Is there a shorter way of declaring CodingKeys? - swift

Say you have a struct for a model of your API response. Let's say it has 50 members. However, 5-7 members are non-standard casing, you could have AUsernAme or _BTmember, but the rest are all snake case credit_score or status_code.
Rather than writing all members like this:
struct MyStruct {
let aUserName: String
// +50 more...
private enum CodingKeys: String, CodingKey {
case aUserName = "AUsernAme"
// +50 more...
}
}
Is there a way that we can write it like this?
struct MyStruct {
#CodingKey("AUsernAme") let aUserName: String
let creditScore: Int
// +50 more ...
}
Edit: I guess this is not possible with the current Swift version, but does anyone know if this would somehow be included in the future versions of Swift?

The solution which Sweeper provided is a great solution to your problem, but IMO it may display great complexity to your problem and to the next developers who will read this code.
If I were you, I would just stick to writing all the CodingKeys for simplicity. If your worry is writing a lot of lines of cases, you can write all the cases that doesn't need custom keys in one line and just add the keys with unusual/non-standard casing on new lines:
case property1, property2, property3, property4, property5...
case property50 = "_property50"
And since you mentioned that the rest are in snake case, not sure if you know yet, but we have JSONDecoder.KeyDecodingStrategy.convertFromSnakeCase.
Hope this helps `tol! :)

How about setting a custom keyDecodingStrategy just before you decode instead?
struct AnyCodingKey: CodingKey, Hashable {
var stringValue: String
init(stringValue: String) {
self.stringValue = stringValue
}
var intValue: Int?
init(intValue: Int) {
self.intValue = intValue
self.stringValue = "\(intValue)"
}
}
let mapping = [
"AUsernAme": "aUserName",
// other mappings...
]
let decoder = JSONDecoder()
decoder.keyDecodingStrategy = .custom({ codingPath in
let key = codingPath[0].stringValue
guard let mapped = mapping[key] else { return codingPath.last! }
return AnyCodingKey(stringValue: mapped)
})
This assumes your JSON has a single level flat structure. You can make this into an extension:
extension JSONDecoder.KeyDecodingStrategy {
static func mappingRootKeys(_ dict: [String: String]) -> JSONDecoder.KeyDecodingStrategy {
return .custom { codingPath in
let key = codingPath[0].stringValue
guard let mapped = dict[key] else { return codingPath.last! }
return AnyCodingKey(stringValue: mapped)
}
}
}
let decoder = JSONDecoder()
decoder.keyDecodingStrategy = .mappingRootKeys(mapping)
If your JSON has more levels, you can change the type of the dictionary to [JSONPath: String], where JSONPath is a type that you can create that represents a key in a nested JSON. Then add a bit of code that converts the coding path, which is just an array of coding keys, to JSONPath. This should not be hard to write on your own.
A simple way is to just use [AnyCodingKey] as JSONPath, but there are many other ways too, and I encourage you to experiment and find the one you like the best.
typealias JSONPath = [AnyCodingKey]
extension AnyCodingKey {
init(codingKey: CodingKey) {
if let int = codingKey.intValue {
self.init(intValue: int)
} else {
self.init(stringValue: codingKey.stringValue)
}
}
}
extension JSONDecoder.KeyDecodingStrategy {
static func mappingRootKeys(_ dict: [JSONPath: String]) -> JSONDecoder.KeyDecodingStrategy {
return .custom { codingPath in
guard let mapped = dict[codingPath.map(AnyCodingKey.init(codingKey:))] else { return codingPath.last! }
return AnyCodingKey(stringValue: mapped)
}
}
}
let mapping = [
[AnyCodingKey(stringValue: "AUsernAme")]: "aUserName"
]
It is not possible to use a property wrapper for this. Your property wrapper #CodingKey("AUsernAme") let aUserName: String will be compiled to something like this (as per here):
private var _aUserName: CodingKey<String> = CodingKey("AUsernAme")
var aUserName: String {
get { _aUserName.wrappedValue }
set { _aUserName.wrappedValue = newValue }
}
There are two main problems with this:
Assuming you don't want to write init(from:) for all the 50+ properties in MyStruct, code will be synthesised to decode it, assigning to its _aUserName property. You only have control over the init(from:) initialiser of the CodingKey property wrapper, and you cannot do anything about how MyStruct is decoded in there. If MyStruct is contained in another struct:
struct AnotherStruct: Decodable {
let myStruct: MyStruct
}
Then you can indeed control the coding keys used to decode myStruct by marking it with a property wrapper. You can do whatever you want in the decoding process by implementing the property wrapper's init(from:), which brings us to the second problem:
The coding key you pass to the CodingKey property wrapper is passed via an initialiser of the form init(_ key: String). But you control the decoding via the initialiser init(from decoder: Decoder) because that is what will be called when the struct is decoded. In other words, there is no way for you to send the key mappings to the property wrapper.

