The "proper" way to update views with Android seems to be LiveData. But I can't determine the "proper" way to connect that to a model. Most of the documentation I have seen shows connecting to Room which returns a LiveData object. But (assuming I am not using Room), returning a LiveData object (which is "lifecycle aware", so specific to the activity/view framework of Android) in my model seems to me to violate the separation of concerns?
Here is an example with Activity...
class MainActivity: AppCompatActivity() {
override fun onCreate(savedInstanceState: Bundle?) {
super.onCreate(savedInstanceState);
setContentView(R.layout.activity_main_activity);
val viewModel = ViewModelProvider(this).get(UserViewModel::class.java)
val nameText = findViewById<TextView>(R.id.nameTextBox)
viewModel.getName().observe(this, { name ->
nameText.value = name
})
}
}
And ViewModel...
class UserViewModel(): ViewModel() {
private val name: MutableLiveData<String> = MutableLiveData()
fun getName() : LiveData<String> {
return name
}
}
But how do I then connect that to my Model without putting a "lifecycle aware" object that is designed for a specific framework in my model (LiveData)...
class UserModel {
val uid
var name
fun queryUserInfo() {
/* API query here ... */
val request = JSONObjectRequest( ...
{ response ->
if( response.name != this.name ) {
this.name = response.name
/* Trigger LiveData update here somehow??? */
}
}
)
}
}
I am thinking I can maybe put an Observable object in my model and then use that to trigger the update of the LiveData in my ViewModel. But don't find any places where anyone else says that is the "right" way of doing it. Or, can I instantiate the LiveData object in the ViewModel from an Observable object in my model?
Or am I just thinking about this wrong or am I missing something?
This is from official documentation. Check comments in code...
UserModel should remain clean
class UserModel {
private val name: String,
private val lastName: String
}
Create repository to catch data from network
class UserRepository {
private val webservice: Webservice = TODO()
fun getUser(userId: String): LiveData<UserModel > {
val data = MutableLiveData<UserModel>() //Livedata that you observe
//you can get the data from api as you want, but it is important that you
//update the LiveDate that you will observe from the ViewModel
//and the same principle is in the relation ViewModel <=> Fragment
webservice.getUser(userId).enqueue(object : Callback<UserModel > {
override fun onResponse(call: Call<User>, response: Response<UserModel >) {
data.value = response.body()
}
// Error case is left out for brevity.
override fun onFailure(call: Call<UserModel >, t: Throwable) {
TODO()
}
})
return data //you will observe this from ViewModel
}
}
The following picture should explain to you what everything looks like
For more details check this:
https://developer.android.com/jetpack/guide
viewmodels-and-livedata-patterns-antipatterns
Related
I have a favorite way of doing network request on Android (using Retrofit). It looks like this:
// NetworkApi.kt
interface NetworkApi {
#GET("users")
suspend fun getUsers(): List<User>
}
And in my ViewModel:
// MyViewModel.kt
class MyViewModel(private val networkApi: NetworkApi): ViewModel() {
val usersLiveData = flow {
emit(networkApi.getUsers())
}.asLiveData()
}
Finally, in my Activity/Fragment:
//MyActivity.kt
class MyActivity: AppCompatActivity() {
private viewModel: MyViewModel by viewModels()
override fun onCreate(savedInstanceState: Bundle?) {
super.onCreate(savedInstanceState)
setContentView(R.layout.activity_main)
viewModel.usersLiveData.observe(this) {
// Update the UI here
}
}
}
The reason I like this way is because it natively works with Kotlin flow, which is very easy to use, and has a lot of useful operations (flatMap, etc).
However, I am not sure how to elegantly handle network errors using this method. One approach that I can think of is to use Response<T> as the return type of the network API, like this:
// NetworkApi.kt
interface NetworkApi {
#GET("users")
suspend fun getUsers(): Response<List<User>>
}
Then in my view model, I can have an if-else to check the isSuccessful of the response, and get the real result using the .body() API if it is successful. But it will be problematic when I do some transformation in my view model. E.g.
