setup_dev.py

#!/usr/bin/env python3
"""
GraphFlow Development Setup and Testing Script

This script helps with development setup, testing, and validation of the GraphFlow framework.
"""

import subprocess
import sys
from pathlib import Path


def run_command(cmd, description):
    """Run a command and handle errors."""
    print(f"\n🔧 {description}")
    print(f"   Running: {cmd}")
    
    try:
        result = subprocess.run(cmd, shell=True, check=True, capture_output=True, text=True)
        print(f"   ✅ Success")
        if result.stdout.strip():
            print(f"   Output: {result.stdout.strip()}")
        return True
    except subprocess.CalledProcessError as e:
        print(f"   ❌ Failed: {e}")
        if e.stdout:
            print(f"   Stdout: {e.stdout}")
        if e.stderr:
            print(f"   Stderr: {e.stderr}")
        return False


def main():
    """Main setup and testing routine."""
    print("🚀 GraphFlow Development Setup and Testing")
    print("=" * 50)
    
    # Check Python version
    python_version = sys.version_info
    print(f"🐍 Python version: {python_version.major}.{python_version.minor}.{python_version.micro}")
    
    if python_version < (3, 11):
        print("❌ Python 3.11+ is required")
        sys.exit(1)
    
    # Install package in development mode
    print("\n📦 Installing GraphFlow in development mode...")
    if not run_command("pip install -e .", "Installing package in editable mode"):
        print("❌ Failed to install package")
        sys.exit(1)
    
    # Install development dependencies
    print("\n📚 Installing development dependencies...")
    dev_deps = [
        "pytest>=7.4.0",
        "pytest-cov>=4.1.0", 
        "black>=23.7.0",
        "isort>=5.12.0",
        "mypy>=1.5.0",
        "rich>=13.0.0",
        "typer>=0.9.0"
    ]
    
    for dep in dev_deps:
        if not run_command(f"pip install '{dep}'", f"Installing {dep}"):
            print(f"⚠️ Warning: Failed to install {dep}")
    
    # Run code formatting
    print("\n🎨 Formatting code with black...")
    run_command("black graphflow/ tests/ examples/ --line-length 88", "Formatting code")
    
    print("\n📋 Sorting imports with isort...")
    run_command("isort graphflow/ tests/ examples/ --profile black", "Sorting imports")
    
    # Run type checking
    print("\n🔍 Running type checks with mypy...")
    run_command("mypy graphflow/ --ignore-missing-imports", "Type checking")
    
    # Run tests
    print("\n🧪 Running tests...")
    if not run_command("python -m pytest tests/ -v", "Running test suite"):
        print("⚠️ Some tests failed, but continuing...")
    
    # Test basic imports
    print("\n📥 Testing basic imports...")
    try:
        from graphflow import Pipeline, node, context, dataset
        print("   ✅ Core imports successful")
        
        # Test basic functionality
        ctx = context(test_param="test_value")
        pipeline = Pipeline("test", "s3://test", ctx)
        
        @node()
        def test_node(df, test_param):
            return f"processed-{test_param}"
        
        pipeline.add_node(test_node)
        print("   ✅ Basic functionality test passed")
        
    except Exception as e:
        print(f"   ❌ Import/functionality test failed: {e}")
    
    # Test CLI
    print("\n🖥️ Testing CLI...")
    if not run_command("python -m graphflow.cli.main --help", "Testing CLI help"):
        print("⚠️ CLI test failed")
    
    # Run examples
    print("\n📖 Testing examples...")
    
    print("\n   Running basic example...")
    if not run_command("python examples/basic_example.py", "Running basic example"):
        print("⚠️ Basic example failed")
    
    print("\n   Running advanced example (limited)...")
    # Advanced example might be too resource intensive, so we just check syntax
    if not run_command("python -m py_compile examples/advanced_example.py", "Compiling advanced example"):
        print("⚠️ Advanced example has syntax errors")
    
    # Project structure validation
    print("\n📁 Validating project structure...")
    required_files = [
        "graphflow/__init__.py",
        "graphflow/core/__init__.py",
        "graphflow/core/context.py",
        "graphflow/core/dataset.py",
        "graphflow/core/decorators.py",
        "graphflow/core/pipeline.py",
        "graphflow/core/executors.py",
        "graphflow/cli/__init__.py",
        "graphflow/cli/main.py",
        "pyproject.toml",
        "README.md",
        "LICENSE",
        "tests/test_core.py",
        "examples/basic_example.py",
        "examples/advanced_example.py"
    ]
    
    missing_files = []
    for file_path in required_files:
        if not Path(file_path).exists():
            missing_files.append(file_path)
    
    if missing_files:
        print(f"   ❌ Missing files: {missing_files}")
    else:
        print("   ✅ All required files present")
    
    # Final summary
    print("\n" + "=" * 50)
    print("🎯 Setup and Testing Complete!")
    print("\nNext steps:")
    print("1. 📖 Read the documentation in README.md")
    print("2. 🚀 Try the examples: python examples/basic_example.py")
    print("3. 🔧 Create a new project: python -m graphflow.cli.main init my_project")
    print("4. 🧪 Run tests: pytest tests/ -v")
    print("5. 📊 Check coverage: pytest tests/ --cov=graphflow")
    
    print("\n🎉 GraphFlow is ready for development!")
    
    # Show quick usage
    print("\n" + "=" * 50)
    print("📚 Quick Usage Example:")
    print("""
from graphflow import Pipeline, node, context, dataset
import pandas as pd

# Create context
ctx = context(multiplier=2, suffix="_processed")

# Create pipeline  
pipeline = Pipeline("demo", "data/", ctx)

# Define node with auto-context injection
@node(outputs=[dataset("result")])
def process_data(multiplier, suffix):  # Auto-injected!
    return f"result_{multiplier}{suffix}"

# Add and run
pipeline.add_node(process_data)
result = pipeline.run()
print(result.summary())
""")


if __name__ == "__main__":
    main()