This is my FirebaseOTPVerificationOperation class, where my MutableStateFlow properties are defined, and values are changed,
#ExperimentalCoroutinesApi
class FirebaseOTPVerificationOperation #Inject constructor(
private val activity: Activity,
val logger: Logger
) {
private val _phoneAuthComplete = MutableStateFlow<PhoneAuthCredential?>(null)
val phoneAuthComplete: StateFlow<PhoneAuthCredential?>
get() = _phoneAuthComplete
private val _phoneVerificationFailed = MutableStateFlow<String>("")
val phoneVerificationFailed: StateFlow<String>
get() = _phoneVerificationFailed
private val _phoneCodeSent = MutableStateFlow<Boolean?>(null)
val phoneCodeSent: StateFlow<Boolean?>
get() = _phoneCodeSent
private val _phoneVerificationSuccess = MutableStateFlow<Boolean?>(null)
val phoneVerificationSuccess: StateFlow<Boolean?>
get() = _phoneVerificationSuccess
fun resendPhoneVerificationCode(phoneNumber: String) {
_phoneVerificationFailed.value = "ERROR_RESEND"
}
}
This is my viewmodal, from where i am listening the changes in stateflow properties, as follows,
class OTPVerificationViewModal #AssistedInject constructor(
private val coroutinesDispatcherProvider: AppCoroutineDispatchers,
private val firebasePhoneVerificationListener: FirebaseOTPVerificationOperation,
#Assisted private val savedStateHandle: SavedStateHandle
) : ViewModel() {
#AssistedInject.Factory
interface Factory {
fun create(savedStateHandle: SavedStateHandle): OTPVerificationViewModal
}
val phoneAuthComplete = viewModelScope.launch {
firebasePhoneVerificationListener.phoneAuthComplete.filter {
Log.e("1","filter auth $it")
it.isNotNull()
}.collect {
Log.e("2","complete auth $it")
}
}
val phoneVerificationFailed = viewModelScope.launch {
firebasePhoneVerificationListener.phoneVerificationFailed.filter {
Log.e("3","filter failed $it")
it.isNotEmpty()
}.collect {
Log.e("4","collect failed $it")
}
}
val phoneCodeSent = viewModelScope.launch {
firebasePhoneVerificationListener.phoneCodeSent.filter {
Log.e("5","filter code $it")
it.isNotNull()
}.collect {
Log.e("6","collect code $it")
}
}
val phoneVerificationSuccess = viewModelScope.launch {
firebasePhoneVerificationListener.phoneVerificationSuccess.filter {
Log.e("7","filter success $it")
it.isNotNull()
}.collect {
Log.e("8","collect success $it")
}
}
init {
resendVerificationCode()
secondCall()
}
private fun secondCall() {
viewModelScope.launch(coroutinesDispatcherProvider.io) {
delay(10000)
resendVerificationCode()
}
}
fun resendVerificationCode() {
viewModelScope.launch(coroutinesDispatcherProvider.io) {
firebasePhoneVerificationListener.resendPhoneVerificationCode(
getNumber()
)
}
}
private fun getNumber() =
"+9191111116055"
}
The issue is that
firebasePhoneVerificationListener.phoneVerificationFailed
is fired in viewmodal for first call of,
init {
resendVerificationCode()
}
but for second call of:
init {
secondCall()
}
firebasePhoneVerificationListener.phoneVerificationFailed is not fired in viewmodal, I don't know why it happened, any reason or explanation will be very appericated.
Current Output:
filter auth null
filter failed
filter code null
filter success null
filter failed ERROR_RESEND
collect failed ERROR_RESEND
Expected Output:
filter auth null
filter failed
filter code null
filter success null
filter failed ERROR_RESEND
collect failed ERROR_RESEND
filter failed ERROR_RESEND
collect failed ERROR_RESEND
Pankaj's answer is correct, StateFlow won't emit the same value twice. As the documentation suggests:
Values in state flow are conflated using Any.equals comparison in a similar way to distinctUntilChanged operator. It is used to conflate incoming updates to value in MutableStateFlow and to suppress emission of the values to collectors when new value is equal to the previously emitted one.
