Duo::new("name", r"^[a-zA-Z_]+", true),
        ]
    })
}

fn main() -> Result<(), Box<dyn std::error::Error>> {
    let mut tokenizer = Tokenizer::new(SOURCE, duos());

    while let Some(token) = tokenizer.consume()? {
        println!("{:?}", token);
    }

    Ok(())
}

结果

Token { kind: "name",      value: "Excalibur", pos: (1, 1)  }
Token { kind: "operator",  value: "=",         pos: (1, 11) }
Token { kind: "price",     value: "5000$",     pos: (1, 13) }
Token { kind: "semicolon", value: ";",         pos: (1, 18) }

信息

Tokenizer::consume 返回 Result Option Token

1. Ok Some 匹配找到
2. Ok None 源文件结束
3. Err 发生错误

Tokenizer::consume_all 返回 Result Vec Token

1. Ok Vec Token 匹配到标记
2. Err 发生错误

性能

初始化耗时约 3 纳秒，对 Yuumi 的分词耗时约 4 毫秒

jayce = { version = "X.X.X", default-features = false }