Swiftpack.co - inDriver/UDF as Swift Package

Swiftpack.co is a collection of thousands of indexed Swift packages. Search packages.
See all packages published by inDriver.
inDriver/UDF 2.5.7
Unidirectional Data Flow Framework
⭐️ 55
🕓 2 weeks ago
.package(url: "https://github.com/inDriver/UDF.git", from: "2.5.7")

Swift Version Swift Package Manager compatible



UDF (Unidirectional Data Flow) is a library based on Unidirectional Data Flow pattern. It lets you build maintainable, testable, and scalable apps.

Unidirectional Data Flow Design Pattern

A unidirectional data flow is a design pattern where a state (data) flows down and events (actions) flow up. It's important that UI never edits or sends back data. That's why UI is usually provided with immutable data. It allows having a single source of truth for a whole app and effectively separates domain logic from UI.


Unidirectional Data Flow design pattern has been popular for a long time in the web development. Now it's time for the mobile development. Having started from multi-platform solutions like React Native and Flutter, Unidirectional Data Flow now becomes a part of native. SwiftUI for Swift and Jetpack Compose for Kotlin are implemented based on ideas of UDF. That's why we in inDriver decided to develop our own UDF library for our purposes.


Here are the main advantages of this UDF implementation:

  • Testable. All domain logic implements in pure functions, so it's easy to unit test it. All UI depends only on provided data, so it's easy to configure and cover by snapshot tests.
  • Scalable and Reusable. Low Coupling and High Cohesion are examples of the basic principles of good software design. UDF implements such principles in practice. It allows to decouple UI and Domain, create reusable features, and scale business logic in a convenient way.
  • Easy working with concurrency. The UDF obviously doesn't solve all potential concurrent problems. But it alleviates working with concurrency in regular cases. State updating always runs on separate serial queue. It guarantees the consistency of a state after any changes. For UI or an async task you can use ViewComponent or ServiceComponent protocols respectively. They will subscribe your components on main or background thread so you can concentrate on business tasks rather than concurrency.
  • Free of FRP frameworks. We decided not to use functional reactive frameworks in our library. Instead, we provided it with a convenient way for subscribing to state updates, so in most cases, you don't even need to know how it works. The absence of FRP frameworks also means that you can't use the UDF with SwiftUI right now. But We plan to add Combine version of the UDF in near future. It will only affect the subscription process, so you will not have to rewrite your domain logic.

Differences from other popular UDF implementations:

RxFeedback - requires RxSwift

The Composable Architecture - iOS13+ only because of Combine

ReSwift - no instruments for modularization

Basic Usage

Let's imagine a simple counter app. It shows a counter label and two buttons "+" and "-" to increment and decrement the counter. Let's consider the stages of its creation.

Building Domain

Firstly we need to declare the state of the app:

struct AppState: Equatable {
   var counter = 0

State is all the data of an app. In our case, it's just an int counter. Next, we need to know when buttons are tapped:

enum CounterAction: Action {
 case decrementButtonTapped
 case incrementButtonTapped

We use an enum and an Action protocol for that. Action describes all of the actions that can occur in your app. Next, we need to update our state according to an action:

func counterReducer(state: inout AppState, action: Action) {
   guard let action = action as? CounterAction else { return }

   switch action {
   case .decrementButtonTapped:
       state.counter -= 1
   case .incrementButtonTapped:
       state.counter += 1

Reducer is a pure function that updates a state. Now we need to glue it together:

let store = Store<AppState>(state: .init(), reducer: counterReducer)

Store combines all the above things together.

View Component

Let's take a look at UI now:

class CounterViewController: UIViewController, ViewComponent {

  typealias Props = Int

   @IBOutlet var counterLabel: UILabel!
   var disposer = Disposer()

   var props: Int = 0 {
       didSet {
           guard isViewLoaded else { return }
           counterLabel.text = "\(props)"
   @IBAction func decreaseButtonTapped(_ sender: UIButton) {

   @IBAction func increaseButtonTapped(_ sender: UIButton) {

CounterViewController implements ViewComponent protocol. It guarantees that a component receives a new state only if the state was changed and this process always occurs in the main thread. In CounterViewController we declare props property and update UI in its didSet. Now we have to connect our ViewController to the store:

let counterViewController = CounterViewController()
counterViewController.connect(to: store, state: \.counter)

Notice that we can choose witch part of the state we want to observe.


Imagine that you would like to reuse your CounterViewController in another app. Or you have a much bigger reusable feature with many View Controllers. In this case, your AppState will look like this:

struct AppState: Equatable {
  var counter = 0
  var bigFeature = BigFeature()

Obviously you don't want your features to know about AppState. You can easily decouple them by scope:

let store = Store<AppState>(state: .init(), reducer: counterReducer)

connectCounter(to: store.scope(\.counter))
connectBigFeature(to: store.scope(\.bigFeature))


//Somewhere in Counter.framework
func connectCounter(to store: Store<Int>) {
  counterViewController.connect(to: store)

//Somewhere in BigFeature.framework
func connectCounter(to store: Store<BigFeature>) {
 firstViewController.connect(to: store, state: \.first)

Now you can move your features to separate frameworks and use them wherever you want.


You can add the UDF to an Xcode project by adding it as a package dependency.

  1. Open File › Swift Packages › Add Package Dependency…
  2. Enter https://github.com/inDriver/UDF
  3. Choose the last version

Inspiration & Acknowledgments

The UDF idea is based on Alexey Demedetskiy ideas and MaximBazarov implementation of Unidirection Data Flow Pattern. Originally the UDF was a fork of Unicore.

The Composable Architecture inspired our implementation of a scope function and modularisation.

Also, we would like to thank all people that took part in development, testing, and using the UDF: Artem Lytkin, Ivan Dyagilev, Andrey Zamorshchikov, Dmitry Filippov, Eldar Adelshin, Anton Nochnoy, Denis Sidorenko, Ilya Kuznetsov and Viktor Gordienko.

If you have any questions or suggestions, please do not hesitate to contact Anton Goncharov or Yuri Trykov.


UDF is released under the Apache 2.0 license. See LICENSE for details.

Copyright 2021  Suol Innovations Ltd.

Licensed under the Apache License, Version 2.0 (the "License");
you may not use this file except in compliance with the License.
You may obtain a copy of the License at


Unless required by applicable law or agreed to in writing, software
distributed under the License is distributed on an "AS IS" BASIS,
See the License for the specific language governing permissions and
limitations under the License.


Stars: 55
Last commit: 1 week ago
jonrohan Something's broken? Yell at me @ptrpavlik. Praise and feedback (and money) is also welcome.

Release Notes

7 weeks ago

minor bug fixes

Swiftpack is being maintained by Petr Pavlik | @ptrpavlik | @swiftpackco | API | Analytics