server/tests/test_enhanced_modules.py

"""
Test Enhanced Modules with Datasets

Tests healthcare, education, and intent modules with dataset integration.

Run with: python test_enhanced_modules.py
"""

import sys
import os

# Add parent directory to path
sys.path.append(os.path.dirname(os.path.dirname(os.path.abspath(__file__))))

from app.services.enhanced_healthcare import get_enhanced_healthcare_module
from app.services.enhanced_education import get_enhanced_education_module
from app.services.intent_trainer import get_intent_trainer
from app.services.dataset_manager import get_dataset_manager


def test_enhanced_healthcare():
    """Test enhanced healthcare with MedQuAD and SymCAT"""
    print("=" * 60)
    print("TESTING ENHANCED HEALTHCARE MODULE")
    print("=" * 60)
    
    healthcare = get_enhanced_healthcare_module()
    
    # Show loaded knowledge stats
    stats = healthcare.get_knowledge_stats()
    print("\n📊 Knowledge Base Stats:")
    for key, value in stats.items():
        print(f"  {key}: {value}")
    
    # Test medical Q&A
    print("\n🏥 Testing Medical Q&A:")
    questions = [
        "What is diabetes?",
        "What are the symptoms of flu?",
        "How is pneumonia treated?"
    ]
    
    for question in questions:
        print(f"\n  Q: {question}")
        answer = healthcare.answer_medical_question(question)
        if answer:
            print(f"  A: {answer['answer'][:100]}...")
            print(f"  Source: {answer['source']}, Confidence: {answer['confidence']:.2f}")
        else:
            print(f"  No answer found")
    
    # Test enhanced symptom checker
    print("\n🩺 Testing Enhanced Symptom Checker:")
    symptoms = ["fever", "cough"]
    result = healthcare.check_symptoms_enhanced(symptoms)
    print(f"  Symptoms: {', '.join(symptoms)}")
    print(f"  Possible Conditions: {', '.join(result['possible_conditions'][:5])}")
    print(f"  Severity: {result['severity']}")
    print(f"  Data Source: {result.get('data_source', 'Base')}")


def test_enhanced_education():
    """Test enhanced education with CourseQ and STACKED"""
    print("\n" + "=" * 60)
    print("TESTING ENHANCED EDUCATION MODULE")
    print("=" * 60)
    
    education = get_enhanced_education_module()
    
    # Show loaded knowledge stats
    stats = education.get_knowledge_stats()
    print("\n📊 Knowledge Base Stats:")
    for key, value in stats.items():
        print(f"  {key}: {value}")
    
    # Test academic Q&A
    print("\n🎓 Testing Academic Q&A:")
    questions = [
        "What is the GPA requirement for admission?",
        "How do I register for courses?",
        "What is a prerequisite course?"
    ]
    
    for question in questions:
        print(f"\n  Q: {question}")
        answer = education.answer_academic_question(question)
        if answer:
            print(f"  A: {answer['answer'][:100]}...")
            print(f"  Source: {answer['source']}, Confidence: {answer['confidence']:.2f}")
        else:
            print(f"  No answer found")
    
    # Test admission information
    print("\n📝 Testing Admission Information:")
    admission_info = education.get_admission_information()
    print(f"  Total admission resources: {admission_info['total_resources']}")
    print(f"  Checklist steps: {len(admission_info['admission_checklist'])}")


def test_intent_trainer():
    """Test intent classifier training and evaluation"""
    print("\n" + "=" * 60)
    print("TESTING INTENT CLASSIFIER TRAINER")
    print("=" * 60)
    
    trainer = get_intent_trainer()
    
    # Evaluate on CLINC150 test set
    print("\n📈 Evaluating on CLINC150 Test Set:")
    results = trainer.evaluate_classifier(dataset="clinc150", split="test")
    
    print(f"\n  Dataset: {results['dataset']}")
    print(f"  Accuracy: {results['accuracy']:.2%} ({results['correct']}/{results['total']})")
    