Therefore, to resolve this issue you can create a wrapping class and override the equals (and hashCode) method to return false even if the classes are in fact the same:
sealed class VerificationError {
object Resend: VerificationError()
override fun equals(other: Any?): Boolean {
return false
}
override fun hashCode(): Int {
return Random.nextInt()
}
}
StateFlow is SharedFlow:
https://github.com/Kotlin/kotlinx.coroutines/issues/2034
Described in more detail in my article: https://veldan1202.medium.com/kotlin-setup-sharedflow-31debf613b91
val shared = MutableSharedFlow(
replay = 1,
onBufferOverflow = BufferOverflow.DROP_OLDEST,
)
shared.tryEmit(value)
The value emitted by state flow is conflated and doesn't emit the same consecutive result twice, you can think as if a condition check is validating the old emitted value is not equal to the newly emitted value.
Current Output:
filter auth null
filter failed
filter code null
filter success null
filter failed ERROR_RESEND
collect failed ERROR_RESEND
(filter failed ERROR_RESEND
collect failed ERROR_RESEND) This being the same old value which was emitted so you will not see them getting emitted.
Use a Channel: this does emit after sending the same value twice.
Add this to your ViewModel
val _intent = Channel<Intent>(Channel.CONFLATED)
Put values using send / trySend
_intent.send(intentLocal)
observe as flow
_intent.consumeAsFlow().collect { //do something }
I think I have some more in-depth understanding of this issue. The first thing to be sure is that for StateFlow, it is not recommended to use variable collection types (such as MutableList, etc.). Because MutableList is not thread safe. If there are multiple references in the core code, it may cause the program to crash.
Before, the method I used was to wrap the class and override the equals method. However, I think this solution is not the safest method. The safest way is for deep copy, Kotlin provides toMutableList() and toList() methods are both deep copy. The emit method judges whether there is a change depends on whether the result of equals() is equal.
The reason I have this problem is that the data type using emit() is: SparseArray<MutableList>. StateFlow calls the equals method for SparseArray. When MutableList changes, the result of equals does not change at this time (even if the equals and hashcode methods of MutableList change).
Finally, I changed the type to SparseArray<List>. Although the performance loss caused by adding and deleting data, this also solves the problem fundamentally.
As mentioned above, LiveData emits data every time, while StateFlow emits only different values. tryEmit() doesn't work. In my case I found two solutions.
If you have String data, you can emit again this way:
private fun emitNewValue() {
subscriber.value += " "
subscriber.value.dropLast(1)
}
For another class you can use this (or create an extension function):
private fun <T> emitNewValue(value: T) {
if (subscriber.value == value) {
subscriber.value = null
}
subscriber.value = value
}
But it's a bad and buggy way (values are emitted twice additionally).
Try to find all subscribers that change their values. It can be not evident. For instance, focus change listener, Switch (checkbox). When you toggle Switch, a text can also change, so you should subscribe to this listener. The same way when you focus other view, an error text can change.
Use wrapper object with any unique id, for example:
class ViewModel {
private val _listFlow = MutableStateFlow(ListData(emptyList()))
val listFlow: StateFlow<ListData> get() = _listFlow
fun update(list:List<String>){
_listFlow.value = ListData(list)
}
data class ListData constructor(
val list: List<String>,
private val id: UUID = UUID.randomUUID(),//added unique id
)
}
I had a similar problem after merging the streams.
The emit() function will not be executed if == is used to determine equality.
The way to solve the problem: You can wrap a layer and rewrite the hashCode() and equals() methods. The equals() method directly returns false.
This solution works in my code. The stream after the combine has also changed.
Pankaj's answer is correct, StateFlow will not emit the same value twice.
Before wrapping, the result of == is still true even if the content is different.
You could make _phoneVerificationFailed nullable and send null between the two calls!
