test_playwright.py

#!/usr/bin/env python3
"""
Test script for Playwright integration
Run this to verify Playwright is working correctly
"""

import asyncio
import sys
import os

# Add backend to path
sys.path.insert(0, os.path.join(os.path.dirname(__file__), 'backend'))

from backend.preprocessing.scraper import fetch
from backend.config import PLAYWRIGHT_HEADLESS, PLAYWRIGHT_BROWSER

def test_playwright_fetch():
    """Test basic Playwright fetch functionality"""
    print("🧪 Testing Playwright integration...")
    print(f"Browser: {PLAYWRIGHT_BROWSER}")
    print(f"Headless: {PLAYWRIGHT_HEADLESS}")
    
    # Test URLs
    test_urls = [
        "https://httpbin.org/html",
        "https://example.com",
        "https://httpbin.org/json"
    ]
    
    for url in test_urls:
        print(f"\n🔍 Testing: {url}")
        try:
            html = fetch(url, retries=1, timeout=10)
            if html:
                print(f"✅ Success: Got {len(html)} characters")
                # Check for basic HTML structure
                if "<html" in html.lower() or "<!doctype" in html.lower():
                    print("✅ Valid HTML detected")
                else:
                    print("⚠️  No HTML structure found")
            else:
                print("❌ Failed: No content returned")
        except Exception as e:
            print(f"❌ Error: {e}")
    
    print("\n🎉 Playwright test complete!")

def test_playwright_import():
    """Test if Playwright can be imported"""
    try:
        from playwright.async_api import async_playwright
        print("✅ Playwright import successful")
        return True
    except ImportError as e:
        print(f"❌ Playwright import failed: {e}")
        print("Please install Playwright: pip install playwright")
        return False

def main():
    print("🦕 DataSaurus Playwright Test")
    print("=" * 40)
    
    # Test import
    if not test_playwright_import():
        sys.exit(1)
    
    # Test basic functionality
    try:
        test_playwright_fetch()
    except Exception as e:
        print(f"❌ Test failed: {e}")
        sys.exit(1)

if __name__ == "__main__":
    main()