I want to call a suspend function within an apply { } block.
I have a:
private suspend fun retrieve(accountAction: AccountAction): Any
suspend fun login() {
accountEvent.apply {
retrieve(it)
}
I tried to surround it with suspend { retrieve(it) } runblocking { retrieve(it) } but it seems that even if it’s not generating an error (Suspension functions can be called only within coroutine body) the code is not getting inside the retrieve function, but just passes through it and that’s why my unit tests fails.
FYI: this is a class, not an activity or a fragment.
Edit:
This is the actual code (from comment):
override suspend fun login(webView: WebView) = trackingId()
.flatMap { id -> AccountAction(client, id, WeakReference(webView), upgradeAccount) }
.map {
it.apply {
upgradeWebViewProgress(webView)
suspend { retrieve(it) }
}
}
.flatMap { updateAuth(it) }
You can use the Flow-API when you want to do asynchronous (suspend) operations on a list of elements like this. You can read about that API here: https://kotlin.github.io/kotlinx.coroutines/kotlinx-coroutines-core/kotlinx.coroutines.flow/-flow/
Probably the simplest way to get your example working is by converting your list to a Flow, performing the suspending operations, then converting back to a List. Like this:
override suspend fun login(webView: WebView) = trackingId()
.flatMap { id -> AccountAction(client, id, WeakReference(webView), upgradeAccount) }
.asFlow()
.map {
it.apply {
upgradeWebViewProgress(webView)
retrieve(it)
}
}
.toList()
.flatMap { updateAuth(it) }
Note that this might not be the most efficient, because it will perform the retrieve-operations sequentially. You can use other operators on Flow to perform the operations in parallel for example.
Edited:
This shows an alternative without using map as it is not really required in my opionion for this example (except you really wanna chain all your calls)
suspend fun login(webView: WebView) {
val result = trackingId().flatMap { id -> AccountAction(client, id, WeakReference(webView), upgradeAccount) }
upgradeWebViewProgress(webView)
return retrieve(result).flatMap { updateAuth(it) } }
Related
I have method, which returns response from server. For example:
fun uploadVideo(link: String, completionHandler: (Result<String>) -> Unit) {
// some action
completionHandler(Result.success(""))
}
I want to call this method one by one. Wait for a response from the previous one to call the next one. For example
uploadVideo("https://stackoverflow.com/video1.mp4") {
}
// call this only when i have response from preview request
uploadVideo("https://stackoverflow.com/video2.mp4") {
}
// call this only when i have response from preview request
uploadVideo("https://stackoverflow.com/video3.mp4") {
}
I tried use suspendCancellableCoroutine, like this
suspend fun uploadVideo(link: String?): String? = suspendCancellableCoroutine { cont ->
uri?.let {
uploadVideo(link,
completionHandler = {
it.onSuccess { uri ->
cont.resume(uri.toString())
}.onFailure {
cont.resumeWithException(it)
}
}
)
} ?: kotlin.run {
cont.resume(null)
}
}
and then call like this:
uploadVideo("https://stackoverflow.com/video1.mp4")
uploadVideo("https://stackoverflow.com/video2.mp4")
uploadVideo("https://stackoverflow.com/video3.mp4")
but these methods are not called sequentially, but in parallel
Note, the contents of your example API function don't quite make sense. If the callback were simply called inside the body of the function, then that would mean the function was blocking the whole time, which would mean there would be no reason for it to even have a callback. It could just directly return the value.
The actual contents of the API function might look more like this:
fun uploadVideo(link: String, completionHandler: (Result<String>) -> Unit) {
val callbackHandler = Handler(Looper.myLooper())
someOtherHandlerOrThreadPool.run {
// some action
callbackHandler.post {
completionHandler(Result.success(""))
}
}
}
The reason I bring that up is that the alternative to nesting a bunch of callbacks is use suspend functions and coroutines, but the code to convert the above to a suspend function doesn't make sense if it were a blocking function like in your version of it.
The basic pattern to convert a callback-based function into a suspend function is to use suspendCoroutine or suspendCancellableCoroutine. If uploadVideo was a function in some api class, you can define it as an extension function:
suspend fun SomeApiClass.uploadVideo(link: String): Result<String> = withContext(Dispatchers.Main) {
suspendCoroutine { cont ->
uploadVideo(link) { cont.resume(it) }
}
}
Now you can call this suspend function repeatedly in sequence if you're inside a coroutine or another suspend function:
fun foo() {
viewModelScope.launch {
val result1 = uploadVideo("https://stackoverflow.com/video1.mp4")
val result2 = uploadVideo("https://stackoverflow.com/video2.mp4")
val result3 = uploadVideo("https://stackoverflow.com/video3.mp4")
}
}
You could try this. this waits till the previous method called its callback and then runs the next one. Only if you have many images this is not a really nice way to do this.
fun uploadVideo(link: String, completionHandler: () -> Unit) {
// some action
completionHandler()
}
uploadVideo("https://stackoverflow.com/video1.mp4") {
uploadVideo("https://stackoverflow.com/video2.mp4") {
uploadVideo("https://stackoverflow.com/video3.mp4") {}
}
}
My aim to call five apis and to get headers from those api response.