// MyViewModel.kt
class MyViewModel(private val networkApi: NetworkApi): ViewModel() {
val usersLiveData = flow {
val response = networkApi.getUsers()
if (response.isSuccessful) {
emit(response.body()) // response.body() will be List<User>
} else {
// What should I do here?
}
}.map { // it: List<User>
// transform Users to some other class
it?.map { oneUser -> OtherClass(oneUser.userName) }
}.asLiveData()
Note the comment "What should I do here?". I don't know what to do in that case. I could wrap the responseBody (in this case, a list of Users) with some "status" (or simply just pass through the response itself). But that means that I pretty much have to use an if-else to check the status at every step through the flow transformation chain, all the way up to the UI. If the chain is really long (e.g. I have 10 map or flatMapConcat on the chain), it is really annoying to do it in every step.
What is the best way to handle network errors in this case, please?
You should have a sealed class to handle for different type of event. For example, Success, Error or Loading. Here is some of the example that fits your usecases.
enum class ApiStatus{
SUCCESS,
ERROR,
LOADING
} // for your case might be simplify to use only sealed class
sealed class ApiResult <out T> (val status: ApiStatus, val data: T?, val message:String?) {
data class Success<out R>(val _data: R?): ApiResult<R>(
status = ApiStatus.SUCCESS,
data = _data,
message = null
)
data class Error(val exception: String): ApiResult<Nothing>(
status = ApiStatus.ERROR,
data = null,
message = exception
)
data class Loading<out R>(val _data: R?, val isLoading: Boolean): ApiResult<R>(
status = ApiStatus.LOADING,
data = _data,
message = null
)
}
Then, in your ViewModel,
class MyViewModel(private val networkApi: NetworkApi): ViewModel() {
// this should be returned as a function, not a variable
val usersLiveData = flow {
emit(ApiResult.Loading(true)) // 1. Loading State
val response = networkApi.getUsers()
if (response.isSuccessful) {
emit(ApiResult.Success(response.body())) // 2. Success State
} else {
val errorMsg = response.errorBody()?.string()
response.errorBody()?.close() // remember to close it after getting the stream of error body
emit(ApiResult.Error(errorMsg)) // 3. Error State
}
}.map { // it: List<User>
// transform Users to some other class
it?.map { oneUser -> OtherClass(oneUser.userName) }
}.asLiveData()
In your view (Activity/Fragment), observe these state.
viewModel.usersLiveData.observe(this) { result ->
// Update the UI here
when(result.status) {
ApiResult.Success -> {
val data = result.data <-- return List<User>
}
ApiResult.Error -> {
val errorMsg = result.message <-- return errorBody().string()
}
ApiResult.Loading -> {
// here will actually set the state as Loading
// you may put your loading indicator here.
}
}
}
//this class represent load statement management operation
/*
What is a sealed class
A sealed class is an abstract class with a restricted class hierarchy.
Classes that inherit from it have to be in the same file as the sealed class.
This provides more control over the inheritance. They are restricted but also allow freedom in state representation.
Sealed classes can nest data classes, classes, objects, and also other sealed classes.
The autocomplete feature shines when dealing with other sealed classes.
This is because the IDE can detect the branches within these classes.