Related

Decoding objects without knowing their type first

There is a likelihood this is an XY problem, I am open to these suggestions as well !
I am trying to work with Minecraft save data. Minecraft encodes Entities (basically anything that is not strictly a block) with their type inside an id property . The file then contains a big array of entities, which I want to decode and instantiate.
The problem is that, using Decodable, I must know an object's type before I start instantiating it like container.decode(Zombie.self). I can't figure out how to create a function that would read the id and return the right type of entity ?
I think this explains what I need better than any explanation could :
//Entity objects don't actually store their ID since re-encoding it is trivial.
protocol Entity : Decodable {var someProperty : Int {get set}}
struct Zombie : Entity {var someProperty : Int}
struct Skeleton : Entity {var someProperty : Int}
//Using JSON instead of SNBT so we can use JSONDecoder
let jsonData = """
[
{
"id":"zombie",
"someProperty":"3"
},
{
"id" : "skeleton",
"someProperty":"3"
}
]
"""
struct EntityList : Decodable {
var list : [any Entity] = []
init(from decoder : Decoder) throws {
var container = try decoder.unkeyedContainer()
//What should we put here ?
}
}
let decoder = JSONDecoder()
let entityList = try decoder.decode(EntityList.self, from: Data(jsonData.utf8))
//entityList should be [Zombie, Skeleton]
At the moment I'm looking into the Factory pattern, maybe that's an interesting lead ? In any case, thank you for your help !
( Please note this question has nothing to do with decoding the actual binary contents of the file, it was honestly quite hard to do but I already have a working Encoder / Decoder. It is only about unpacking those contents, hence why I just used JSON in the example above, since we have a common Decoder for that. )
I honestly haven't used the new any syntax enough to know if that can help but I have done what you're trying to do numerous times and here is how I do it.
Set up the data first
We first declare what a Zombie and a Skeleton are. They could just inherit from a protocol or they could be separate structs...
struct Zombie: Decodable {
let someProperty: Int
}
struct Skeleton: Decodable {
let someProperty: Int
let skeletonSpecificProperty: String
}
Then we can turn your array of [anyEntityType] into a homogeneous array by using an enum and embedding the entities into it...
enum Entity: Decodable {
case zombie(Zombie)
case skeleton(Skeleton)
}
Decode the enum given your JSON structure
We have to provide a custom decoder for the Entity type...
init(from decoder: Decoder) throws {
let container = try decoder.container(keyedBy: RootKeys.self)
// First get the `id` value from the JSON object
let type = try container.decode(String.self, forKey: .id)
// check the value for each type of entity we can decode
switch type {
// for each value of `id` create the related type
case "zombie":
let zombie = try Zombie(from: decoder)
self = .zombie(zombie)
case "skeleton":
let skeleton = try Skeleton(from: decoder)
self = .skeleton(skeleton)
default:
// throw an error here... unsupported type or something
}
}
This should now let you decode an array of Entities from JSON into an [Entity] array.
Deal with "unknown" types
There is an extra step required for dealing with the "unknown" types. For instance, in the code above. If the JSON contains "id": "creeper" this will error as it can't deal with that. And you'll end up with your whole array failing to decode.
I've created a couple of helper functions that help with that...
If you create an object like...
struct Minecraft: Decodable {
let entities: [Entity]
enum RootKeys: String, CodingKey {
case entities
}
}
And these helpers...
extension KeyedDecodingContainer {
func decodeAny<T: Decodable>(_ type: T.Type, forKey key: K) throws -> [T] {
var items = try nestedUnkeyedContainer(forKey: key)
var itemsArray: [T] = []
while !items.isAtEnd {
guard let item = try? items.decode(T.self) else {
try items.skip()
continue
}
itemsArray.append(item)
}
return itemsArray
}
}
private struct Empty: Decodable { }
extension UnkeyedDecodingContainer {
mutating func skip() throws {
_ = try decode(Empty.self)
}
}
You can create a custom decoder for the Minecraft type like this...
init(from decoder: Decoder) throws {
let container = try decoder.container(keyedBy: RootKeys.self)
self.entities = try container.decodeAny(Entity.self, forKey: .entities)
}