Related
I have an external interface which I cannot change:
interface ExternalApi {
fun onDataReceived(data: String)
}
I need to start consuming data and send it to flow. Data order is a necessity. I'd like to have a cold flow, but I couldn't find a version of cold flow with emit function, so I used hot flow + replay set to Max value as a workaround. Here was my first try:
class FlowProblem {
val flow: MutableSharedFlow<String> = MutableSharedFlow(replay = Int.MAX_VALUE)
fun startConsuming() {
object : ExternalApi {
override fun onDataReceived(data: String) {
flow.emit(data)
}
}
}
}
Unfortunately it doesn't work as emit function is a suspended function. However this is an external interface and I cannot add suspend modifier. I tried to also do something like this:
override fun onDataReceived(data: String) {
val coroutineScope = CoroutineScope(Job())
coroutineScope.launch {
flow.emit(data)
}
}
but for me it's kind a silly to create new coroutine only in order to move data to flow. I'm also wondering about data order.
What should I do? Maybe flow/channel is not suitable here and I should pick something another?
Thanks IR42, callbackFlow was exactly what I needed.
I'm not super sure what I'm doing here so go easy on me:
I'm making a wordle clone and the word that is to be guessed is stored as a string in a pre-populated room database which I am trying to retrieve to my ViewModel and currently getting:
"StandaloneCoroutine{Active}#933049a"
instead of the actual data.
I have tried using LiveData which only returned null which as far as I'm aware is because it was not observed.
Switched to coroutines which seemed to make more sense if my UI doesn't need the data anyway.
I ended up with this so far:
DAO:
#Dao
interface WordListDao {
#Query("SELECT word FROM wordlist WHERE used = 0 ORDER BY id DESC LIMIT 1")
suspend fun readWord(): String
// tried multiple versions here only string can be converted from Job
// #Query("SELECT * FROM wordlist WHERE used = 0 ORDER BY id DESC LIMIT 1")
// fun readWord(): LiveData<WordList>
// #Query("SELECT word FROM wordlist WHERE used = 0 ORDER BY id DESC LIMIT 1")
// fun readWord(): WordList
}
repository:
class WordRepository(private val wordListDao: WordListDao) {
//val readWordData: String = wordListDao.readWord()
suspend fun readWord(): String {
return wordListDao.readWord()
}
}
model:
#Entity(tableName = "wordlist")
data class WordList(
#PrimaryKey(autoGenerate = true)
val id: Int,
val word: String,
var used: Boolean
)
VM:
class HomeViewModel(application: Application) : ViewModel() {
private val repository: WordRepository
private var word: String
init {
val wordDb = WordListDatabase.getDatabase(application)
val wordDao = wordDb.wordlistDao()
repository = WordRepository(wordDao)
word = viewModelScope.launch {
repository.readWord()
}.toString()
Log.d("TAG", ": $word") // does nothing?
}
println(word) // StandaloneCoroutine{Active}#933049a
}
This is the only way that I have managed to not get the result of:
Cannot access database on the main thread
There is a better way to do this, I just can't figure it out.
You can access the return value of repository.readWord() only inside the launch block.
viewModelScope.launch {
val word = repository.readWord()
Log.d("TAG", ": $word") // Here you will get the correct word
}
If you need to update you UI when this word is fetched from database, you need to use an observable data holder like a LiveData or StateFlow.
class HomeViewModel(application: Application) : ViewModel() {
private val repository: WordRepository
private val _wordFlow = MutableStateFlow("") // A mutable version for use inside ViewModel
val wordFlow = _word.asStateFlow() // An immutable version for outsiders to read this state
init {
val wordDb = WordListDatabase.getDatabase(application)
val wordDao = wordDb.wordlistDao()
repository = WordRepository(wordDao)
viewModelScope.launch {
_wordFlow.value = repository.readWord()
}
}
}
You can collect this Flow in your UI layer,
someCoroutineScope {
viewModel.wordFlow.collect { word ->
// Update UI using this word
}
}
Edit: Since you don't need the word immediately, you can just save the word in a simple global variable for future use, easy.
class HomeViewModel(application: Application) : ViewModel() {
private lateinit var repository: WordRepository
private lateinit var word: String
init {
val wordDb = WordListDatabase.getDatabase(application)
val wordDao = wordDb.wordlistDao()
repository = WordRepository(wordDao)
viewModelScope.launch {
word = repository.readWord()
}
// word is not available here, but you also don't need it here
}
// This is the function which is called when user types a word and presses enter
fun submitGuess(userGuess: String) {
// You can access the `word` here and compare it with `userGuess`
}
}
The database operation will only take a few milliseconds to complete so you can be sure that by the time you actually need that original word, it will have been fetched and stored in the word variable.