*/
ٍٍٍٍٍ
sealed class APIResponse<out T>{
class Success<T>(response: Response<T>): APIResponse<T>() {
val data = response.body()
}
class Failure<T>(response: Response<T>): APIResponse<T>() {
val message:String = response.errorBody().toString()
}
class Exception<T>(throwable: Throwable): APIResponse<T>() {
val message:String? = throwable.localizedMessage
}
}
create extention file called APIResponsrEX.kt
and create extextion method
fun <T> APIResponse<T>.onSuccess(onResult :APIResponse.Success<T>.() -> Unit) : APIResponse<T>{
if (this is APIResponse.Success) onResult(this)
return this
}
fun <T> APIResponse<T>.onFailure(onResult: APIResponse.Failure<*>.() -> Unit) : APIResponse<T>{
if (this is APIResponse.Failure<*>)
onResult(this)
return this
}
fun <T> APIResponse<T>.onException(onResult: APIResponse.Exception<*>.() -> Unit) : APIResponse<T>{
if (this is APIResponse.Exception<*>) onResult(this)
return this
}
merge it with Retrofit
inline fun <T> Call<T>.request(crossinline onResult: (response: APIResponse<T>) -> Unit) {
enqueue(object : retrofit2.Callback<T> {
override fun onResponse(call: Call<T>, response: Response<T>) {
if (response.isSuccessful) {
// success
onResult(APIResponse.Success(response))
} else {
//failure
onResult(APIResponse.Failure(response))
}
}
override fun onFailure(call: Call<T>, throwable: Throwable) {
onResult(APIResponse.Exception(throwable))
}
})
}
I'm working with MVVM, and I have made different implementations of it, but one thing that is still making me doubt is how do I get data from a Repository (Firebase) from my ViewModel without attaching any lifecycle to the ViewModel.
I have implemented observeForever() from the ViewModel, but I don't think that is a good idea because I think I should communicate from my repository to my ViewModel either with a callback or a Transformation.
I leave here an example where I fetch a device from Firebase and update my UI, if we can see here, I'm observing the data coming from the repo from the UI, but from the ViewModel I'm also observing data from the repo, and here is where I really doubt if I'm using the right approach, since I don't know if observeForever() will be cleared on onCleared() if my view is destroyed, so it won't keep the observer alive if the view dies.
UI
override fun onCreate(savedInstanceState: Bundle?) {
super.onCreate(savedInstanceState)
setContentView(R.layout.activity_main)
button.setOnClickListener {
val deviceId = editText.text.toString().trim()
observeData(deviceId)
}
}
fun observeData(deviceId:String){
viewModel.fetchDeviceData(deviceId).observe(this, Observer {
textView.text = "Tipo: ${it.devType}"
})
ViewModel
class MainViewmodel: ViewModel() {
private val repo = Repo()
fun fetchDeviceData(deviceId:String):LiveData<Device>{
val mutableData = MutableLiveData<Device>()
repo.getDeviceData(deviceId).observeForever {
mutableData.value = it
}
return mutableData
}
}
Repository
class Repo {
private val db = FirebaseDatabase.getInstance().reference
fun getDeviceData(deviceId:String):LiveData<Device>{
val mutableData = MutableLiveData<Device>()
db.child(deviceId).child("config/device").addListenerForSingleValueEvent(object: ValueEventListener{
override fun onDataChange(dataSnapshot: DataSnapshot) {
val device = dataSnapshot.getValue(Device::class.java)
mutableData.value = device
}
override fun onCancelled(dataError: DatabaseError) {
Log.e("Error","handle error callback")
}
})
return mutableData
}
}
This example just shows how to fetch the device from Firebase, it works, but from my ViewModel, it keeps making me think that observeForever() is not what I'm looking for to communicate data between the repository to the ViewModel.
I have seen Transformations, but I, in this case, I just need to deliver the entire Device object to my UI, so I don't need to transform the Object I'm retrieving to another Object
What should be here the right approach to communicate the repository and the ViewModel properly?
is observeForever lifecycle aware?
No, that's why it's called observeForever.
I have implemented observeForever() from the ViewModel, but I don't think that is a good idea
No, it's not, you should be using Transformations.switchMap {.
since I don't know if observeForever() will be cleared on onCleared() if my view is destroyed, so it won't keep the observer alive if the view dies.