enum encoded value is nil while storing the class object in UserDefaults. Codable Protocol is already inherited

I am new to iOS and trying to store User object in UserDefaults. So that when the app is launched again, I can check user type and based on it, I need to navigate to relevant screen.
For that, I have created a User class as below (Codable) and it has one userType enum property!
enum UserType: Int, Codable {
case userType1 = 0
case userType2 = 1
case notDetermined = 2
init(from decoder: Decoder) throws {
let label = try decoder.singleValueContainer().decode(Int.self)
self = UserType(rawValue: label) ?? .notDetermined
}
}
class User: Codable {
public var userFullName: String? = ""
public var userType: UserType? //= .notDetermined
enum CodingKeys: String, CodingKey {
case userFullName
}
}
In my view Controller class, I am creating a new instance for User object and trying to store in user defaults as below:
let newUser = User()
newUser.userFullName = "Test"
newUser.userType = userTypeBtn.isSelected ? .userType1 : .userType2
when I print the newUser's userType, I can see proper value whichever is selected. But after that, when I am trying to store it in userDefaults as below, it returns nil for userType property.
do {
let encoded = try JSONEncoder().encode(newValue)
UserDefaults.standard.set(encoded, forKey: UserDefaultKey.currentUser)
UserDefaults.standard.sync()
} catch {
print("Unable to Encode User Object: (\(error))")
}
when I tried to print this encoded variable, and decoded it in console
JSONDecoder().decode(User.self, from: encoded).userType
it prints nil.
Please help me how can I store optional enum property in UserDefaults and retrieve it when needed using Codable
You should include userType in your CodingKeys enum:
enum CodingKeys: String, CodingKey {
case userFullName
case userType
}
Or just delete the CodingKeys enum entirely, since by default, all the properties are included as coding keys. The keys in the CodingKeys enum determines what the synthesised Codable implementation will encode and decode. If you don't include userType, userType will not be encoded, so it will not be stored into UserDefaults.
I am not getting it from Server and userType is an external property outside the JSON response
This is fine, because userType is optional. If the JSON does not have the key, it will be assigned nil. This might be a problem if you are also encoding User and sending it to the server, and that the server can't handle extra keys in the request, in which case you need two structs - one for storing to/loading from UserDefaults, one for parsing/encoding server response/request.
Remember to encode a new User to UserDefaults when you try this out, since the old one still doesn't have the userType encoded with it.
Observations
Having a custom implementation for Decodable part of enum UserType: Int, Codable is probably not the best idea. Swift compiler supports encoding/decoding enum X: Int out of the box without having you to write custom implementation for it. (In fact, starting with Swift 5.5, Swift compiler can now do this for enums that have cases with associated values as well.)
You should try to avoid having cases like .notDetermined. Either user has a type that's well defined or user.type is nil. You can easily define convenience getters on user itself to know about it's type.
Swift allows nesting of types, so having User.Kind instead of UserType is more natural in Swift.
Following implementation takes care of all of these points.
import Foundation
class User: Codable {
enum Kind: Int, Codable {
case free = 1
case pro = 2
}
public var fullName: String?
public var kind: Kind?
}
let newUser = User()
newUser.fullName = "Test"
newUser.kind = .free
do {
let encoded = try JSONEncoder().encode(newUser)
UserDefaults.standard.set(encoded, forKey: "appUser")
if let fetched = UserDefaults.standard.value(forKey: "appUser") as? Data {
let decoded = try JSONDecoder().decode(User.self, from: fetched)
print(decoded)
}
}
Above code includes definition, construction, encodeAndStore, fetchAndDecode and it does everything you need without any custom implementation.
Bonus
Above code does not print a nice description for the User. For that, you can add CustomStringConvertible conformance like this.
extension User: CustomStringConvertible {
var description: String {
"""
fullName: \(fullName ?? "")
kind: \(kind?.description ?? "")
"""
}
}
extension User.Kind: CustomStringConvertible {
var description: String {
switch self {
case .free: return "free"
case .pro: return "pro"
}
}
}
If you try print(decoded) after implementing this, you will clearly see what you want to see for User instance.
User.kind can be nil and I don't want to handle it with if let every time I need to check this from different screens in the app.
No worries, it can be simplified to this.
extension User {
var isFreeUser: Bool { kind == .free }
var isProUser: Bool { kind == .pro }
}

