Generic Response Object Serialization using Swift 4 Codable
If you don't want to use another dependency like ObjectMapper you can do the following way but you may have to make some chagnes.
Following is a typical model which we use to deserialize JSON data with generics using Alamofire. There is plenty of examples and excellent documentation on Alamofire.
struct User: ResponseObjectSerializable, ResponseCollectionSerializable, CustomStringConvertible {
let username: String
let name: String
var description: String {
return "User: { username: \(username), name: \(name) }"
}
init?(response: HTTPURLResponse, representation: Any) {
guard
let username = response.url?.lastPathComponent,
let representation = representation as? [String: Any],
let name = representation["name"] as? String
else { return nil }
self.username = username
self.name = name
}
}
Using Codable protocol introduced in Swift 4
typealias Codable = Decodable & Encodable
The first step in this direction is to add helper functions that
will do half of the work in deserialization JSON data and handle
errors. Using Swift extensions we add functions to decode incoming
JSON into our model struct/class that we will write afterward.
let decoder = JSONDecoder()
let responseObject = try? decoder.decode(T.self, from: jsonData)
The decoder (1) is an object that decodes instances of a data type from JSON objects.
Helper functions
extension DataRequest{
/// @Returns - DataRequest
/// completionHandler handles JSON Object T
@discardableResult func responseObject<T: Decodable> (
queue: DispatchQueue? = nil ,
completionHandler: @escaping (DataResponse<T>) -> Void ) -> Self{
let responseSerializer = DataResponseSerializer<T> { request, response, data, error in
guard error == nil else {return .failure(BackendError.network(error: error!))}
let result = DataRequest.serializeResponseData(response: response, data: data, error: error)
guard case let .success(jsonData) = result else{
return .failure(BackendError.jsonSerialization(error: result.error!))
}
// (1)- Json Decoder. Decodes the data object into expected type T
// throws error when failes
let decoder = JSONDecoder()
guard let responseObject = try? decoder.decode(T.self, from: jsonData)else{
return .failure(BackendError.objectSerialization(reason: "JSON object could not be serialized \(String(data: jsonData, encoding: .utf8)!)"))
}
return .success(responseObject)
}
return response(queue: queue, responseSerializer: responseSerializer, completionHandler: completionHandler)
}
/// @Returns - DataRequest
/// completionHandler handles JSON Array [T]
@discardableResult func responseCollection<T: Decodable>(
queue: DispatchQueue? = nil, completionHandler: @escaping (DataResponse<[T]>) -> Void
) -> Self{
let responseSerializer = DataResponseSerializer<[T]>{ request, response, data, error in
guard error == nil else {return .failure(BackendError.network(error: error!))}
let result = DataRequest.serializeResponseData(response: response, data: data, error: error)
guard case let .success(jsonData) = result else{
return .failure(BackendError.jsonSerialization(error: result.error!))
}
let decoder = JSONDecoder()
guard let responseArray = try? decoder.decode([T].self, from: jsonData)else{
return .failure(BackendError.objectSerialization(reason: "JSON array could not be serialized \(String(data: jsonData, encoding: .utf8)!)"))
}
return .success(responseArray)
}
return response(responseSerializer: responseSerializer, completionHandler: completionHandler)
}
}
Second, I earlier mentioned “using Swift 4 Codable” but if all we
want is to decode JSON from the server, we only need is a model
struct/class that conforms to protocol Decodable. (If you have the
same structure you want to upload you can use Codable to handle both
decoding and encoding) So, now our User model struct now looks like
this.
struct User: Decodable, CustomStringConvertible {
let username: String
let name: String
/// This is the key part
/// If parameters and variable name differ
/// you can specify custom key for mapping "eg. 'user_name'"
enum CodingKeys: String, CodingKey {
case username = "user_name"
case name
}
var description: String {
return "User: { username: \(username), name: \(name) }"
}
}
Lastly, our function call to API looks like.
Alamofire.request(Router.readUser("mattt"))).responseObject{ (response: DataResponse<User>) in
// Process userResponse, of type DataResponse<User>:
if let user = response.value {
print("User: { username: \(user.username), name: \(user.name) }")
}
}
For more complex (nested) JSON, the logic remains the same and only modifications you need in model struct/class is that all structs/classes must conform to Decodable protocol and Swift takes care of everything else.
