setup_playwright.py

#!/usr/bin/env python3
"""
Setup script for Playwright browsers
Run this after installing requirements to download browser binaries
"""

import subprocess
import sys
import os

def install_playwright_browsers():
    """Install Playwright browser binaries"""
    print("Installing Playwright browser binaries...")
    try:
        # Install browsers
        result = subprocess.run([
            sys.executable, "-m", "playwright", "install"
        ], check=True, capture_output=True, text=True)
        
        print("✅ Playwright browsers installed successfully!")
        print(result.stdout)
        
        # Install system dependencies (for Linux)
        if os.name == 'posix' and sys.platform != 'darwin':
            print("\nInstalling system dependencies...")
            try:
                subprocess.run([
                    sys.executable, "-m", "playwright", "install-deps"
                ], check=True, capture_output=True, text=True)
                print("✅ System dependencies installed!")
            except subprocess.CalledProcessError as e:
                print(f"⚠️  System dependencies installation failed: {e}")
                print("You may need to install them manually:")
                print("sudo apt-get install libnss3 libatk-bridge2.0-0 libdrm2 libxkbcommon0 libxcomposite1 libxdamage1 libxrandr2 libgbm1 libxss1 libasound2")
        
        return True
        
    except subprocess.CalledProcessError as e:
        print(f"❌ Failed to install Playwright browsers: {e}")
        print(f"Error output: {e.stderr}")
        return False
    except FileNotFoundError:
        print("❌ Playwright not found. Please install it first:")
        print("pip install playwright")
        return False

def check_playwright_installation():
    """Check if Playwright is properly installed"""
    try:
        import playwright
        print("✅ Playwright installed successfully")
        return True
    except ImportError:
        print("❌ Playwright not installed. Please run:")
        print("pip install playwright")
        return False

def main():
    print("🦕 DataSaurus Playwright Setup")
    print("=" * 40)
    
    # Check if Playwright is installed
    if not check_playwright_installation():
        sys.exit(1)
    
    # Install browsers
    if install_playwright_browsers():
        print("\n🎉 Setup complete! You can now use Playwright for scraping.")
        print("\nConfiguration options in backend/config.py:")
        print("- PLAYWRIGHT_HEADLESS: True/False")
        print("- PLAYWRIGHT_BROWSER: 'chromium', 'firefox', or 'webkit'")
        print("- PLAYWRIGHT_VIEWPORT: {'width': 1920, 'height': 1080}")
        print("- PLAYWRIGHT_WAIT_FOR_LOAD: 'load', 'domcontentloaded', or 'networkidle'")
        print("- PLAYWRIGHT_SCREENSHOT: True/False (for debugging)")
    else:
        print("\n❌ Setup failed. Please check the errors above.")
        sys.exit(1)

if __name__ == "__main__":
    main()