"""
Flask Web Application for Multimodal Glaucoma Detection
Provides web interface for Fundus, OCT, and Multimodal predictions
"""

from flask import Flask, render_template, request, jsonify, send_from_directory
import os
import sys

# Ensure the project root is in sys.path
sys.path.append(os.path.dirname(os.path.abspath(__file__)))

from pathlib import Path
import base64
from io import BytesIO
from models.load_models import ModelLoader
from models.predict import predict_fundus, predict_oct, predict_multimodal
from utils.gradcam import generate_gradcam_image
import traceback

app = Flask(__name__)
app.config['UPLOAD_FOLDER'] = 'static/uploads'
app.config['MAX_CONTENT_LENGTH'] = 16 * 1024 * 1024  # 16MB max file size

# Ensure upload directory exists
os.makedirs(app.config['UPLOAD_FOLDER'], exist_ok=True)

# Load models on startup
print("Loading models...")
model_loader = ModelLoader()
print("Models loaded successfully!")

@app.route('/')
def index():
    """Landing page with mode selection"""
    return render_template('index.html')

@app.route('/fundus')
def fundus_page():
    """Fundus analysis page"""
    return render_template('upload.html', mode='fundus')

@app.route('/oct')
def oct_page():
    """OCT analysis page"""
    return render_template('upload.html', mode='oct')

@app.route('/multimodal')
def multimodal_page():
    """Multimodal analysis page"""
    return render_template('upload.html', mode='multimodal')

@app.route('/about')
def about_page():
    """About page"""
    return render_template('about.html')

@app.route('/predict/fundus', methods=['POST'])
def predict_fundus_endpoint():
    """Fundus prediction endpoint"""
    try:
        if 'fundus_image' not in request.files:
            return jsonify({'error': 'No fundus image provided'}), 400
        
        file = request.files['fundus_image']
        if file.filename == '':
            return jsonify({'error': 'No file selected'}), 400
        
        # Save uploaded file
        filepath = os.path.join(app.config['UPLOAD_FOLDER'], 'fundus_temp.jpg')
        file.save(filepath)
        
        # Get prediction
        result = predict_fundus(filepath, model_loader)
        
        return jsonify(result)
    
    except Exception as e:
        print(f"Error in fundus prediction: {e}")
        traceback.print_exc()
        return jsonify({'error': str(e)}), 500

@app.route('/predict/oct', methods=['POST'])
def predict_oct_endpoint():
    """OCT prediction endpoint"""
    try:
        if 'oct_image' not in request.files:
            return jsonify({'error': 'No OCT image provided'}), 400
        
        file = request.files['oct_image']
        if file.filename == '':
            return jsonify({'error': 'No file selected'}), 400
        
        # Save uploaded file
        filepath = os.path.join(app.config['UPLOAD_FOLDER'], 'oct_temp.jpg')
        file.save(filepath)
        
        # Get prediction
        result = predict_oct(filepath, model_loader)
        
        return jsonify(result)
    
    except Exception as e:
        print(f"Error in OCT prediction: {e}")
        traceback.print_exc()
        return jsonify({'error': str(e)}), 500

@app.route('/predict/multimodal', methods=['POST'])
def predict_multimodal_endpoint():
    """Multimodal prediction endpoint"""
    try:
        if 'fundus_image' not in request.files or 'oct_image' not in request.files:
            return jsonify({'error': 'Both fundus and OCT images required'}), 400
        
        fundus_file = request.files['fundus_image']
        oct_file = request.files['oct_image']
        
        if fundus_file.filename == '' or oct_file.filename == '':
            return jsonify({'error': 'No file selected'}), 400
        
        # Save uploaded files
        fundus_path = os.path.join(app.config['UPLOAD_FOLDER'], 'fundus_temp.jpg')
        oct_path = os.path.join(app.config['UPLOAD_FOLDER'], 'oct_temp.jpg')
        fundus_file.save(fundus_path)
        oct_file.save(oct_path)
        
        # Get prediction
        result = predict_multimodal(fundus_path, oct_path, model_loader)
        
        return jsonify(result)
    
    except Exception as e:
        print(f"Error in multimodal prediction: {e}")
        traceback.print_exc()
        return jsonify({'error': str(e)}), 500

if __name__ == '__main__':
    print("\n" + "="*60)
    print("Multimodal Glaucoma Detection System")
    print("="*60)
    print("Server starting on http://localhost:5000")
    print("Press Ctrl+C to stop")
    print("="*60 + "\n")
    app.run(debug=False, host='0.0.0.0', port=7860)