Xcode warning: Immutable property will not be decoded because it is declared with an initial value which cannot be overwritten

Running Xcode 12, my Swift 5 Xcode project now has warnings whenever a Decodable or Codable type declares a let constant with an initial value.
struct ExampleItem: Decodable {
let number: Int = 42 // warning
}
Immutable property will not be decoded because it is declared with an initial value which cannot be overwritten
Xcode suggests changing the let to a var:
Fix: Make the property mutable instead
var number: Int = 42
It also suggests the fix:
Fix: Set the initial value via the initializer or explicitly define a CodingKeys enum including a 'title' case to silence this warning
What is the purpose of this new warning? Should it be heeded, or ignored? Can this type of warning be silenced?
Should Xcode's fix be implemented? Or is there a better solution?
Noah's explanation is correct. It’s a common source of bugs and it's not immediately obvious what’s happening due to the “magical” behaviour of Codable synthesis, which is why I added this warning to the compiler, since it brings your attention to the fact that the property won't be decoded and makes you explicitly call it out if that's the expected behaviour.
As the fix-it explains, you have a couple of options if you want to silence this warning - which one you choose depends on the exact behaviour you want:
Pass the initial value via an init:
struct ExampleItem: Decodable {
let number: Int
init(number: Int = 42) {
self.number = number
}
}
This will allow number to be decoded, but you can also pass around instances of ExampleItem where the default value is used.
You can also use it directly inside init instead, during decoding:
struct ExampleItem: Decodable {
let number: Int
private enum CodingKeys: String, CodingKey {
case number
}
init(from decoder: Decoder) throws {
let container = try decoder.container(keyedBy: CodingKeys.self)
number = try container.decodeIfPresent(Int.self, forKey: .number) ?? 42
}
}
This will allow number to be decoded, but use 42 as the default value if the decoding fails.
Make the property a var, although you can also make it a private(set) var:
struct ExampleItem: Decodable {
var number: Int = 42
}
Making it a var will allow number to be decoded, but it will also allow callers to modify it. By marking it as private(set) var instead, you can disallow this if you want.
Define an explicit CodingKeys enum:
struct ExampleItem: Decodable {
let number: Int = 42
private enum CodingKeys: CodingKey {}
}
This will prevent number from being decoded. Since the enum has no cases, this makes it clear to the compiler that there are no properties that you want to decode.
This warning appears because immutable properties with initial values don't participate in decoding - after all, they're immutable and they have an initial value, which means that initial value will never be changed.
For example, consider this code:
struct Model: Decodable {
let value: String = "1"
}
let json = """
{"value": "2"}
"""
let decoder = JSONDecoder()
let model = try! decoder.decode(Model.self, from: json.data(using: .utf8)!)
print(model)
This will actually print Model(value: "1"), even though the json we gave it had value as "2".
In fact, you don't even need to provide the value in the data you're decoding, since it has an initial value anyway!
let json = """
{}
"""
let decoder = JSONDecoder()
let model = try! decoder.decode(Model.self, from: json.data(using: .utf8)!)
print(model) // prints "Model(value: "1")"
Changing the value to a var means it will decode correctly:
struct VarModel: Decodable {
var value: String = "1"
}
let json = """
{"value": "2"}
"""
let varModel = try! decoder.decode(VarModel.self, from: json.data(using: .utf8)!)
print(varModel) // "VarModel(value: "2")"
If you're seeing this error, it means your code has never correctly parsed the property in question when decoding. If you change it to a var, the property will be parsed correctly, which might be what you want - however, you should make sure that the data you're decoding always has that key set. For example, this will throw an exception (and crash since we're using try!):
let json = """
{}
"""
let decoder = JSONDecoder()
struct VarModel: Decodable {
var value: String = "1"
}
let varModel = try! decoder.decode(VarModel.self, from: json.data(using: .utf8)!)
In conclusion, Xcode's suggestion is probably viable in many cases, but you should evaluate on a case by case basis whether changing the property to a var will break your app's functionality.
If you want the property to always return the hard-coded initial value (which is what's happening right now), consider making it a computed property or a lazy var.
Solution: define an explicit CodingKeys enum to prevent id from decoded.
For example,
struct Course: Identifiable, Decodable {
let id = UUID()
let name: String
private enum CodingKeys: String, CodingKey {
case name
}
init(name: String) { self.name = name }
init(from decoder: Decoder) throws {
let container = try decoder.container(keyedBy: CodingKeys.self)
let name = try container.decodeIfPresent(String.self, forKey: .name)
self.name = name ?? "default-name"
}
}
The suggested workarounds by #SuyashSrijan suppresses the warning but may also lead to further developer errors.
I've written an alternative work around here:
public struct IdentifierWrapper<T>: Identifiable {
public let id = UUID()
public let value: T
}
Usage:
struct Model: Codable, Identifiable {
public let name: String
}
let wrapper = IdentifierWrapper(value: Model(name: "ptrkstr"))