(Now that I'm at a computer I can write a bit more.)
The problems with your current code:
You cannot safely read from the database on the main thread synchronously. That's why the suspend keyword would be used in your DAO/repository. Which means, there is no way you can have a non-nullable word property in your ViewModel class that is initialized in an init block.
Coroutines are asychronous. When you call launch, it is queuing up the coroutine to start its work, but the launch function returns a Job, not the result of the coroutine, and your code beneath the launch call continues on the same thread. The code inside the launch call is sent off to the coroutines system to be run and suspend calls will in most cases, as in this case, be switching to background threads back and forth. So when you call toString() on the Job, you are just getting a String representation of the coroutine Job itself, not the result of its work.
Since the coroutine does its work asynchronously, when you try to log the result underneath the launch block, you are logging it before the coroutine has even had a chance to fetch the value yet. So even if you had assigned the result of the coroutine to some String variable, it would still be null by the time you are logging it.
For your database word to be usable outside a coroutine, you need to put it in something like a LiveData or SharedFlow so that other places in code can subscribe to it and do something with the value when it arrives.
SharedFlow is a pretty big topic to learn, so I'll just use LiveData for the below samples.
One way to create a LiveData using your suspend function to retrieve the word is to use the liveData builder function, which returns a LiveData that uses a coroutine under the hood to get the value to publish via the LiveData:
class HomeViewModel(application: Application) : ViewModel() {
private val repository: WordRepository = WordListDatabase.getDatabase(application)
.wordDb.wordlistDao()
.let(::WordRepository)
private val word: LiveData<String> = liveData {
repository.readWord()
}
val someLiveDataForUi: LiveData<Something> = Transformations.map(word) { word ->
// Do something with word and return result. The UI code can
// observe this live data to get the result when it becomes ready.
}
}
To do this in a way that is more similar to your code (just to help with understanding, since this is less concise), you can create a MutableLiveData and publish to the LiveData from your coroutine.
class HomeViewModel(application: Application) : ViewModel() {
private val repository: WordRepository
private val word = MutableLiveData<String>()
init {
val wordDb = WordListDatabase.getDatabase(application)
val wordDao = wordDb.wordlistDao()
repository = WordRepository(wordDao)
viewModelScope.launch {
word.value = repository.readWord()
}
}
val someLiveDataForUi: LiveData<Something> = Transformations.map(word) { word ->
// Do something with word and return result. The UI code can
// observe this live data to get the result when it becomes ready.
}
}
If you're not ready to dive into coroutines yet, you can define your DAO to return a LiveData instead of suspending. It will start reading the item from the database and publish it through the live data once it's ready.
#Dao
interface WordListDao {
#Query("SELECT word FROM wordlist WHERE used = 0 ORDER BY id DESC LIMIT 1")
fun readWord(): LiveData<String>
}
class HomeViewModel(application: Application) : ViewModel() {
private val repository: WordRepository = WordListDatabase.getDatabase(application)
.wordDb.wordlistDao()
.let(::WordRepository)
private val word: LiveData<String> = repository.readWord()
//...
}
The return value is as expected, because launch does always return a Job object representing the background process.
I do not know how you want to use the String for, but all operations which should be done after receiving the String must be moved inside the Coroutine or in a function which is called from the Coroutine.
viewModelScope.launch {
val word = repository.readWord()
// do stuff with word
// switch to MainThread if needed
launch(Dispatchers.Main){}
}
Consider a sealed class State.
sealed class State {
object Unknown : State()
object Loading : State()
object Success : State()
data class Failure(val exception: Exception)
}
I have a stateflow where consumers can actively listen to the state updates.
val state:State = MutableStateFlow(State.Unknown)
Now, I also want to have a simple suspend method which waits till the state reaches either Success or Failure, so consumers who just need the result once need not be aware of the stateflow.