    # Show category distribution
    print(f"\n  Category Distribution:")
    dist = results['category_distribution']
    print(f"    Predicted: {dist['predicted']}")
    print(f"    Expected: {dist['expected']}")
    
    # Sample predictions
    print(f"\n  Sample Predictions:")
    for pred in results['predictions'][:3]:
        status = "✓" if pred['correct'] else "✗"
        print(f"\n  {status} '{pred['text']}'")
        print(f"    Expected: {pred['expected_category']}")
        print(f"    Predicted: {pred['predicted_category']} (conf: {pred['confidence']:.2f})")
    
    # Analyze misclassifications
    print("\n🔍 Misclassification Analysis:")
    analysis = trainer.analyze_misclassifications(results)
    print(f"  Total misclassified: {analysis['total_misclassified']}")
    print(f"  Low confidence count: {analysis.get('low_confidence_count', 0)}")
    
    if analysis.get('confusion_matrix'):
        print(f"\n  Confusion Matrix:")
        for key, count in list(analysis['confusion_matrix'].items())[:5]:
            print(f"    {key}: {count}")
    
    # Get improvement suggestions
    print("\n💡 Improvement Suggestions:")
    suggestions = trainer.suggest_improvements(results)
    for suggestion in suggestions:
        print(f"  {suggestion}")


def test_dataset_manager():
    """Test dataset manager"""
    print("\n" + "=" * 60)
    print("TESTING DATASET MANAGER")
    print("=" * 60)
    
    manager = get_dataset_manager()
    
    # Get dataset info
    print("\n📦 Available Datasets:")
    info = manager.get_dataset_info()
    
    for name, details in info.items():
        status = "✓ Downloaded" if details['downloaded'] else "⏳ Not downloaded"
        print(f"\n  {name} ({details['type']})")
        print(f"    {status}")
        print(f"    {details['description']}")


def test_full_workflow():
    """Test complete workflow with all modules"""
    print("\n" + "=" * 60)
    print("TESTING COMPLETE WORKFLOW")
    print("=" * 60)
    
    healthcare = get_enhanced_healthcare_module()
    education = get_enhanced_education_module()
    
    # Simulate user queries
    test_queries = [
        {
            "query": "What is diabetes?",
            "industry": "healthcare",
            "expected_module": "healthcare"
        },
        {
            "query": "How to check my GPA?",
            "industry": "education",
            "expected_module": "education"
        },
        {
            "query": "I have fever and headache",
            "industry": "healthcare",
            "expected_module": "healthcare"
        }
    ]
    
    for test in test_queries:
        print(f"\n📩 Query: '{test['query']}'")
        print(f"   Industry: {test['industry']}")
        
        if test['industry'] == "healthcare":
            # Try medical Q&A first
            answer = healthcare.answer_medical_question(test['query'])
            if answer:
                print(f"   📖 Answer: {answer['answer'][:80]}...")
                print(f"   Source: {answer['source']}")
        
        elif test['industry'] == "education":
            # Try academic Q&A
            answer = education.answer_academic_question(test['query'])
            if answer:
                print(f"   📖 Answer: {answer['answer'][:80]}...")
                print(f"   Source: {answer['source']}")


if __name__ == "__main__":
    print("\n🚀 Starting Enhanced Module Tests with Datasets\n")
    
    try:
        # Test dataset manager
        test_dataset_manager()
        
        # Test enhanced modules
        test_enhanced_healthcare()
        test_enhanced_education()
        
        # Test intent trainer
        test_intent_trainer()
        
        # Test full workflow
        test_full_workflow()
        
        print("\n" + "=" * 60)
        print("✅ ALL ENHANCED MODULE TESTS COMPLETED")
        print("=" * 60 + "\n")
        
    except Exception as e:
        print(f"\n❌ TEST FAILED: {e}")
        import traceback
        traceback.print_exc()