ReactiveState for Kotlin Multiplatform and Android
Easy reactive state management and ViewModels for Kotlin Multiplatform. No boilerplate. Compatible with Android.
ReactiveState-Kotlin provides these foundations for building multiplatform ViewModels and lower-level logic:
See the ReactiveState documentation for more details.
Supported platforms
android, jvm, ios, tvos, watchos, macosArm64, macosX64, mingwX64, linuxX64
Installation
Add the package to your build.gradle
's dependencies {}
:
dependencies {
// Add the BOM using the desired ReactiveState version
api platform("com.ensody.reactivestate:reactivestate-bom:VERSION")
// Leave out the version number from now on:
implementation "com.ensody.reactivestate:reactivestate"
// Utils for unit tests that want to use coroutines
implementation "com.ensody.reactivestate:reactivestate-test"
// Note: kotlin-coroutines-test only supports the "jvm" target,
// so reactivestate-test has the same limitation
}
Also, make sure you've integrated the Maven Central repo, e.g. in your root build.gradle
:
subprojects {
repositories {
// ...
mavenCentral()
// ...
}
}
Quick intro
The following two principles are here to give you a quick idea of the reactive programming aspect only.
The "Guide" section in the documentation describes how to work with the more advanced aspects like multiplatform ViewModels, lifecycle handling, etc.
Note: While the discussion is about StateFlow
, you can also use LiveData
or even implement extensions for other observable values.
Observing StateFlow
Imagine you have an input form with first and last name and want to observe two StateFlow
values at the same time:
isFirstNameValid: StateFlow<Boolean>
isLastNameValid: StateFlow<Boolean>
This is how you'd do it by using the autoRun
function:
autoRun {
submitButton.isEnabled = get(isFirstNameValid) && get(isLastNameValid)
}
With get(isFirstNameValid)
you retrieve isFirstNameValid.value
and at the same time tell autoRun
to re-execute the block whenever the value is changed.
That code is similar to writing this:
lifecycleScope.launchWhenStarted {
isFirstNameValid
.combine(isLastNameValid) { firstNameValid, lastNameValid ->
firstNameValid to lastNameValid
}
.conflate()
.collect { (firstNameValid, lastNameValid) ->
try {
submitButton.isEnabled = firstNameValid && lastNameValid
} catch (e: CancellationException) {
throw e
} catch (e: Throwable) {
onError(e)
}
}
}
Reactive StateFlow / reactive data
The same principle can be used to create a derived
, reactive StateFlow
:
val isFormValid: StateFlow<Boolean> = derived {
get(isFirstNameValid) && get(isLastNameValid)
}
Now you can use autoRun { submitButton.isEnabled = get(isFormValid) }
in the rest of your code.
Going even further, isFirstNameValid
itself would usually also be the result of a derived
computation.
So, you can have multiple layers of reactive derived
StateFlow
s.
Relation to Jetpack Compose / Flutter / React
Reactive UI frameworks like Jetpack Compose automatically rebuild the UI whenever e.g. a StateFlow
changes.
So, in the UI layer autoRun
can usually be replaced with a Composable
.
However, below the UI your data still needs to be reactive, too.
Here ReactiveState provides derived
to automatically recompute a StateFlow
based on other StateFlow
s.
This pattern is very useful in practice and provides the perfect foundation for frameworks like Jetpack Compose which primarily focus on the UI aspect.
ReactiveState's derived
and autoRun
provide the same reactivity for your data and business logic.
In Jetpack Compose you even have derivedStateOf
which is very similar to derived
.
So, you can choose whether you want to build your business logic based on the official coroutines library (StateFlow
/derived
) or Jetpack Compose (State
/derivedStateOf
). However, the coroutines library has the advantage that it's available for more platforms and it's fully independent of any UI frameworks. Finally, most open-source non-UI libraries will probably be based on coroutines, so StateFlow
based code might also be better for compatibility/interoperability.
In other words, the combination of both solutions used together results in a fully reactive, multiplatform codebase - which improves code simplicity and avoids many bugs.
Moreover, Jetpack Compose currently doesn't provide any multiplatform ViewModel support or any large-scale architecture.
So, this library solves that by providing BaseReactiveState
for ViewModels.
It also comes with a lifecycle-aware event system (eventNotifier
) and loading state handling (so you can track one or multiple different loading indicators based on coroutines that you launch).
See also
This library is based on reactive_state for Flutter and adapted to Kotlin Multiplatform and Android patterns.
License
Copyright 2024 Ensody GmbH, Waldemar Kornewald
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
http://www.apache.org/licenses/LICENSE-2.0
Unless required by applicable law or agreed to in writing, software
distributed under the License is distributed on an "AS IS" BASIS,
WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
See the License for the specific language governing permissions and
limitations under the License.