Writing custom parsers

If you want to use your own parser and provide additional capabilities for your rules, you can specify your own custom parser. The parser should expose, as default, a single function that take the source code as the first argument. The function should simply return an AST.

⚠️The first node of the tree should have the property type with the value root.

You can find an LintHTML parser project here.

{
    "parser": "./path/to/my-parser.js"
}

Parser CJS format

// my-parser.js
const myParser = require("customParser");

module.exports = function(code) {
    return myParser(code)
};

Parser CJS format

// my-parser.js
import myParser from "customParser";

export default function(code) {
    return myParser(code)
};

The AST specification

Please refer to the API documentation to know how to construct nodes matching the AST specification.

interface Node {
 type: 'tag' | 'text' | 'directive' | 'comment' | 'root' | 'script' | 'style' | 'cdata' | 'doctype';
 
 name: string;
 data?: string // only 'text', 'directive' and 'comment' nodes
 attributes?: NodeAttribute[]; // only 'tag' nodes

 parent?: Node;
 previousSibling: Node;
 nextSibling?: Node;

 startIndex: number; 
 endIndex: number; 
 openIndex: number;
 closeIndex: number;
 
 children: Node[];
 
 loc: Position;
 open: CharValue;
 close: CharValue; 
}

interface CharValue {
 chars: string; // tag name, attribute name or attribute value without `<, >, ', ", \s, \t` chars
 raw?: string; // Same as `chars` but with `<, >, ', ",` chars
 loc: Position;
}

interface NodeAttribute {
 type: string;
 name: CharValue;
 value: CharValue;
 equal: CharValue;
 loc: Position;
}

interface Position {
 start: Range;
 end: Range;
}

interface Range {
 line: number;
 column: number;
}