import pandas as pd
import numpy as np

def get_model_expected_features():
    """Return all features that the trained model expects"""
    # Based on the original training data, these are typical features
    expected_features = [
        # Basic transaction features
        'TransactionAmt', 'TransactionDT',
        
        # Card features
        'card1', 'card2', 'card3', 'card4', 'card5', 'card6',
        
        # Address features
        'addr1', 'addr2',
        
        # Distance features  
        'dist1', 'dist2',
        
        # Email features
        'P_emaildomain', 'R_emaildomain',
        
        # Count features (C1-C14)
        'C1', 'C2', 'C3', 'C4', 'C5', 'C6', 'C7', 'C8', 'C9', 'C10', 'C11', 'C12', 'C13', 'C14',
        
        # Time delta features (D1-D15)
        'D1', 'D2', 'D3', 'D4', 'D5', 'D6', 'D7', 'D8', 'D9', 'D10', 'D11', 'D12', 'D13', 'D14', 'D15',
        
        # Match features (M1-M9)
        'M1', 'M2', 'M3', 'M4', 'M5', 'M6', 'M7', 'M8', 'M9',
        
        # Vesta features (V1-V339) - Full range based on error message
        *[f'V{i}' for i in range(1, 340)],
        
        # Identity features (id_01 to id_38)
        *[f'id_{i:02d}' for i in range(1, 39)],
        
        # Device features
        'DeviceType', 'DeviceInfo',
        
        # Additional features that might be present
        'ProductCD'
    ]
    
    return expected_features

def fill_missing_features(transaction_data):
    """Fill missing features with appropriate default values"""
    
    # Get all expected features
    expected_features = get_model_expected_features()
    
    # Default values for different feature types
    defaults = {
        # Numeric features default to 0 or reasonable values
        'card1': 13553, 'card2': 150.0, 'card3': 150.0, 'card5': 142.0,
        'addr1': 325.0, 'addr2': 87.0,
        'dist1': 19.0, 'dist2': 19.0,
        
        # Count features (C1-C14) - mostly 0 or 1
        **{f'C{i}': 0.0 for i in range(1, 15)},
        'C1': 1.0, 'C2': 1.0, 'C6': 1.0, 'C9': 1.0, 'C11': 1.0, 'C12': 1.0, 'C13': 1.0, 'C14': 1.0,
        
        # Time delta features (D1-D15) - mostly 0
        **{f'D{i}': 0.0 for i in range(1, 16)},
        'D5': 20.0,  # Common non-zero value
        
        # Match features (M1-M9) - mostly F with some T
        **{f'M{i}': 'F' for i in range(1, 10)},
        'M1': 'T', 'M2': 'T', 'M3': 'T',
        'M4': 'M0',  # Special case
        
        # Vesta features (V1-V339) - default to 1.0
        **{f'V{i}': 1.0 for i in range(1, 340)},
        
        # Identity features (id_01 to id_38) - default to 0.0
        **{f'id_{i:02d}': 0.0 for i in range(1, 39)},
        
        # Categorical features
        'card4': 'visa',
        'card6': 'credit',
        'P_emaildomain': 'gmail.com',
        'R_emaildomain': 'gmail.com',
        'DeviceType': 'desktop',
        'DeviceInfo': 'Windows',
        'ProductCD': 'W',
        
        # Transaction defaults
        'TransactionDT': 86400,  # Default timestamp
    }
    
    # Create complete transaction data
    complete_data = {}
    
    # First, add all provided data
    complete_data.update(transaction_data)
    
    # Then fill missing features with defaults
    for feature in expected_features:
        if feature not in complete_data:
            complete_data[feature] = defaults.get(feature, 0.0)
    
    return complete_data

def create_simple_transaction(amount, card_type="visa", email_domain="gmail.com", hour=12):
    """Create a transaction with minimal inputs and smart defaults"""
    
    transaction_data = {
        'TransactionAmt': float(amount),
        'TransactionDT': hour * 3600,
        'card4': card_type,
        'P_emaildomain': email_domain,
        'R_emaildomain': email_domain,
    }
    
    # Fill all missing features
    return fill_missing_features(transaction_data)

def validate_features(df, expected_features):
    """Validate that DataFrame has all expected features"""
    missing_features = set(expected_features) - set(df.columns)
    extra_features = set(df.columns) - set(expected_features)
    
    return {
        'missing': list(missing_features),
        'extra': list(extra_features),
        'is_valid': len(missing_features) == 0
    }