ObjectMapper is a framework written in Swift that makes it easy for you to convert your model objects (classes and structs) to and from JSON.
- Features
- The Basics
- Mapping Nested Objects
- Custom Transformations
- Subclassing
- Generic Objects
- Mapping Context
- ObjectMapper + Alamofire
- ObjectMapper + Realm
- To Do
- Contributing
- Installation
- Mapping JSON to objects
- Mapping objects to JSON
- Nested Objects (stand alone, in arrays or in dictionaries)
- Custom transformations during mapping
- Struct support
To support mapping, a class or struct just needs to implement the Mappable protocol.
public protocol Mappable {
init?(_ map: Map)
mutating func mapping(map: Map)
static func objectForMapping(map: Map) -> Mappable? // Optional
}ObjectMapper uses the <- operator to define how each member variable maps to and from JSON.
class User: Mappable {
var username: String?
var age: Int?
var weight: Double!
var array: [AnyObject]?
var dictionary: [String : AnyObject] = [:]
var bestFriend: User? // Nested User object
var friends: [User]? // Array of Users
var birthday: NSDate?
required init?(_ map: Map) {
}
// Mappable
func mapping(map: Map) {
username <- map["username"]
age <- map["age"]
weight <- map["weight"]
array <- map["arr"]
dictionary <- map["dict"]
bestFriend <- map["best_friend"]
friends <- map["friends"]
birthday <- (map["birthday"], DateTransform())
}
}
struct Temperature: Mappable {
var celsius: Double?
var fahrenheit: Double?
init?(_ map: Map) {
}
mutating func mapping(map: Map) {
celsius <- map["celsius"]
fahrenheit <- map["fahrenheit"]
}
}Once your class implements Mappable, the Mapper class handles everything else for you:
Convert a JSON string to a model object:
let user = Mapper<User>().map(JSONString)Convert a model object to a JSON string:
let JSONString = Mapper().toJSONString(user, prettyPrint: true)ObjectMapper can map classes composed of the following types:
IntBoolDoubleFloatStringRawRepresentable(Enums)Array<AnyObject>Dictionary<String, AnyObject>Object<T: Mappable>Array<T: Mappable>Array<Array<T: Mappable>>Set<T: Mappable>Dictionary<String, T: Mappable>Dictionary<String, Array<T: Mappable>>- Optionals of all the above
- Implicitly Unwrapped Optionals of the above
This failable initializer can be used for JSON validation prior to object serialization. Returning nil within the function will prevent the mapping from occuring. You can inspect the JSON stored within the Map object to do your validation:
required init?(_ map: Map){
// check if a required "name" property exists within the JSON.
if map.JSONDictionary["name"] == nil {
return nil
}
}This function is where all mapping definitions should go. When parsing JSON, it is executed after a successful object initialization. When generating JSON, it is the only function that is called on the object.
This is a sub protocol of Mappable that provides an extra static function that can be used instead of init?(_ map: Map)
If this function is implemented, init?(_ map: Map) will no longer be called by ObjectMapper. This function should be used to:
- provide an existing cached object to be used for mapping
- return an object of another type (which also conforms to Mappable) to be used for mapping. For instance, you may inspect the JSON to infer the type of object that should be used for mapping (see example)
ObjectMapper supports dot notation within keys for easy mapping of nested objects. Given the following JSON String:
"distance" : {
"text" : "102 ft",
"value" : 31
}You can access the nested objects as follows:
func mapping(map: Map) {
distance <- map["distance.value"]
}Nested keys also support accessing values from an array. Given a JSON response with an array of distances, the value could be accessed as follows:
distance <- map["distances.0.value"]
If you have a key that contains ., you can individually disable the above feature as follows:
func mapping(map: Map) {
identifier <- map["app.identifier", nested: false]
}ObjectMapper also supports custom transforms that convert values during the mapping process. To use a transform, simply create a tuple with map["field_name"] and the transform of your choice on the right side of the <- operator:
birthday <- (map["birthday"], DateTransform())The above transform will convert the JSON Int value to an NSDate when reading JSON and will convert the NSDate to an Int when converting objects to JSON.
You can easily create your own custom transforms by adopting and implementing the methods in the TransformType protocol:
public protocol TransformType {
typealias Object
typealias JSON
func transformFromJSON(value: AnyObject?) -> Object?
func transformToJSON(value: Object?) -> JSON?
