task_parser.py

#!/usr/bin/env python3
"""
Task parser utility for Antigravity CLI.

Extracts and analyzes task.md files to provide structured access to task status
and metadata. Designed to be imported and used by antigravity-cli.py.
"""

import os
import json
from pathlib import Path
from typing import Dict, List, Any

def parse_task_md(task_path: Path) -> Dict[str, List[Dict[str, Any]]]:
    """
    Read task.md and extract all checklist items.
    
    Parses each line to identify status:
      - [x] = completed
      - [/] = in_progress
      - [ ] = todo
    
    Extracts the ID from each line (e.g., <!-- id: 0 -->)
    
    Returns structured data with lists for completed, in_progress, and todo items.
    Each item is a dict: { 'text': str, 'id': int }
    
    Args:
        task_path: Path to task.md file
        
    Returns:
        Dict with keys: 'completed', 'in_progress', 'todo'
    """
    result = {
        'completed': [],
        'in_progress': [],
        'todo': []
    }
    
    task_md_path = task_path / "task.md"
    if not task_md_path.exists():
        print(f"Error: task.md not found at {task_md_path}", file=sys.stderr)
        return result
    
    try:
        with open(task_md_path, 'r') as f:
            lines = f.readlines()
    except Exception as e:
        print(f"Error reading task.md: {e}", file=sys.stderr)
        return result
    
    for line in lines:
        # Skip empty lines and headers
        line = line.strip()
        if not line or line.startswith('#'):
            continue
            
        # Check for checklist item pattern
        if line.startswith('[x]') or line.startswith('[/]') or line.startswith('[ ]'):
            # Extract status
            if line.startswith('[x]'):
                status = 'completed'
                text = line[3:].strip()
            elif line.startswith('[/]'):
                status = 'in_progress'
                text = line[3:].strip()
            elif line.startswith('[ ]'):
                status = 'todo'
                text = line[3:].strip()
            else:
                continue
                
            # Extract ID from HTML comment
            item_id = None
            if '<!-- id:' in line:
                try:
                    # Extract between <!-- id: and -->
                    start = line.find('<!-- id:') + len('<!-- id:')
                    end = line.find('-->', start)
                    if end != -1:
                        item_id_str = line[start:end].strip()
                        item_id = int(item_id_str)
                except ValueError:
                    item_id = None
            
            # Add to result
            result[status].append({
                'text': text,
                'id': item_id
            })
    
    return result

def get_task_summary(task_path: Path) -> Dict[str, Any]:
    """
    Return task summary including:
      - task name (first line after # in task.md)
      - version and updatedAt from task.md.metadata.json
      - GUID from the directory name
    
    Args:
        task_path: Path to task directory
        
    Returns:
        Dict with keys: 'name', 'version', 'updatedAt', 'guid'
    """
    # Extract GUID from directory name
    guid = task_path.name
    
    # Extract task name from task.md
    task_name = "Unknown Task"
    task_md_path = task_path / "task.md"
    if task_md_path.exists():
        try:
            with open(task_md_path, 'r') as f:
                first_line = f.readline().strip()
                if first_line.startswith('#'):
                    task_name = first_line[1:].strip()
                else:
                    task_name = first_line
        except Exception:
            pass
    
    # Extract version and updatedAt from metadata
    version = "unknown"
    updated_at = "unknown"
    meta_path = task_path / "task.md.metadata.json"
    if meta_path.exists():
        try:
            with open(meta_path, 'r') as f:
                metadata = json.load(f)
                version = metadata.get('version', 'unknown')
                updated_at = metadata.get('updatedAt', 'unknown')
        except json.JSONDecodeError:
            pass
        except Exception:
            pass
    
    return {
        'name': task_name,
        'version': version,
        'updatedAt': updated_at,
        'guid': guid
    }