KeyPathMapper

KeyPathMapper is a thin library based on swift #KeyPath feature which facilitates mapping between two different models.

Installation

CocoaPods

pod "KeyPathMapper"

Swift Package Manager

Create a file Package.swift

// swift-tools-version:4.0

import PackageDescription

let package = Package(
    name: "YourProject",
    dependencies: [
        .package(url: "https://github.com/marshallxxx/KeyPathMapper.git", "1.0.0" ..< "2.0.0")
    ],
    targets: [
        .target(name: "YourProject", dependencies: ["KeyPathMapper"])
    ]
)

How to use?

// MARK: - Models

struct TypeA {
    var name: String
    var age: Int
    var submodel: TypeASubmodel
}

struct TypeASubmodel {
    var value: String = ""
}

struct TypeB {
    var name: String
    var age: Int?
    var submodel: TypeBSubmodel
}

struct TypeBSubmodel {
    var value: String = ""
}

// MARK: - Mapper generation

var mapper: KeyPathMapper<TypeA, TypeB> {
        var mapper = KeyPathMapper<TypeA, TypeB>()
        mapper.map(\TypeA.name, to: \TypeB.name)
        mapper.map(\TypeA.age, to: \TypeB.age, fallbackValue: 0)
        
        var submodelMapper = KeyPathMapper<TypeASubmodel, TypeBSubmodel>()
        submodelMapper.map(\TypeASubmodel.value, to: \TypeBSubmodel.value)
        mapper.map(\TypeA.submodel, to: \TypeB.submodel, with: NestedMapperTransformer(keyPathMapper: submodelMapper))
        
        return mapper
    }

// MARK: - Use of mapper

var a = TypeA(name: "TypeA", age: 20, submodel: TypeASubmodel(value: "TypeASubmodel"))
Var b = TypeB(name: "TypeB", age: nil, submodel: TypeBSubmodel(value: "TypeBSubmodel"))
        
let updated: TypeA = mapper.updated(value: a, from: b)
// OR
mapper.update(value: &a, from: b)

KeyPathMapper

KeyPathMapper represent a base class for mapping two models.

Initialisation:

KeyPathMapper<TypeA, TypeB>()

Add mapping:

• If it’s mapping same type then you can avoid transformer part

mapper.map(\TypeA.name, to: \TypeB.name)

• If it’s mapping same type, but one of them is optional library is providing a simplified mapping function. Fallback value is used when you try to map nil to non optional value.

mapper.map(\TypeA.age, to: \TypeB.age, fallbackValue: 0)

• In case you are sure that value will exist at time of mapping you can force unwrap KeyPath

mapper.map(\TypeA.age, to: \TypeB.age!)

• If you have nested subtype, you can use NestedMapperTransformer with their own KeyPathMapper

var submodelMapper = KeyPathMapper<TypeASubmodel, TypeBSubmodel>()
submodelMapper.map(\TypeASubmodel.value, to: \TypeBSubmodel.value)
mapper.map(\TypeA.submodel, to: \TypeB.submodel, with: NestedMapperTransformer(keyPathMapper: submodelMapper))

Conversion:

Sometimes you may want to initialise a model from another one, through KeyPathMapper. You can do that by making your model conform to MappableInitializable, then you can use:

let updated: TypeA = mapper.convert(from: b)

Observing:

You can observe KeyPath change and map it into another model whenever new value is set. This functionality is provided only if observed object is instance of NSObject and observing model is a reference type.

mapper.observe(object: d, mapInto: c)

MapperTransfomer

MapperTransfomer are transforming value on mapping. You can create your own transformations by creating a type which conforms to MapperTransfomer.

Example transforming Int to String and vice versa will look like:

public struct IntStringMapperTransformer: MapperTransfomer {
    public typealias TypeA = Int
    public typealias TypeB = String
    
    public func transform(from previous: TypeB, with value: TypeA) -> TypeB {
        return TypeB(value)
    }
    
    public func transform(from previous: TypeA, with value: TypeB) -> TypeA {
        return TypeA(value)!
    }
}