Combining Attributes

Carapace attributes can be combined to create powerful, flexible DTOs. This advanced guide shows how to leverage multiple attributes together for complex scenarios.

Attribute Processing Order

When multiple attributes are applied to a property, they are processed in a specific order:

1. Pre-hydration (MapFrom, etc.)
2. Hydration (value assignment)
3. Post-hydration (not currently implemented)
4. Serialization (MapTo, Hidden, etc.)

Understanding this order is important when combining attributes, as it determines how they interact.

Basic Combinations

You can apply multiple attributes to a single property to achieve combined effects:

use Alamellama\Carapace\Attributes\CastWith;
use Alamellama\Carapace\Attributes\MapFrom;
use Alamellama\Carapace\Attributes\MapTo;
use Alamellama\Carapace\Attributes\Hidden;
use Alamellama\Carapace\ImmutableDTO;

final class User extends ImmutableDTO
{
    public function __construct(
        #[MapFrom('display_name')]
        #[MapFrom('user_name')]
        #[MapTo('full_name')]
        public string $name,

        #[MapFrom('user_email')]
        #[MapTo('email_address')]
        public string $email,

        #[Hidden]
        public string $password,
    ) {}
}

In this example:

• name is mapped from user_name or display_name in the input and to full_name in the output
• email is mapped from user_email in the input and to email_address in the output
• password is excluded from serialization

Mapping and Casting

Combine MapFrom with CastWith to both map and transform data:

final class Order extends ImmutableDTO
{
    public function __construct(
        public string $id,

        #[MapFrom('order_date')]
        #[CastWith(new DateTimeCaster('Y-m-d'))]
        public DateTimeInterface $date,

        #[MapFrom('items')]
        #[CastWith(OrderItem::class)]
        public array $orderItems,
    ) {}
}

This allows you to:

1. Accept data with different key names
2. Automatically cast that data to the appropriate types