Well if you're not clearing it in onCleared() using removeObserver(observer), then it won't clear itself, because it observes forever.
here is where I really doubt if I'm using the right approach,
No, you can do much better than this following a reactive approach.
override fun onCreate(savedInstanceState: Bundle?) {
super.onCreate(savedInstanceState)
setContentView(R.layout.activity_main)
button.setOnClickListener {
val deviceId = editText.text.toString().trim()
viewModel.onSelectedDeviceChanged(deviceId)
}
viewModel.selectedDevice.observe(this, Observer { device ->
textView.text = "Tipo: ${device.devType}"
})
}
And
class MainViewModel(
private val savedStateHandle: SavedStateHandle,
): ViewModel() {
private val repo = Repo() // TODO: move to Constructor Argument with ViewModelProvider.Factory
private val selectedDeviceId: MutableLiveData<String> = savedStateHandle.getLiveData<String>("selectedDeviceId")
fun onSelectedDeviceChanged(deviceId: String) {
selectedDeviceId.value = deviceId
}
val selectedDevice = Transformations.switchMap(selectedDeviceId) { deviceId ->
repo.getDeviceData(deviceId)
}
}
And
class Repo {
private val db = FirebaseDatabase.getInstance().reference // TODO: move to constructor arg? Probably
fun getDeviceData(deviceId:String) : LiveData<Device> {
return object: MutableLiveData<Device>() {
private val mutableLiveData = this
private var query: Query? = null
private val listener: ValueEventListener = object: ValueEventListener {
override fun onDataChange(dataSnapshot: DataSnapshot) {
val device = dataSnapshot.getValue(Device::class.java)
mutableLiveData.value = device
}
override fun onCancelled(dataError: DatabaseError) {
Log.e("Error","handle error callback")
}
}
override fun onActive() {
query?.removeEventListener(listener)
val query = db.child(deviceId).child("config/device")
this.query = query
query.addValueEventListener(listener)
}
override fun onInactive() {
query?.removeEventListener(listener)
query = null
}
}
}
}
This way, you can observe for changes made in Firebase (and therefore be notified of future changes made to your values) using LiveData, rather than only execute a single fetch and then not be aware of changes made elsewhere to the same data.
To use ObserveForever, you need to remove the observer inside onClear in the ViewModel.
In this case, I would suggest to use Transformation even though you just need a direct mapping without any processing of the data, which is actually the same as what you are doing with the observer for observerForever.
observeForever() is not Lifecycle aware and will continue to run until removeObserver() is called.
In your ViewModel do this instead,
class MainViewmodel: ViewModel() {
private val repo = Repo()
private var deviceData : LiveData<Device>? = null
fun fetchDeviceData(deviceId:String):LiveData<Device>{
deviceData = repo.getDeviceData(deviceId)
return deviceData!!
}
}
I Am using MVVM architecture to simple project. Then i stack in this case, when i have to return value from Model DataSource (Lambda function) to Repository then ViewModel will observe this repository. Please correct me if this not ideally and give me some advise for the true MVVM in android. i want to use LiveData only instead of RxJava in this case, because many sample in Github using RxJava.
