Installation

``bash
npm install openspg-schema-antlr4
`

> It will be pnpm/yarn add openspg-schema-antlr4 if you use pnpm or yarn.

Usage

$3

* parse(code, [options]): parse() parses the provided code as an entire Schema source unit.
* options:
* tolerant: boolean, default is false. If true, the parser will try to parse as much as possible, even if the input is invalid, and never throw an error.
* selector: function, default is (p) => p.sourceUnit(). If provided, the parser will only return the nodes that match the selector. It will be useful when you want to
parse a specific node.
* output: SyntaxNode, the root node of the AST.

`js
// parse.mjs
import {parse} from 'openspg-schema-antlr4';

const code = // Sample
namespace BaiKe

Person(人物): EntityType
properties:
desc(描述): Text
index: TextAndVector
semanticType(语义类型): SemanticConcept
synonyms(同义词): Person
officialName(标准名): Person

Transport(运输): EntityType
properties:
desc(描述): Text
index: TextAndVector
semanticType(语义类型): SemanticConcept
synonyms(同义词): Transport
officialName(标准名): Transport

Works(作品): EntityType
properties:
desc(描述): Text
index: TextAndVector
semanticType(语义类型): SemanticConcept
synonyms(同义词): Works
officialName(标准名): Works

Event(事件): EventType
properties:
subject(主体): Person
participants(参与者): Person
constraint: MultiValue
time(时间): Date
location(地点): GeographicLocation
abstract(摘要): Text
index: TextAndVector
type(事件类型): Text
index: Text
;

const ast = parse(code, {tolerant: true, selector: (p) => p.sourceUnit()});

`

$3

* tokenizer(code, [options]): tokenizer() parses the provided code as tokens.
* options:
* tolerant: boolean, default is false.
* output: SyntaxToken[].

`js
// tokenizer.mjs
import {tokenizer} from 'openspg-schema-antlr4';

const tokens = tokenizer(code, {tolerant: true});

`

$3

We can use it alongside the parser to traverse nodes.

`js
import {parse, visit, serialize} from 'openspg-schema-antlr4';

const ast = parse(code);

// Use visit to traverse ast by enter/exit node type.
visit(ast, {
enter: ({node, parent}) => {
console.log(node.type, parent?.type); // print node type
},
exit: () => {
}, // will call when exit node
Identifier: ({node: identifierNode}) => {
console.log(identifierNode.name); // print identifier name
},
exitContractDefinition: ({node: contractDefinitionNode}) => {
// will call when exit ContractDefinition node
}
});

// Use serialize to modify ast.
const newAST = serialize(ast, ({node}) => {
// do something
if (node.type === 'Identifier') {
return node.name;
}
return node;
})
`

`js
// traverse.mjs
import {parse, traverse} from 'openspg-schema-antlr4';

const ast = parse(code);

const newAST = traverse(ast, (path) => {
// path.path => SourceUnit.Namespace ...
// path.node => current node
// path.parentPath => parent node path
// path.depth => current node depth
// path.stop(); => stop traverse
// path.rewrite({...}); => rewrite current node
// path.matches({ type: 'xxx' }); => check if current node matches the given filter
// return () => {}; => will call when exit node
});
`

$3

> Not recommended, but you can use it if you want.

`ts
import {SchemaLexer, SchemaParser, CharStream, CommonTokenStream} from 'openspg-schema-antlr4';

const code = ...; // code here

const input = CharStream.fromString(code);
const lexer = new SolidityLexer(input);
const tokens = new CommonTokenStream(lexer);
const parser = new SolidityParser(tokens);

const parseTree = parser.sourceUnit();

// do something with parseTree
``

License

Apache