Swift generic problem with array and individual item generic support

I have problem with declaring an Array and initialize it with many different generic items and in the same time use those items individually with generic support. Lets look at this:
protocol Item {
associatedtype ValueType: Any
var value: ValueType? { get set }
}
class ItemClass<T: Any>: Item {
typealias ValueType = T
var value: ValueType?
}
let intItem = ItemClass<Int>()
let stringItem = ItemClass<String>()
let array: [ItemClass<Any>] = [intItem, stringItem]
// Iterate over items and use `value` property as Any?
for item in array {
let val: Any? = item.value
// do something with val
}
// Individual usage with generic types support
let intValue: Int? = intItem.value
let stringValue: String? = stringItem.value
Why there is an error in array declaration like this:
Cannot convert value of type 'ItemClass<Int>' to expected element type 'ItemClass<Any>'
This entire approach is incorrect, and you can see that by the consuming code:
// Iterate over items and use `value` property as Any?
for item in array {
let val: Any? = item.value
// do something with val
}
In that "do something with val," what can you possibly do? There are no methods on Any. If you're going to do something like as? T, then you've broken the whole point of the types, because you don't mean "any". You mean "some list of types I know about." If you want "some list of types I know about," that's an enum with associated data, not a protocol with an associated type.
enum Item {
case string(String)
case int(Int)
var stringValue: String? {
guard case .string(let value) = self else { return nil }
return value
}
var intValue: Int? {
guard case .int(let value) = self else { return nil }
return value
}
}
let intItem = Item.int(4)
let stringItem = Item.string("value")
let array: [Item] = [intItem, stringItem]
// Iterate over items and use `value` property as Any?
for item in array {
switch item {
case let .string(value): break // Do something with string
case let .int(value): break // Do something with int
}
}
// Individual usage with generic types support
let intValue: Int? = intItem.intValue
let stringValue: String? = stringItem.stringValue
If, on the other hand, you really mean "any type," then you're not going to be able to put them in a collection without hiding the values in a box that gets rid of any information about that type (i.e. "a type eraser"). Which you need comes down to your actual use case. There isn't a single answer; it's going to be driven by how you want to consume this data.
But if you need as? very much at all, you've done something wrong.

