import tensorflow as tf
import pandas as pd
import numpy as np
from sklearn.preprocessing import MinMaxScaler
import os

# Define the model path
model_path = "models/lstm_energy_model"

# Check if the model directory exists and load it
if not os.path.exists(model_path):
    raise FileNotFoundError(f"Error: Model directory {model_path} does not exist. Please run train_lstm_model.py to generate the model.")
try:
    model = tf.keras.models.load_model(model_path)
    print(f"✅ Model loaded successfully from {model_path}")
except Exception as e:
    raise RuntimeError(f"Error loading model from {model_path}: {e}")

def detect_energy_anomalies(file_path):
    """
    Detect anomalies in energy generation data (time-series) using LSTM.
    Args:
    - file_path (str): Path to the CSV/JSON file containing energy generation data

    Returns:
    - anomaly (str): Energy anomaly detection result
    """
    # Load energy generation data
    if not os.path.exists(file_path):
        raise FileNotFoundError(f"Error: Data file {file_path} not found.")
    try:
        data = pd.read_csv(file_path)
    except Exception as e:
        raise RuntimeError(f"Failed to load {file_path}: {e}")
    if 'energy_generation' not in data.columns:
        raise ValueError("CSV file must contain an 'energy_generation' column.")
    if len(data) < 60:
        raise ValueError(f"Dataset must have at least 60 rows for LSTM prediction. Found {len(data)} rows.")
    energy_data = data['energy_generation'].values.reshape(-1, 1)

    # Normalize data for LSTM
    scaler = MinMaxScaler(feature_range=(0, 1))
    energy_data = scaler.fit_transform(energy_data)

    # Prepare data for LSTM
    X = []
    for i in range(60, len(energy_data)):
        X.append(energy_data[i-60:i, 0])
    X = np.array(X)
    if X.shape[0] == 0:
        raise ValueError("Not enough data points to prepare for LSTM prediction (need at least 60 time steps).")
    X = X.reshape((X.shape[0], X.shape[1], 1))

    # Predict with LSTM model
    prediction = model.predict(X)
    last_predicted_value = prediction[-1][0]  # Get last prediction
    
    # Define threshold for anomaly
    threshold = 0.1
    anomaly = "Anomaly Detected" if abs(last_predicted_value - energy_data[-1][0]) > threshold else "No Anomaly"

    return anomaly