Network service layer Combine REST API CRUD

Easy and lightweight network layer for creating different set of network requests like GET, POST, PUT, DELETE customizable with coders conforming to TopLevelDecoder, TopLevelEncoder

Features

• Stand alone package without any dependencies using just Apple's Combine facilities
• Error handling from forming URLRequest object to gettting data on a subscription
• Customizable for different environments development, production...
• Customizable for different requests schemes from classic CRUD Rest to what suits to your own custom routes
• Based on interfaces not implementations
• Ability to log and customize logers for different environment
• Ability to chain requests layering with predicate logic analyzing previous result
• Do in parallel infinite amount of requests collecting their results in Array and waiting until they are all done

1. Environment

Define enum with interface IEnvironment

enum Environment: IEnvironment {

    case development

    case production

    var baseURL: String {
        switch self {
            case .development: return "http://localhost:3000"
            case .production: return "https://apple.com"
        }
    }
    
    var headers: [IRequestHeader]? {
        switch self {
            case .development: return [ContentType.applicationJSON]
            case .production: return [ContentType.textJSON]
        }
    }
    
    var logger : ILogger? {
        switch self {
            case .development: return ServiceLogger()
            case .production: return nil
        }
    }  
}

Request headers

All headers for a request have to conform to the interface IRequestHeader

The example implemetation for content type headers is here ContentType.swift

Logger

You can use out of the box standard logger ServiceLogger if you don't need some specific data from URLRequest and URLResponse or define your own with interface ILogger

2. API for endpoint

Define endpoint API enum

enum UserRestAPI {

    case index
    case read(id: Int)
    case create
    case update
    case delete(id: Int)
}

Extend the enum with interface IRequest

extension UserRestAPI: IRequest {
    
    var route: String {
        switch self {
            case .index: return "/user"
            case .read(let id): return "/user/\(id)"
            case .create: return "/user"
            case .update: return "/user"
            case .delete(let id): return "/user/\(id)"
        }
    }
    
    var method: RequestMethod {
        switch self {
            case .index: return .get
            case .read(_): return .get
            case .create: return .post
            case .update: return .put
            case .delete(_): return .delete
        }
    }
}

The example implemetation is here UserRestAPI.swift UserRestAPI.swift

3. Create network sevice

    let network = NetworkService(environment: Environment.development)

• execute - Do request from the endpont configuration GET, POST, PUT, DELETE

There are four methods are available currently GET, POST, PUT, DELETE

Parameters

Pass a [String: CustomStringConvertible] dictionary to the parameter that avalable for GET, POST, PUT requests. It's an optinal parameter.

Read

   let cfg = UserRestAPI.read(id: 1)
   
   let publisher: Output = network.execute(with: cfg, ["token" : 65678])

Create

    let cfg = UserRestAPI.create
    let user = Model(id: 11, name: "Igor")

    let publisher: Output = network.execute(body: user, with: cfg)

Update

    let cfg = UserRestAPI.update
    let user = Model(id: 11, name: "Igor")    

    let publisher: Output = network.execute(body: user, with: cfg)

Delete

    let cfg = UserRestAPI.delete(id: 11)
    
    let publisher: Output = network.execute(with: cfg)

4. Managing requests

Chaining requests

        let read: Output = network.execute(with: UserRestAPI.index)

        let user = Model(id: 11, name: "Igor")
        let create: Output = network.execute(body: user, with: UserRestAPI.create)
        
        read.then(create)
            
        // or chain it using predicate to analyze previous result
        
        read.then(ifTrue : {$0.count > 1}, create)

Do in parallel infinite amount of requests

Collecting their results in Array and waiting until they are all done

All requests are expected the same output and failure

       [read, create, delete, update, read, read ].doTogether

Mixing

First create and update requests as a group and then read

        [create, update]
                       .waitEverybody
                       .then(read)
                      
        [create, delete]
                        .waitEverybody
                        .then([delete, read].doTogether)               

Package installation

In Xcode - Select Xcode>File> Swift Packages>Add Package Dependency...
and add https://github.com/The-Igor/d3-network-service

Try it in the real environment

Simple server instalation (mocking with NodeJS Express)

To try it in the real environment. I suggest installing the basic NodeJS Express boilerplate. Take a look on the video snippet how easy it is to get it through Webstorm that is accessible for free for a trial period.

• Get WebStorm Early Access
• Get index.js file from here and replace it with the one in the boilerplate and luanch the server.

Real SwiftUI example

d3-rest-combine-swift-example

• Run server NodeJS Express
• Run SwiftUI project

Documentation(API)

• You need to have Xcode 13 installed in order to have access to Documentation Compiler (DocC)
• Go to Product > Build Documentation or ⌃⇧⌘ D