Reference Types/Subclassing, and Changes to Swift 4 Codable & encoder/decoders

I'm struggling to understand class/reference type behavior and how this relates to changes as I try to upgrade and reduce code using Codable in Swift 4.
I have two classes – a SuperClass with all of the data that will be persistent and that I save to UserDefaults (a place name & string with coordinates), and a SubClass that contains additional, temporary info that I don't need (weather data for the SuperClass coordinates).
In Swift 3 I used to save data like this:
func saveUserDefaults() {
var superClassArray = [SuperClass]()
// subClassArray is of type [SubClass] and contains more data per element.
superClassArray = subClassArray
let superClassData = NSKeyedArchiver.archivedData(withRootObject: superClassArray)
UserDefaults.standard.set(superClassData, forKey: " superClassData")
}
SuperClass conformed to NSObject & NSCoding
It also included the required init decoder & the encode function.
It all worked fine.
In trying to switch to Swift 4 & codable I've modified SuperClass to conform to Codable.
SuperClass now only has one basic initializer and none of the encoder/decoder stuff from Swift 3. There is no KeyedArchiving happening with this new approach (below). SubClass remains unchanged. Unfortunately I crash on the line where I try? encoder.encode [giving a Thread 1: EXC_BAD_ACCESS (code=1, address=0x10)]. My assumption is that the encoder is getting confused with identical reference types where one is SuperClass and one SubClass (subClassArray[0] === superClassArray[0] is true).
I thought this might work:
func saveUserDefaults() {
var superClassArray = [SuperClass]()
superClassArray = subClassArray
// assumption was that the subclass would only contain parts of the superclass & wouldn't produce an error when being encoded
let encoder = JSONEncoder()
if let encoded = try? encoder.encode(superClassArray){
UserDefaults.standard.set(encoded, forKey: " superClassArray ")
} else {
print("Save didn't work!")
}
}
Then, instead of creating an empty superClassArray, then using:
superClassArray = subClassArray, as shown above, I replace this with the single line:
let superClassArray: [SuperClass] = subClassArray.map{SuperClass(name: $0.name, coordinates: $0.coordinates)}
This works. Again, assumption is because I'm passing in the values inside of the class reference type & haven't made the superClassArray = subClassArray. Also, as expected, subClassArray[0] === superClassArray[0] is false
So why did the "old stuff" in Swift 3 work, even though I used the line superClassArray = subClassArray before the let superClassData = NSKeyedArchiver.archivedData(withRootObject: superClassArray)
? Am I essentially achieving the same result by creating the array in Swift 4 that was happening with the old Swift 3 encoder/decoder? Is the looping / recreation
Thanks!
Polymorphic persistence appears to be broken by design.
The bug report SR-5331 quotes the response they got on their Radar.
Unlike the existing NSCoding API (NSKeyedArchiver), the new Swift 4 Codable implementations do not write out type information about encoded types into generated archives, for both flexibility and security. As such, at decode time, the API can only use the concrete type your provide in order to decode the values (in your case, the superclass type).
This is by design — if you need the dynamism required to do this, we recommend that you adopt NSSecureCoding and use NSKeyedArchiver/NSKeyedUnarchiver
I am unimpressed, having thought from all the glowing articles that Codable was the answer to some of my prayers. A parallel set of Codable structs that act as object factories is one workaround I'm considering, to preserve type information.
Update I have written a sample using a single struct that manages recreating polymorphic classes. Available on GitHub.
I was not able to get it to work easily with subclassing. However, classes that conform to a base protocol can apply Codable for default encoding. The repo contains both keyed and unkeyed approaches. The simpler is unkeyed, copied below
// Demo of a polymorphic hierarchy of different classes implementing a protocol