I have added my code below
Api service class
#GET("users")
suspend fun getUserList(): Call<List<FriendListModel>>
Repo class
suspend fun getList(): Response<List<FriendListModel>> {
return apiService.getUserList().execute()
}
ViewModel class
fun getFriends() {
viewModelScope.launch(Dispatchers.IO) {
val data =
async {
try {
val data = friendListRepo.getList()
val header = data.headers().get("id")
/*
* need to add header logic
*/
Resource.success(data)
} catch (throwable: Throwable) {
when (throwable) {
is HttpException -> {
Resource.error(false, throwable.response()?.message()?:"")
}
else -> {
Resource.error(false, "")
}
}
}
}
val res = data.await()
mutableFriendsList.postValue(res)
}
}
My question is, am I doing it in right way because I am getting a warning in repo class saying that "Inappropriate blocking method call" since I am calling execute() method though I am calling it in suspend function.
[I referred] Kotlin coroutines await for 2 or more different concurrent requests.
Is there any other approach to achieve this?
You should not combine suspend with Call. Call is for asynchronous work. suspend does asynchronous work synchronously by suspending. It can't be both at once. execute does a blocking synchronous fetch of the data, which shouldn't be done in a coroutine.
So, your functions should look like:
#GET("users")
suspend fun getUserList(): List<FriendListModel>
suspend fun getList(): List<FriendListModel> {
return apiService.getUserList()
}
Then when you use it in a coroutine, you don't need async because you're just calling a synchronous suspend function. You also don't need to fool with Dispatchers.IO since you're only using a suspend function (not doing blocking work). I also simplified your catch block in this example, but that's not related to the solution (I just couldn't help myself).
fun getFriends() {
viewModelScope.launch {
mutableFriendsList.value = try {
val data = friendListRepo.getList()
val header = data.headers().get("id")
/*
* need to add header logic
*/
Resource.success(data)
} catch (throwable: Throwable) {
Resource.error(false, (throwable as? HttpException)?.response()?.message.orEmpty())
}
}
}
Side note, even when you are calling blocking code, you should never need to use async immediately followed by an await() call on it. That is just a convoluted alternative to withContext.
I'm working on a pet project where I'm trying to create a hybrid app using a WebView. The web platform that I run in the WebView sends events to the WebView/App through a #JavascriptInterface object. I can also command the web navigation by running a set of javascript functions against the web platform via the WebView using the evaluateJavascript(String, (String) -> Unit) function.
What I'm trying to achieve right now is that these commands that I execute through the evaluateJavascript(String, (String) -> Unit) function run sequentially. I might execute these commands from many different places at the same time, so I want them to run, wait for the callback from the evaluateJavascript() function to get called, and then execute the next command in the queue.
This is what I have in my custom WebView class:
val scriptQueue = mutableListOf<String>()
fun queueEvaluateJavascript(script: String) {
if (webViewIsLoading) {
scriptQueue.add(script)
} else {
scriptQueue.add(script)
runScriptQueue()
}
}
fun runScriptQueue() {
for (script in scriptQueue) {
evaluateJavascript(script, { })
}
scriptQueue.clear()
}
As you can see this is a super basic approach, and I don't really account for the evaluateJavascript() callback. Ideally, I'd like to find a way to flat map each of this evaluateJavascript() calls so we execute one after another, but waiting for the callback to go through.
With RxJava I think I'd create an Observable and then have the evaluateJavascript() callback trigger the subscriber's onNext(). Since, I'm using Kotlin Coroutines I wanted to do something with Coroutines, so I can queue these evaulateJavascript() calls. But I'm not 100% sure what would be the equivalent here.
That would be a nice problem to approach with coroutines.
The usual way to convert callback based APIs to suspend functions is the following:
suspend fun evaluateJs(script: String) = suspendCoroutine<String> { cont ->
evaluateJavascript(script) { result ->
cont.resume(result)
}
}
You can then use that in combination maybe with a Channel (to serve as a queue) and a coroutine that processes this channel:
class MyWebView(context: Context) : WebView(context) {
private val jsQueue = Channel<String>(BUFFERED)
fun startJsProcessingLoopIn(scope: CoroutineScope) {
scope.launch {
for (script in jsQueue) {
evaluateJs(script)
}
}
}
// you could also make this function non-suspend if necessary by calling
// sendBlocking (or trySend depending on coroutines version)
suspend fun queueEvaluateJavascript(script: String) {
jsQueue.send(script)
}
private suspend fun evaluateJs(script: String) = suspendCoroutine<String> { cont ->
evaluateJavascript(script) { result ->
cont.resume(result)
}
}
}
Alternatively you can create your own coroutine scope and make sure to tie it with some sort of lifecycle of your webview (I'm not familiar with WebView so I'll let you judge which kind of method is correct):
class MyWebView2(context: Context) : WebView(context) {
// you can even further customize the exact thread pool used here
// by providing a particular dispatcher
private val jsProcessingScope = CoroutineScope(CoroutineName("js-processing"))
private val jsQueue = Channel<String>(BUFFERED)
// this starts the loop right away but you can also put this in a method
// to start it at a more appropriate moment
init {
jsProcessingScope.launch {
for (script in jsQueue) {
evaluateJs(script)
}
}
}
// you could also make this function non-suspend if necessary by calling
// sendBlocking (or trySend depending on coroutines version)
suspend fun queueEvaluateJavascript(script: String) {
jsQueue.send(script)
}
private suspend fun evaluateJs(script: String) = suspendCoroutine<String> { cont ->
evaluateJavascript(script) { result ->
cont.resume(result)
}
}
fun someCloseOrDisposeCallback() {
jsProcessingScope.cancel()
}
}
I need to call a suspending function inside a suspendCoroutine block, before I call continuation.resume().
