LiveData Transformation to StateFlow / SharedFlow

Question

What is the equivalent code for this live data transformation in StateFlow / SharedFlow?

val myLiveData: LiveData<MyLiveData> = Transformations
                    .switchMap(_query) {
                        if (it == null) {
                           AbsentLiveData.create()
                        } else {
                           repository.load()
                     }

Basically, I want to listen to every query changes to react what to return. So, anything similar to that using StateFlow / SharedFlow is welcome.

Answer 1

switchMap is deprecated in flows and should use either of flatMap, transform or transformLatest to convert one type of flows to others. An example for that would be

val myFlow = flowOf<Int>().transform<Int, String> { flowOf("$it") }} 

I guess you can use same logic for StateFlow or SharedFlows.

Answer 2

First, create an helper extension function:

fun <R> Flow<R>.toStateFlow(coroutineScope: CoroutineScope, initialValue: R) = stateIn(coroutineScope, SharingStarted.Lazily, initialValue)

---

Use mapLatest{} for Transformations.map():

val studentNames = _students.mapLatest { students ->
    students.map { "${it.name}" }
}.toStateFlow(uiScope, emptyList())    //uiScope is viewModelScope

Use flatMapLatest{} for Transformations.switchMap():

val asteroids = _asteroidFilter.flatMapLatest { filter ->
    asteroidRepository.getAsteroidsFlow(filter)
}.toStateFlow(uiScope, emptyList())

Use combine() for MediatorLiveData:

val sumScore = combine(_team1Score, _team2Score) { score1, score2 ->
    score1 + score2
}.toStateFlow(uiScope, 0)