详解Rust调用tree-sitter支持自定义语言解析

mkdir my_language
cd my_language
touch grammar.js

module.exports = grammar({
    name: 'my_language',
    rules: {
        // 规则定义
        program: $ => repeat($.statement),
        statement: $ => choice(
            $.expression,
            $.assignment
        ),
        expression: $ => /[a-zA-Z_][a-zA-Z0-9_]*/,
        assignment: $ => seq(
            $.expression,
            '=',
            $.expression
        ),
    }
});

npm install -g tree-sitter-cli

tree-sitter generate

cargo new tree_sitter_my_language
cd tree_sitter_my_language

[dependencies]
tree-sitter = "0.23"
[build-dependencies]
cc = "1.0"

extern crate cc;
fn main() {
    cc::Build::new()
        .include("my_language/src") // 指向自定义语言的 src 目录
        .file("my_language/src/parser.c")
        .compile("tree-sitter-my_language");
    println!("cargo:rerun-if-changed=my_language/src/parser.c");
}

use tree_sitter::{Parser, Language};
// 引入自定义语言
extern "C" { fn tree_sitter_my_language() -> Language; }
fn main() {
    // 初始化解析器
    let mut parser = Parser::new();
    // 设置自定义语言
    let language = unsafe { tree_sitter_my_language() };
    parser.set_language(&language).expect("Error loading custom language grammar");
    // 要解析的自定义语言代码
    let source_code = r#"
    x = 10
    y = 20
    z = x + y
    "#;
    // 解析源代码
    let tree = parser.parse(source_code, None).unwrap();
    // 获取语法树的根节点
    let root_node = tree.root_node();
    // 输出解析结果
    println!("Parsed custom language code:\n{:?}", root_node);
}

tree_sitter_my_language/
├── Cargo.toml
├── build.rs
├── my_language/        # 自定义语言目录
│   ├── grammar.js
│   ├── src/
│   │   ├── parser.c
│   │   └── ... (其他生成的文件)
└── src/
    └── main.rs

cargo build
cargo run