// and still being Codable
// This variant uses unkeyed containers so less data is pushed into the encoded form.
import Foundation
protocol BaseBeast {
func move() -> String
func type() -> Int
var name: String { get }
}
class DumbBeast : BaseBeast, Codable {
static let polyType = 0
func type() -> Int { return DumbBeast.polyType }
var name:String
init(name:String) { self.name = name }
func move() -> String { return "\(name) Sits there looking stupid" }
}
class Flyer : BaseBeast, Codable {
static let polyType = 1
func type() -> Int { return Flyer.polyType }
var name:String
let maxAltitude:Int
init(name:String, maxAltitude:Int) {
self.maxAltitude = maxAltitude
self.name = name
}
func move() -> String { return "\(name) Flies up to \(maxAltitude)"}
}
class Walker : BaseBeast, Codable {
static let polyType = 2
func type() -> Int { return Walker.polyType }
var name:String
let numLegs: Int
let hasTail: Bool
init(name:String, legs:Int=4, hasTail:Bool=true) {
self.numLegs = legs
self.hasTail = hasTail
self.name = name
}
func move() -> String {
if numLegs == 0 {
return "\(name) Wriggles on its belly"
}
let maybeWaggle = hasTail ? "wagging its tail" : ""
return "\(name) Runs on \(numLegs) legs \(maybeWaggle)"
}
}
// Uses an explicit index we decode first, to select factory function used to decode polymorphic type
// This is in contrast to the current "traditional" method where decoding is attempted and fails for each type
// This pattern of "leading type code" can be used in more general encoding situations, not just with Codable
//: **WARNING** there is one vulnerable practice here - we rely on the BaseBeast types having a typeCode which
//: is a valid index into the arrays `encoders` and `factories`
struct CodableRef : Codable {
let refTo:BaseBeast //In C++ would use an operator to transparently cast CodableRef to BaseBeast
typealias EncContainer = UnkeyedEncodingContainer
typealias DecContainer = UnkeyedDecodingContainer
typealias BeastEnc = (inout EncContainer, BaseBeast) throws -> ()
typealias BeastDec = (inout DecContainer) throws -> BaseBeast
static var encoders:[BeastEnc] = [
{(e, b) in try e.encode(b as! DumbBeast)},
{(e, b) in try e.encode(b as! Flyer)},
{(e, b) in try e.encode(b as! Walker)}
]
static var factories:[BeastDec] = [
{(d) in try d.decode(DumbBeast.self)},
{(d) in try d.decode(Flyer.self)},
{(d) in try d.decode(Walker.self)}
]
init(refTo:BaseBeast) {
self.refTo = refTo
}
init(from decoder: Decoder) throws {
var container = try decoder.unkeyedContainer()
let typeCode = try container.decode(Int.self)
self.refTo = try CodableRef.factories[typeCode](&container)
}
func encode(to encoder: Encoder) throws {
var container = encoder.unkeyedContainer()
let typeCode = self.refTo.type()
try container.encode(typeCode)
try CodableRef.encoders[typeCode](&container, refTo)
}
}
struct Zoo : Codable {
var creatures = [CodableRef]()
init(creatures:[BaseBeast]) {
self.creatures = creatures.map {CodableRef(refTo:$0)}
}
func dump() {
creatures.forEach { print($0.refTo.move()) }
}
}
//: ---- Demo of encoding and decoding working ----
let startZoo = Zoo(creatures: [
DumbBeast(name:"Rock"),
Flyer(name:"Kookaburra", maxAltitude:5000),
Walker(name:"Snake", legs:0),
Walker(name:"Doggie", legs:4),
Walker(name:"Geek", legs:2, hasTail:false)
])
startZoo.dump()
print("---------\ntesting JSON\n")
let encoder = JSONEncoder()
encoder.outputFormatting = .prettyPrinted
let encData = try encoder.encode(startZoo)
print(String(data:encData, encoding:.utf8)!)
let decodedZoo = try JSONDecoder().decode(Zoo.self, from: encData)
print ("\n------------\nAfter decoding")
decodedZoo.dump()
Update 2020-04 experience
This approach continues to be more flexible and superior to using Codable, at the cost of a bit more programmer time. It is used very heavily in the Touchgram app which provides rich, interactive documents inside iMessage.
There, I need to encode multiple polymorphic hierarchies, including different Sensors and Actions. By storing signatures of decoders, it not only provides with subclassing but also allows me to keep older decoders in the code base so old messages are still compatible.