What is the appropriate way of doing that?
private suspend fun someFunction() = suspendCoroutine { cont ->
//...
val myResult = mySuspendingFunction() //<--- The IDE says "Suspension functions can be called only within coroutine body"
cont.resume(myResult)
}
You can't call a suspend function in suspendCoroutine block, because it accepts non suspend block as parameter:
suspend inline fun <T> suspendCoroutine(
crossinline block: (Continuation<T>) -> Unit
): T
'suspendCoroutine' mainly used when we have some legacy code with callbacks, e.g.:
suspend fun getUser(id: String): User = suspendCoroutine { continuation ->
Api.getUser(id) { user ->
continuation.resume(user)
}
}
If function someFunction() doesn't call Api with callbacks then you should reconsider your approach getting rid of 'suspendCoroutine':
private suspend fun someFunction() {
// ...
val myResult = mySuspendingFunction()
// ...
}
If you still want to use suspendCoroutine move call of mySuspendingFunction out of suspendCoroutine block:
private suspend fun someFunction(): String {
val myResult = mySuspendingFunction()
return suspendCoroutine { cont ->
//...
cont.resume(myResult)
}
}
suspend fun mySuspendingFunction(): String {
delay(1000) // simulate request
return "result"
}
It's best to avoid this and call the suspending function before suspendCoroutine, as others have answered. That is possible for the specific case in question.
However, that is not possible if you need the continuation.
(The following is for those, who found this question for the this reason, as #Zordid and I have. chan.send is an example of this.)
In which case, the following is a possible, but error prone way to do it, that I do not recommend:
suspend fun cont1() {
//btw. for correct implementation, this should most likely be at least suspendCancellableCoroutine
suspendCoroutine<Unit> { uCont ->
val x = suspend { chan.send(foo(uCont)) }
x.startCoroutine(Continuation(uCont.context) {
if (it.isFailure)
uCont.resumeWith(it)
// else resumed by whatever reads from chan
})
}
}
(I think the error handling alone illustrates why it's not a great option, despite other problems.)
A better, safer and cheaper way is to use CompletableDeferred if you can.
If you must pass in a Continuation, it's still safer and probably cheaper to do:
suspend fun cont2() {
val rslt = CompletableDeferred<Unit>()
chan.send(foo(Continuation(currentCoroutineContext()) {
rslt.completeWith(it)
}))
rslt.await()
}
I'm trying to create a Flow that needs to emit values from a callback but I can't call the emit function since the SAM is a normal function
Here's the class with the SAM from a library that I can't really modify it the way I need it to be.
class ValueClass {
fun registerListener(listener: Listener) {
...
}
interface Listener {
fun onNewValue(): String
}
}
And here's my take on creating the Flow object
class MyClass(private val valueClass: ValueClass) {
fun listenToValue = flow<String> {
valueClass.registerListener { value ->
emit(value) // Suspension functions can only be called on coroutine body
}
}
}
I guess it would've been simple if I could change the ValueClass but in this case, I can't. I've been wrapping my head around this and trying to look for implementations.
At least from what I know so far, one solution would be to use GlobalScope like this
class MyClass(private val valueClass: ValueClass) {
fun listenToValue = flow<String> {
valueClass.registerListener { value ->
GlobalScope.launch {
emit(value)
}
}
}
}
Now, this works but I don't want to use GlobalScope since I'll be using viewModelScope to tie it to my app's lifecycle.
Is there any way to work around this?
Thanks in advance. Any help would be greatly appreciated!
You can use callbackFlow to create a Flow from the callback. It will look something like:
fun listenToValue(): Flow<String> = callbackFlow {
valueClass.registerListener { value ->
trySend(value)
channel.close() // close channel if no more values are expected
}
awaitClose { /*unregister listener*/ }
}
Or if only one value is expected from the callback, you can use suspendCoroutine or suspendCancellableCoroutine. It this case listenToValue() function must be suspend and later called from a coroutine(e.g. someScope.launch):
suspend fun listenToValue(): String = suspendCoroutine { continuation ->
valueClass.registerListener { value ->
continuation.resumeWith(value)
}
}