How to achieve this?
Although you already came up with a working solution, you might want to make use of the built-in Flow.first { ... } operator for simplicity.
suspend fun waitForResult(): State {
val resultStates = setOf(State.Success::class, State.Failure::class)
return state.first { it::class in resultStates }
}
I was able to come up with the following extension function which looks to be working fine.
suspend fun waitForResult(): State {
val resultStates = setOf(State.Success::class, State.Failure::class)
return state.waitForStates(resultStates)
}
suspend fun <T : Any> StateFlow<T>.waitForStates(states: Set<KClass<out T>>): T = coroutineScope {
var currentValue = value
// not needed for correctness, just an optimisation
if (currentValue::class in states) {
return currentValue
}
coroutineScope {
collect {
if (it::class in states) {
currentValue = it
cancel()
}
}
}
return currentValue
}
The following code is from the project.
The function of tasksRepository.refreshTasks() is to insert data from remote server to local DB, it's a time consuming operation.
In class TasksViewModel, asksRepository.refreshTasks() is wrapped with viewModelScope.launch{}, it means launch and careless.
1: How can I guarantee tasksRepository.observeTasks().distinctUntilChanged().switchMap { filterTasks(it) } to return the latest result?
2: I don't know how distinctUntilChanged() work, will it keep listening to return the latest result in whole Lifecycle ?
3: What's happened if I use tasksRepository.observeTasks().switchMap { filterTasks(it) } instead of tasksRepository.observeTasks().distinctUntilChanged().switchMap { filterTasks(it) }
Code
class TasksViewModel(..) : ViewModel() {
private val _items: LiveData<List<Task>> = _forceUpdate.switchMap { forceUpdate ->
if (forceUpdate) {
_dataLoading.value = true
viewModelScope.launch {
tasksRepository.refreshTasks()
_dataLoading.value = false
}
}
tasksRepository.observeTasks().distinctUntilChanged().switchMap { filterTasks(it) }
}
...
}
class DefaultTasksRepository(...) : TasksRepository {
override suspend fun refreshTask(taskId: String) {
updateTaskFromRemoteDataSource(taskId)
}
private suspend fun updateTasksFromRemoteDataSource() {
val remoteTasks = tasksRemoteDataSource.getTasks()
if (remoteTasks is Success) {
tasksLocalDataSource.deleteAllTasks()
remoteTasks.data.forEach { task ->
tasksLocalDataSource.saveTask(task)
}
} else if (remoteTasks is Result.Error) {
throw remoteTasks.exception
}
}
override fun observeTasks(): LiveData<Result<List<Task>>> {
return tasksLocalDataSource.observeTasks()
}
}
switchMap - The returned LiveData delegates to the most recent LiveData created by calling switchMapFunction with the most recent value set to source, without changing the reference. Doc
Yes, it'll keep listening to return the latest result in whole Lifecycle. distinctUntilChanged creates a new LiveData object that does not emit a value until the source LiveData value has been changed. The value is considered changed if equals() yields false.
Yes you can use that too but it'll keep emitting the values even the values are the same as the last emitted value.
e.g. first emitted value is ["aman","bansal"] and the second is the same ["aman","bansal"] which you don't want to emit since the values are same. So you use distinctUntilChanged to make sure it won't emit the same value until changed.
I hope this helped.
So I was following the Algorithms book by Sedgewick and trying to translate the implementations from Java to Kotlin and when I tried to implement one Iterator for the Bag data structure (that's essentially a one-way linked list), I got stuck with nullability issues and thread safety in Kotlin.
