paulcote/serializedswift

A significant enhancement to the current Codable protocol for better and easier Serializing and Deserializing of JSON

SerializedSwift

A GSON inspired JSON decoding strategy in Swift using @propertyWrappers.

Features:

• No need to write own init(from decoder: Decoder)
• No need to write own CodingKeys subclass
• Works with inheritance and composition out of the box
• Custom Transformer classes
• Alternative coding keys
• Default values if JSON key is missing

struct Foo: Serializable {
    @Serialized
    var bar: String
    
    @Serialized("globalId")
    var id: String?
    
    @Serialized(alternateKey: "mobileNumber")
    var phoneNumber: String?
    
    @Serialized(default: 0)
    var score: Int
}

Installation

Codoapods

Add this line to your Podfile:

pod 'SerializedSwift'

Swift Package manager

If you are using SPM for your dependency manager, add this to the dependencies in your Package.swift file:

dependencies: [
    .package(url: "https://github.com/dejanskledar/SerializedSwift.git")
]

Requirements

• iOS 10.0+ / macOS 10.12+ / tvOS 10.0+ / watchOS 3.0+
• Xcode 11+
• Swift 5.1+

Usage

class User: Serializable {

    @Serialized
    var name: String
    
    @Serialized("globalId")
    var id: String?
    
    @Serialized(alternateKey: "mobileNumber")
    var phoneNumber: String?
    
    @Serialized(default: 0)
    var score: Int
    
    required init() {}
}

Works with inheritance

No additional decoding logic needed

class PowerUser: User {
    @Serialized
    var powerName: String?

    @Serialized(default: 0)
    var credit: Int
}

Works with composition

No additional decoding logic needed

class ChatRoom: Serializable {
    @Serialized
    var admin: PowerUser?

    @Serialized(default: [])
    var users: [User]
}

Custom transformer classes

You can create own custom Transformable classes, for custom transformation logic.

class DateTransformer: Transformable {
    static func transformFromJSON(from value: String?) -> Date? {
        let formatter = DateFormatter()
        return formatter.date(from: value ?? "")
    }
    
    static func transformToJson(from value: Date?) -> String? {
        let formatter = DateFormatter()
        return formatter.string(from: value ?? Date())
    }
}

struct User: Serializable {
    @SerializedTransformable<DateTransformer>
    var birthDate: Date?
}

Features

Serializable

• typealias-ed from SerializableEncodable & SerializableDecodable
• Custom decoding and encoding using propertyWrappers (listed below)
• Use this protocol for your classes and structures in the combination with the property wrappers below

Serialized

• Serialization propertyWrapper for all properties, optional and non-optionals!
• Custom decoding Key
• By default using the propertyName as a Decoding Key
• Alternative Decoding Key support
• Optional Default value (if the key is missing). By default, the Default value is nil. For non-optionals, default value is recommended, to avoid crashes

@Serialized("mainKey", alternativeKey: "backupKey", default: "")
var key: String?

SerializedTransformable

• Custom transforming property wrapper
• Create own Transformable classes
• Transforming Decodable objects to own types, like custom String Date format to native Date

 class DateTransformer: Transformable {
     static func transformFromJSON(from value: String?) -> Date? {
         let formatter = DateFormatter()
         return formatter.date(from: value ?? "")
     }
     
     static func transformToJson(from value: Date?) -> String? {
         let formatter = DateFormatter()
         return formatter.string(from: value ?? "")
     }
 }

 // Usage of `SerializedTransformable`
 struct User: Serializable {
     @SerializedTransformable<DateTransformer>
     var birthDate: Date?
 } 

Contribute

This is only a tip of the iceberg of what can one achieve using Property Wrappers and how we can improve Decoding and Encoding JSON in Swift. Feel free to colaborate.