}In a lot of situations you can use the built-in transform class TransformOf to quickly perform a desired transformation. TransformOf is initialized with two types and two closures. The types define what the transform is converting to and from and the closures perform the actual transformation.
For example, if you want to transform a JSON String value to an Int you could use TransformOf as follows:
let transform = TransformOf<Int, String>(fromJSON: { (value: String?) -> Int? in
// transform value from String? to Int?
return Int(value!)
}, toJSON: { (value: Int?) -> String? in
// transform value from Int? to String?
if let value = value {
return String(value)
}
return nil
})
id <- (map["id"], transform)Here is a more condensed version of the above:
id <- (map["id"], TransformOf<Int, String>(fromJSON: { Int($0!) }, toJSON: { $0.map { String($0) } }))Classes that implement the Mappable protocol can easily be subclassed. When subclassing mappable classes, follow the structure below:
class Base: Mappable {
var base: String?
required init?(_ map: Map) {
}
func mapping(map: Map) {
base <- map["base"]
}
}
class Subclass: Base {
var sub: String?
required init?(_ map: Map) {
super.init(map)
}
override func mapping(map: Map) {
super.mapping(map)
sub <- map["sub"]
}
}Make sure your subclass implemenation calls the right initializers and mapping functions to also apply the mappings from your superclass.
ObjectMapper can handle classes with generic types as long as the generic type also conforms to Mappable. See the following example:
class Result<T: Mappable>: Mappable {
var result: T?
required init?(_ map: Map){
}
func mapping(map: Map) {
result <- map["result"]
}
}
let result = Mapper<Result<User>>().map(JSON)The Map object which is passed around during mapping, has an optional MapContext object that is available for developers to use if they need to pass information around during mapping.
To take advantage of this feature, simple create an object that implements MapContext (which is an empty protocol) and pass it into Mapper during initialization.
struct Context: MapContext {
var importantMappingInfo = "Info that I need during mapping"
}
class User: Mappable {
var name: String?
required init?(_ map: Map){
}
func mapping(map: Map){
if let context = map.context as? Context {
// use context to make decisions about mapping
}
}
}
let context = Context()
let user = Mapper<User>(context: context).map(JSONString)#ObjectMapper + Alamofire
If you are using Alamofire for networking and you want to convert your responses to Swift objects, you can use AlamofireObjectMapper. It is a simple Alamofire extension that uses ObjectMapper to automatically map JSON response data to Swift objects.
#ObjectMapper + Realm
ObjectMapper and Realm can be used together. Simply follow the class structure below and you will be able to use ObjectMapper to generate your Realm models:
class Model: Object, Mappable {
dynamic var name = ""
required convenience init?(_ map: Map) {
self.init()
}
func mapping(map: Map) {
name <- map["name"]
}
}Note: Generating a JSON string of a Realm Object using ObjectMappers' toJSON function only works within a Realm write transaction. This is caused because ObjectMapper uses the inout flag in its mapping functions (<-) which are used both for serializing and deserializing. Realm detects the flag and forces the toJSON function to be called within a write block even though the objects are not being modified.
- Improve error handling. Perhaps using
throws - Class cluster documentation
- Realm List Transform
Contributions are very welcome 👍😃.
Before submitting any pull request, please ensure you have run the included tests and they have passed. If you are including new functionality, please write test cases for it as well.
ObjectMapper can be added to your project using CocoaPods 0.36 or later by adding the following line to your Podfile:
pod 'ObjectMapper', '~> 1.3'If you're using Carthage you can add a dependency on ObjectMapper by adding it to your Cartfile:
github "Hearst-DD/ObjectMapper" ~> 1.3
Otherwise, ObjectMapper can be added as a submodule:
- Add ObjectMapper as a submodule by opening the terminal,
cd-ing into your top-level project directory, and entering the commandgit submodule add https://github.com/Hearst-DD/ObjectMapper.git - Open the
ObjectMapperfolder, and dragObjectMapper.xcodeprojinto the file navigator of your app project. - In Xcode, navigate to the target configuration window by clicking on the blue project icon, and selecting the application target under the "Targets" heading in the sidebar.
- Ensure that the deployment target of
ObjectMapper.frameworkmatches that of the application target. - In the tab bar at the top of that window, open the "Build Phases" panel.
- Expand the "Target Dependencies" group, and add
ObjectMapper.framework. - Click on the
+button at the top left of the panel and select "New Copy Files Phase". Rename this new phase to "Copy Frameworks", set the "Destination" to "Frameworks", and addObjectMapper.framework.