The implementation in java from the book is done this way:
public class Bag<Item> {
private Node first;
private class Node {
Item item;
Node next;
}
/* some methods */
private class Iterator<Item> {
private Node current = first;
public boolean hasNext() { current != null; }
public Item next() {
if (!hasNext()) throw new NoSuchElementException();
Item item = current.item;
current = current.next;
return item;
}
}
}
which I tried to implement in Kotlin like this:
class Bag<Item> : Iterable<Item> {
private inner class Node(val item: Item, val next: Node?)
private var first : Node? = null
/* Some methods */
override fun iterator() = object : Iterator<Item> {
private var current : Bag<Item>.Node? = first
override fun hasNext() : Boolean = current != null
override fun next() : Item {
if (current == null) throw NoSuchElementException()
val item = current.item
current = current.next
return item
}
}
}
But I got the following error:
Smart cast to 'Bag.Node' is impossible, because 'current' is a mutable property that could have been changed by this time
I understand this is due to race condition between checking if the variable is null and actually accessing the variable attributes, since the variable could be set to null by other threads. After some time I arrived into the following implementation:
override fun iterator() = object : Iterator<Item> {
private var current : Bag<Item>.Node? = first
override fun hasNext() : Boolean = current != null
override fun next() : Item {
current?.let {
val item = it.item
current = it.next
return item
} ?: throw NoSuchElementException()
}
}
And the compiler thought this was fine. But I still have some doubts. Which leads to my questions:
1) Is the assignment current = it.next thread safe or should I assign to the implicit it instead?
2) Is there a idiomatic Kotlin way of implementing an iterator of non-nullable type that ends with a null value? (meaning that all values in it are non-null except for the ending condition)
Is the assignment current = it.next thread safe
It is not thread safe.
Imagine a list of integers and two threads A and B who want to use iterator instance I.
1 -> 2 -> 3 -> 4 -> 5 A: item=1, next=(2)
^ A: item=1, next=(2)
I
Both threads begin to iterate. Both path inside current?.let. Both read current item (val item = it.item) and got item=1, next=(2). Then, first thread A is frozen and second thread B advances the iterator, let's say, by three items forward:
1 -> 2 -> 3 -> 4 -> 5 A: item=1, next=(2)
^ B: item=4, next=(5)
I
Now B enters let and reads next item: item=4, next=(5). Remember that A is still in his loop and it's item is item=1, next=(2). If B is now frozen and A advances one line of code (current = it.next) then things are broken: current is a shared state (as it is stored in the iterator) and, thus, B will see the change as well. On the next iteration in B it will be "thrown back" to the item #2. Nothing bad will happen and the program will not fail, but most probably that is not the behavior you need.
Even more: for the described reasons iterators are not meant to be thread-safe and every thread should have it's own, independent one. Things become more interesting with iterators that change a collection (insertions / deletions), but it's another story as it is about the collection, not about the iterator.
Should I assign to the implicit it instead?
You cannot assign to it, as it is a function parameter and it is passed by value, and, thus, cannot be changed. The compiler will ban the assignment with a message like "Val cannot be reassigned"
Is there a idiomatic Kotlin way of implementing an iterator of non-nullable type that ends with a null value?
I'd say: yes. You could potentially use sealed classes to designate different types of nodes like:
sealed class Node<out T>;
object Empty : Node<Nothing>();
data class Full<T>(val item: T, val next: Node<T>) : Node<T>();
class Bag<T>(private val first: Node<T>) : Iterable<T> {
override fun iterator(): Iterator<T> = object : Iterator<T> {
private var current = first
override fun hasNext() = current !is Empty
override fun next() = when (val c = current) {
Empty -> throw IllegalStateException()
is Full -> {
current = c.next
c.item
}
}
}
}
fun main() {
val bag = Bag(Full(1, Full(2, Full(3, Empty))))
bag.forEach(::println)
}
I was surprised to not see this method listed, which has been available since Kotlin 1.3. You can use the iterator {} helper in conjunction with yield() to easily build an iterator that handles null effectively.
For example, in a LinkedList:
class NamedLinkedListNode(
val name: String,
val next: NamedLinkedListNode?
) : Iterable<String> {
override fun iterator() = iterator {
var node: NamedLinkedListNode? = this#NamedLinkedListNode
while (node != null) {
yield(node.name)
node = node.next
}
}
}