Job Markdown Parser and Tokenizer

A simple parser for a job markdown language in TypeScript for Node.js 24+.

Installation

npm install jobmd

Usage

Basic Usage

import { parse } from './src/parser.ts';

const text = `- MARKETDOPRODUCT-0_1 stream | brand | Project specification | Project initiation #brand #additional_tags @productManager (2026-03-13_20d) 〈mdoValueStream: stream followers: @seniorProjectManager, @streamLead, @sourcing, @projectManager〉
  - MARKETDOPRODUCT-0_1_1 stream | brand | Project specification | Presourcing #brand #additional_tags @sourcing (2026-03-13_6d) 〈mdoValueStream: stream followers: @projectManager, @streamLead Parent: MARKETDOPRODUCT-0_1〉`;

const ast = parse(text);

console.log(ast.lines[0].text); // "stream | brand | Project specification | Project initiation"
console.log(ast.lines[0].queue); // "MARKETDOPRODUCT-0_1"
console.log(ast.lines[0].tags); // ["#brand", "#additional_tags"]
console.log(ast.lines[0].user); // "@productManager"
console.log(ast.lines[0].dateRange); // "(2026-03-13_20d)"
console.log(ast.lines[0].children.length); // 1

Using the Tokenizer

import { Tokenizer } from './src/tokenizer.ts';

const text = '- MARKETDOPRODUCT-0_1 stream #brand @productManager';
const tokenizer = new Tokenizer(text);

let token = tokenizer.nextToken();
while (token.type !== 'EOF') {
  console.log(`${token.type}: ${token.value}`);
  console.log(`  Position: line ${token.position.line}, column ${token.position.column}`);
  console.log(`  Bytes: ${token.position.startByte}-${token.position.endByte}`);
  token = tokenizer.nextToken();
}

Working with AST

import { parse } from './src/parser.ts';
import type { LineNode } from './src/ast.ts';

const ast = parse(text);

// Recursive AST traversal
function traverse(node: LineNode, depth: number = 0) {
  const indent = '  '.repeat(depth);
  console.log(`${indent}${node.point || ''} ${node.queue || ''} ${node.text}`);

  if (node.tags.length > 0) {
    console.log(`${indent}  Tags: ${node.tags.join(', ')}`);
  }

  if (node.user) {
    console.log(`${indent}  User: ${node.user}`);
  }

  if (node.dateRange) {
    console.log(`${indent}  Date Range: ${node.dateRange}`);
  }

  if (node.props.length > 0) {
    console.log(`${indent}  Props:`);
    for (const prop of node.props) {
      console.log(`${indent}    ${prop.key}: ${prop.values.join(', ')}`);
    }
  }

  if (node.comment) {
    console.log(`${indent}  Comment: ${node.comment}`);
  }

  for (const child of node.children) {
    traverse(child, depth + 1);
  }
}

for (const line of ast.lines) {
  traverse(line);
}

Language Specification

Tokens

• Queue or Ticket: [A-Z]+-(0|(0_)?[0-9]+)

  • Examples: MARKETDOPRODUCT-0_1, FTECH-123, EDUCATION-41472

• Tag: #[a-z]+

  • Examples: #brand, #additional_tags

• User: @[a-z][a-z0-9\-]+

  • Examples: @productManager, @senior-project-manager

• DateRange: \(<date>_(<date>|\d+[dw])\)

  • Examples: (2026-03-13_20d), (2026-03-13_2026-04-02)

• Date: ([0-9]{4}-[0-9]{2}-[0-9]{2}|[0-9]{2}.[0-9]{2}.[0-9]{4})

  • Examples: 2026-03-13, 13.03.2026

• Props: 〈 and 〉

  • key: [a-zA-Z\-]+
  • separator: :
  • value: [^\s]+|"[^"]+" or , or
  • multiple values separated by ,

• Point: - or *

• Indent: Significant, indicates nesting level

• Empty Line: Resets nesting level

• Comment: -- or // to end of line

Line Structure

(<Point>)? (<Queue>)? <Text> (<Tag> )* (<User>)? (<DateRange>)? (〈(Props)*〉)? (<Comment> <Text>)?

Testing

# Run all tests
npm test

# Run tests in watch mode
npm run test:watch

Building

npm run build

License

MIT