clawd/kanban/api.py

#!/usr/bin/env python3
"""
Simple API server for Echo Task Board.
Handles YouTube summarization requests.
"""

import json
import subprocess
import sys
import re
import os
from http.server import HTTPServer, SimpleHTTPRequestHandler
from urllib.parse import parse_qs, urlparse
from datetime import datetime
from pathlib import Path

BASE_DIR = Path(__file__).parent.parent
TOOLS_DIR = BASE_DIR / 'tools'
NOTES_DIR = BASE_DIR / 'notes' / 'youtube'
KANBAN_DIR = BASE_DIR / 'kanban'

class TaskBoardHandler(SimpleHTTPRequestHandler):
    
    def do_POST(self):
        if self.path == '/api/youtube':
            self.handle_youtube()
        elif self.path == '/api/files':
            self.handle_files_post()
        else:
            self.send_error(404)
    
    def handle_files_post(self):
        """Save file content."""
        try:
            content_length = int(self.headers['Content-Length'])
            post_data = self.rfile.read(content_length).decode('utf-8')
            data = json.loads(post_data)
            
            path = data.get('path', '')
            content = data.get('content', '')
            
            workspace = Path('/home/moltbot/clawd')
            target = (workspace / path).resolve()
            
            if not str(target).startswith(str(workspace)):
                self.send_json({'error': 'Access denied'}, 403)
                return
            
            # Create parent dirs if needed
            target.parent.mkdir(parents=True, exist_ok=True)
            
            # Write file
            target.write_text(content, encoding='utf-8')
            
            self.send_json({
                'status': 'saved',
                'path': path,
                'size': len(content)
            })
        except Exception as e:
            self.send_json({'error': str(e)}, 500)
    
    def do_GET(self):
        if self.path == '/api/status':
            self.send_json({'status': 'ok', 'time': datetime.now().isoformat()})
        elif self.path.startswith('/api/files'):
            self.handle_files_get()
        elif self.path.startswith('/api/'):
            self.send_error(404)
        else:
            # Serve static files
            super().do_GET()
    
    def handle_files_get(self):
        """List files or get file content."""
        from urllib.parse import urlparse, parse_qs
        parsed = urlparse(self.path)
        params = parse_qs(parsed.query)
        
        path = params.get('path', [''])[0]
        action = params.get('action', ['list'])[0]
        
        # Security: only allow access within workspace
        workspace = Path('/home/moltbot/clawd')
        try:
            target = (workspace / path).resolve()
            if not str(target).startswith(str(workspace)):
                self.send_json({'error': 'Access denied'}, 403)
                return
        except:
            self.send_json({'error': 'Invalid path'}, 400)
            return
        
        if action == 'list':
            if not target.exists():
                self.send_json({'error': 'Path not found'}, 404)
                return
            
            if target.is_file():
                # Return file content
                try:
                    content = target.read_text(encoding='utf-8', errors='replace')
                    self.send_json({
                        'type': 'file',
                        'path': path,
                        'name': target.name,
                        'content': content[:100000],  # Limit to 100KB
                        'size': target.stat().st_size,
                        'truncated': target.stat().st_size > 100000
                    })
                except Exception as e:
                    self.send_json({'error': str(e)}, 500)
            else:
                # List directory
                items = []
                try:
                    for item in sorted(target.iterdir()):
                        if item.name.startswith('.'):
                            continue
                        items.append({
                            'name': item.name,
                            'type': 'dir' if item.is_dir() else 'file',
                            'size': item.stat().st_size if item.is_file() else None,
                            'path': str(item.relative_to(workspace))
                        })
                    self.send_json({
                        'type': 'dir',
                        'path': path,
                        'items': items
                    })
                except Exception as e:
                    self.send_json({'error': str(e)}, 500)
        else:
            self.send_json({'error': 'Unknown action'}, 400)
    
    def handle_youtube(self):
        try:
            content_length = int(self.headers['Content-Length'])
            post_data = self.rfile.read(content_length).decode('utf-8')
            data = json.loads(post_data)
            
            url = data.get('url', '').strip()
            
            if not url or 'youtube.com' not in url and 'youtu.be' not in url:
                self.send_json({'error': 'URL YouTube invalid'}, 400)
                return
            
