# test_multimodal_tools.py
import sys
import os
from dotenv import load_dotenv
import tempfile
import base64

# Load environment variables
load_dotenv()

# Add tools to path
sys.path.append(os.path.join(os.path.dirname(__file__), '..'))

def create_test_image():
    """Create a simple test image for testing"""
    try:
        from PIL import Image, ImageDraw
        
        # Create a simple test image
        img = Image.new('RGB', (200, 100), color='white')
        draw = ImageDraw.Draw(img)
        draw.text((10, 40), "Test Image", fill='black')
        
        # Save to temporary file
        temp_file = tempfile.NamedTemporaryFile(delete=False, suffix='.png')
        img.save(temp_file.name)
        return temp_file.name
    except ImportError:
        print("    ⚠️  PIL not available, creating mock image file")
        # Create a mock image file for testing
        temp_file = tempfile.NamedTemporaryFile(delete=False, suffix='.txt')
        temp_file.write(b"This is a mock image file for testing")
        temp_file.close()
        return temp_file.name

def test_multimodal_tools():
    """Test multimodal tool functions"""
    try:
        from tools.multimodal_tools import MultimodalTools, analyze_image, extract_text, analyze_transcript
        
        # Check OpenRouter API key
        openrouter_key = os.getenv("OPENROUTER_API_KEY")
        if not openrouter_key:
            print("    ❌ OpenRouter API key not found!")
            print("    Please add OPENROUTER_API_KEY to your .env file")
            return
        
        print("  🎨 Initializing multimodal tools...")
        multimodal_tools = MultimodalTools()
        print(f"    ✅ API key loaded: {openrouter_key[:8]}...")
        
        # Test with transcript (doesn't require image)
        print("  🎙️  Testing audio transcript analysis...")
        test_transcript = "Hello, this is a test transcript. We're discussing artificial intelligence and machine learning technologies."
        
        transcript_result = multimodal_tools.analyze_audio_transcript(
            test_transcript, 
            "What is the main topic of this transcript?"
        )
        
        if transcript_result and "Error" not in transcript_result:
            print("    ✅ Transcript analysis: PASSED")
            print(f"    Result: {transcript_result[:100]}...")
        else:
            print(f"    ❌ Transcript analysis failed: {transcript_result}")
        
        # Test convenience function
        print("  🔧 Testing convenience functions...")
        convenience_result = analyze_transcript(
            "This is a test about Python programming and software development.",
            "Summarize the main topics"
        )
        
        if convenience_result and "Error" not in convenience_result:
            print("    ✅ Convenience function: PASSED")
        else:
            print(f"    ⚠️  Convenience function result: {convenience_result}")
        
        # Test image analysis (if we can create a test image)
        print("  🖼️  Testing image analysis...")
        try:
            test_image_path = create_test_image()
            
            # Test image description
            image_result = multimodal_tools.analyze_image(
                test_image_path,
                "Describe what you see in this image"
            )
            
            if image_result and "Error" not in image_result:
                print("    ✅ Image analysis: PASSED")
                print(f"    Result: {image_result[:100]}...")
            else:
                print(f"    ⚠️  Image analysis result: {image_result}")
            
            # Test OCR functionality
            ocr_result = multimodal_tools.extract_text_from_image(test_image_path)
            if ocr_result and "Error" not in ocr_result:
                print("    ✅ OCR extraction: PASSED")
                print(f"    Extracted: {ocr_result[:50]}...")
            else:
                print(f"    ⚠️  OCR result: {ocr_result}")
            
            # Clean up test image
            os.unlink(test_image_path)
            
        except Exception as e:
            print(f"    ⚠️  Image testing skipped: {str(e)}")
        
        # Test error handling
        print("  🚨 Testing error handling...")
        error_result = multimodal_tools.analyze_image(
            "nonexistent_file.jpg",
            "This should fail"
        )
        
        if "Error" in error_result:
            print("    ✅ Error handling: PASSED")
        else:
            print("    ⚠️  Error handling unexpected result")
        
        # Test empty transcript
        empty_result = multimodal_tools.analyze_audio_transcript("", "Question")
        if "Error" in empty_result:
            print("    ✅ Empty input handling: PASSED")
        
        print("  ✅ MULTIMODAL TESTS COMPLETED!")
        
        # Demo output
        print("\n  📋 Multimodal Tools Demo:")
        demo_transcript = "We discussed the implementation of AI agents using CrewAI framework for automated workflows."
        demo_result = analyze_transcript(demo_transcript, "What framework was mentioned?")
        print(f"    Question: What framework was mentioned?")
        print(f"    Answer: {demo_result[:150]}...")
        
    except ImportError as e:
        print(f"    ❌ Import error: {e}")
        print("    Make sure all dependencies are installed")
    except Exception as e:
        print(f"    ❌ Test failed: {e}")
        import traceback
        traceback.print_exc()

if __name__ == "__main__":
    test_multimodal_tools()