Spaces:

abidlabs
/

closed2open

Running

File size: 7,468 Bytes

adc9dea

#!/usr/bin/env python3
"""
Script to convert Jupyter notebooks to recipe pages using nbgradio.
This script processes notebooks in the notebooks/ directory and generates
recipe pages in the recipes/ directory with proper styling integration.
"""

import os
import subprocess
import sys
from pathlib import Path
import shutil
import json

def install_nbgradio():
    """Install nbgradio if not already installed."""
    try:
        import nbgradio
        print("✅ nbgradio is already installed")
    except ImportError:
        print("📦 Installing nbgradio...")
        subprocess.run([sys.executable, "-m", "pip", "install", "nbgradio"], check=True)
        print("✅ nbgradio installed successfully")

def get_notebook_files():
    """Get all notebook files from the notebooks directory."""
    notebooks_dir = Path("notebooks")
    if not notebooks_dir.exists():
        print("❌ notebooks/ directory not found")
        return []
    
    notebook_files = list(notebooks_dir.glob("*.ipynb"))
    print(f"📚 Found {len(notebook_file)} notebook(s): {[f.name for f in notebook_files]}")
    return notebook_files

def get_existing_recipes():
    """Get list of existing recipe files."""
    recipes_dir = Path("recipes")
    if not recipes_dir.exists():
        return []
    
    recipe_files = list(recipes_dir.glob("*.html"))
    return [f.stem for f in recipe_files]

def convert_notebook_to_recipe(notebook_path):
    """Convert a single notebook to a recipe page."""
    notebook_name = notebook_path.stem
    print(f"\n🔄 Converting {notebook_name}...")
    
    # Create temporary directory for nbgradio output
    temp_dir = Path(f"temp_{notebook_name}")
    temp_dir.mkdir(exist_ok=True)
    
    try:
        # Run nbgradio with --spaces and --fragment
        cmd = [
            "nbgradio", "build", 
            str(notebook_path),
            "--spaces",
            "--fragment",
            "--output-dir", str(temp_dir)
        ]
        
        print(f"🚀 Running: {' '.join(cmd)}")
        result = subprocess.run(cmd, capture_output=True, text=True)
        
        if result.returncode != 0:
            print(f"❌ Error converting {notebook_name}:")
            print(result.stderr)
            return False
        
        print(f"✅ Successfully converted {notebook_name}")
        
        # Process the generated fragment
        fragment_file = temp_dir / "fragments" / f"{notebook_name}.html"
        if fragment_file.exists():
            create_recipe_page(notebook_name, fragment_file)
            return True
        else:
            print(f"❌ Fragment file not found: {fragment_file}")
            return False
            
    except Exception as e:
        print(f"❌ Error processing {notebook_name}: {e}")
        return False
    finally:
        # Clean up temporary directory
        if temp_dir.exists():
            shutil.rmtree(temp_dir)

def create_recipe_page(notebook_name, fragment_file):
    """Create a complete recipe page from the fragment."""
    recipes_dir = Path("recipes")
    recipes_dir.mkdir(exist_ok=True)
    
    # Read the fragment content
    with open(fragment_file, 'r', encoding='utf-8') as f:
        fragment_content = f.read()
    
    # Extract title from notebook name (convert snake_case to Title Case)
    title = notebook_name.replace('_', ' ').replace('-', ' ').title()
    
    # Create the complete recipe page
    recipe_html = f"""<!doctype html>
<html>
	<head>
		<meta charset="utf-8" />
		<meta name="viewport" content="width=device-width" />
		<title>{title} - ClosedToOpen</title>
		<link rel="stylesheet" href="../style.css" />
	</head>
	<body>
		<nav class="navbar">
			<div class="nav-content">
				<h1 class="nav-title">ClosedToOpen</h1>
				<p class="nav-subtitle">how to stop relying on proprietary APIs for AI and start living freely</p>
			</div>
		</nav>

		<main class="main-content">
			<div class="breadcrumb">
				<a href="../index.html">← Back to Migration Guides</a>
			</div>
			
			<section class="recipe-header">
				<h1 class="recipe-title">{title}</h1>
				<p class="recipe-description">Migrate from proprietary APIs to open-source alternatives</p>
				<div class="recipe-meta">
					<span class="difficulty">Difficulty: Medium</span>
					<span class="time">Time: 45 minutes</span>
					<span class="category">Category: Migration</span>
				</div>
			</section>

			<section class="recipe-content">
				<div class="notebook-content">
					{fragment_content}
				</div>
			</section>
		</main>
	</body>
</html>"""
    
    # Write the recipe page
    recipe_file = recipes_dir / f"{notebook_name}.html"
    with open(recipe_file, 'w', encoding='utf-8') as f:
        f.write(recipe_html)
    
    print(f"📄 Created recipe page: {recipe_file}")

def update_notebook_for_nbgradio(notebook_path):
    """Update notebook to include nbgradio syntax for interactive cells."""
    print(f"🔧 Updating {notebook_path.name} for nbgradio...")
    
    # Read the notebook
    with open(notebook_path, 'r', encoding='utf-8') as f:
        notebook = json.load(f)
    
    updated = False
    
    # Look for cells with gradio code and add nbgradio comments
    for cell in notebook['cells']:
        if cell['cell_type'] == 'code':
            source = ''.join(cell['source'])
            
            # Check if this cell contains gradio code
            if 'import gradio' in source or 'gr.Interface' in source or 'demo.launch()' in source:
                # Check if it already has nbgradio comment
                if not source.startswith('#nbgradio'):
                    # Add nbgradio comment
                    notebook_name = notebook_path.stem
                    nbgradio_comment = f"#nbgradio name=\"{notebook_name}\"\n"
                    cell['source'] = [nbgradio_comment] + cell['source']
                    updated = True
                    print(f"  ✅ Added nbgradio comment to gradio cell")
    
    # Write back if updated
    if updated:
        with open(notebook_path, 'w', encoding='utf-8') as f:
            json.dump(notebook, f, indent=2)
        print(f"  📝 Updated notebook saved")
    else:
        print(f"  ℹ️  No gradio cells found or already updated")

def main():
    """Main function to convert all notebooks to recipes."""
    print("🚀 Starting notebook to recipe conversion...")
    
    # Install nbgradio
    install_nbgradio()
    
    # Get notebook files
    notebook_files = get_notebook_files()
    if not notebook_files:
        print("❌ No notebooks found to convert")
        return
    
    # Get existing recipes
    existing_recipes = get_existing_recipes()
    print(f"📋 Existing recipes: {existing_recipes}")
    
    # Convert each notebook
    converted_count = 0
    for notebook_path in notebook_files:
        notebook_name = notebook_path.stem
        
        # Skip if recipe already exists
        if notebook_name in existing_recipes:
            print(f"⏭️  Skipping {notebook_name} (recipe already exists)")
            continue
        
        # Update notebook for nbgradio
        update_notebook_for_nbgradio(notebook_path)
        
        # Convert to recipe
        if convert_notebook_to_recipe(notebook_path):
            converted_count += 1
    
    print(f"\n🎉 Conversion complete! {converted_count} new recipe(s) created.")
    print("📁 Check the recipes/ directory for the generated files.")

if __name__ == "__main__":
    main()