            # Process synchronously (simpler, avoids fork issues)
            try:
                print(f"Processing YouTube URL: {url}")
                result = process_youtube(url)
                print(f"Processing result: {result}")
                self.send_json({
                    'status': 'done',
                    'message': 'Notița a fost creată! Refresh pagina Notes.'
                })
            except Exception as e:
                import traceback
                print(f"YouTube processing error: {e}")
                traceback.print_exc()
                self.send_json({
                    'status': 'error',
                    'message': f'Eroare: {str(e)}'
                }, 500)
                
        except Exception as e:
            self.send_json({'error': str(e)}, 500)
    
    def send_json(self, data, code=200):
        self.send_response(code)
        self.send_header('Content-Type', 'application/json')
        self.send_header('Access-Control-Allow-Origin', '*')
        self.end_headers()
        self.wfile.write(json.dumps(data).encode())
    
    def do_OPTIONS(self):
        self.send_response(200)
        self.send_header('Access-Control-Allow-Origin', '*')
        self.send_header('Access-Control-Allow-Methods', 'GET, POST, OPTIONS')
        self.send_header('Access-Control-Allow-Headers', 'Content-Type')
        self.end_headers()


def process_youtube(url):
    """Download subtitles, summarize, save note."""
    import time
    
    # Get video info and subtitles
    yt_dlp = os.path.expanduser('~/.local/bin/yt-dlp')
    
    # Get title
    result = subprocess.run(
        [yt_dlp, '--dump-json', '--no-download', url],
        capture_output=True, text=True, timeout=30
    )
    
    if result.returncode != 0:
        print(f"Failed to get video info: {result.stderr}")
        return
    
    info = json.loads(result.stdout)
    title = info.get('title', 'Unknown')
    duration = info.get('duration', 0)
    video_id = info.get('id', 'unknown')
    
    # Download subtitles
    temp_dir = Path('/tmp/yt_subs')
    temp_dir.mkdir(exist_ok=True)
    
    for f in temp_dir.glob('*'):
        f.unlink()
    
    subprocess.run([
        yt_dlp, '--write-auto-subs', '--sub-langs', 'en',
        '--skip-download', '--sub-format', 'vtt',
        '-o', str(temp_dir / '%(id)s'),
        url
    ], capture_output=True, timeout=120)
    
    # Find and read subtitle file
    transcript = None
    for sub_file in temp_dir.glob('*.vtt'):
        content = sub_file.read_text(encoding='utf-8', errors='replace')
        transcript = clean_vtt(content)
        break
    
    if not transcript:
        print("No subtitles found")
        return
    
    # Create note filename
    date_str = datetime.now().strftime('%Y-%m-%d')
    slug = re.sub(r'[^\w\s-]', '', title.lower())[:50].strip().replace(' ', '-')
    filename = f"{date_str}_{slug}.md"
    
    # Create simple note (without AI summary for now - just transcript)
    note_content = f"""# {title}

**Video:** {url}
**Duration:** {duration // 60}:{duration % 60:02d}
**Saved:** {date_str}
**Tags:** #youtube #to-summarize

---

## Transcript

{transcript[:15000]}

---

*Notă: Sumarizarea va fi adăugată de Echo.*
"""
    
    # Save note
    NOTES_DIR.mkdir(parents=True, exist_ok=True)
    note_path = NOTES_DIR / filename
    note_path.write_text(note_content, encoding='utf-8')
    
    # Update index
    subprocess.run([
        sys.executable, str(TOOLS_DIR / 'update_notes_index.py')
    ], capture_output=True)
    
    # Add task to kanban
    subprocess.run([
        sys.executable, str(KANBAN_DIR / 'update_task.py'),
        'add', 'in-progress', f'Sumarizare: {title[:30]}...', url, 'medium'
    ], capture_output=True)
    
    print(f"Created note: {filename}")
    return filename


def clean_vtt(content):
    """Convert VTT to plain text."""
    lines = []
    seen = set()
    
    for line in content.split('\n'):
        if any([
            line.startswith('WEBVTT'),
            line.startswith('Kind:'),
            line.startswith('Language:'),
            '-->' in line,
            line.strip().startswith('<'),
            not line.strip(),
            re.match(r'^\d+$', line.strip())
        ]):
            continue
        
        clean = re.sub(r'<[^>]+>', '', line).strip()
        if clean and clean not in seen:
            seen.add(clean)
            lines.append(clean)
    
    return ' '.join(lines)


if __name__ == '__main__':
    port = 8080
    os.chdir(KANBAN_DIR)
    
    print(f"Starting Echo Task Board API on port {port}")
    httpd = HTTPServer(('0.0.0.0', port), TaskBoardHandler)
    httpd.serve_forever()