In my Model i have class UserDaoImpl, code snippet like below
class UserDaoImpl : UserDao {
private val resultCreateUser = MutableLiveData<AppResponse>()
private val mAuth : FirebaseAuth by lazy {
FirebaseAuth.getInstance()
}
override fun createUser(user: User) {
mAuth.createUserWithEmailAndPassword(user.email, user.password)
.addOnCompleteListener {
//I DID NOT REACH THIS LINE
println("hasilnya ${it.isSuccessful} ")
if(it.isSuccessful){
val appResponse = AppResponse(true, "oke")
resultCreateUser.postValue(appResponse)
}else{
val appResponse = AppResponse(false, "not oke -> ${it.result.toString()}")
resultCreateUser.postValue(appResponse)
}
}
.addOnFailureListener {
println("hasilnya ${it.message}")
val appResponse = AppResponse(false, "not oke -> ${it.message}")
resultCreateUser.postValue(appResponse)
}
}
override fun getResultCreateUser() = resultCreateUser
}
And this is my Repository snippet code
class RegisterRepositoryImpl private constructor(private val userDao: UserDao) : RegisterRepository{
companion object{
#Volatile private var instance : RegisterRepositoryImpl? = null
fun getInstance(userDao: UserDao) = instance ?: synchronized(this){
instance ?: RegisterRepositoryImpl(userDao).also {
instance = it
}
}
}
override fun registerUser(user: User) : LiveData<AppResponse> {
userDao.createUser(user)
return userDao.getResultCreateUser() as LiveData<AppResponse>
}
}
Then this is my ViewModel
class RegisterViewModel (private val registerRepository: RegisterRepository) : ViewModel() {
val signUpResult = MutableLiveData<AppResponse>()
fun registerUser(user: User){
println(user.toString())
val response = registerRepository.registerUser(user)
signUpResult.value = response.value
}
}
If i execute the snippet code above, the result always nullpointer in signUpResult
This is my Activity
lateinit var viewModel: RegisterViewModel
override fun onCreate(savedInstanceState: Bundle?) {
super.onCreate(savedInstanceState)
setContentView(R.layout.activity_register)
initializeUI()
}
private fun initializeUI() {
val factory = InjectorUtils.provideRegisterViewModelFactory()
viewModel = ViewModelProviders.of(this, factory).get(RegisterViewModel::class.java)
viewModel.signUpResult.observe(this, Observer {
//IT always null
if(it.success){
// to HomeActivity
Toast.makeText(this, "Success! ${it.msg}", Toast.LENGTH_SHORT).show()
}else{
Toast.makeText(this, "FALSE! ${it.msg}", Toast.LENGTH_SHORT).show()
}
})
register_btn.setOnClickListener {
val username = name.text.toString()
val email = email.text.toString()
val password = password.text.toString()
val phone = number.text.toString()
val user = User(0, username,"disana", email, password, "disana")
viewModel.registerUser(user)
}
}
Crash occured when i press register button
I'm not 100% sure, but I think the problem is in your ViewModel, where you are trying to pass by reference MutableLiveData. Your Activity is observing signUpResult MutableLiveData, but you are never posting new value, you are trying to change reference of that LiveData to one in Repository.
val signUpResult = MutableLiveData<AppResponse>()
fun registerUser(user: User){
println(user.toString())
val response = registerRepository.registerUser(user)
signUpResult.value = response.value
}
I think that the solution here is to let your ViewModel return LiveData, which is returned from Repository.
fun registerUser(user: User): MutableLiveData<AppResponse> {
println(user.toString())
return registerRepository.registerUser(user)
}
And you need to observe function registerUser(user) in your Activity.
viewModel.registerUser(user).observe(this, Observer {
But now you encountered another problem. By this example you will trigger observe method every time your button is clicked. So you need to split in repository your function, you need to make one only for returning userDao.getResultCreateUser() as LiveData<AppResponse>, and the other to trigger userDao.create(user) .
So you can make two functions in your repository
override fun observeRegistrationResponse() : LiveData<AppResponse> {
return userDao.getResultCreateUser() as LiveData<AppResponse>
}
override fun registerUser(user: User) {
userDao.createUser(user)
}
Now also in ViewModel you need to make separate function for observing result and for sending request for registration.
fun observeRegistrationResponse(): LiveData<AppResponse> {
return registerRepository.observeRegistrationResponse()
}
fun registerUser(user: User){
println(user.toString())
registerRepository.registerUser(user)
}
And finally you can observe in your function initializeUI
viewModel.observeRegistrationResponse().observe(this, Observer {
And send registration request on button click
viewModel.registerUser(user)
Sorry for long response, but I tried to explain why you need to change your approach. I hope I helped you a bit to understand how LiveData works.
I have a repository setup like this
class ServerTimeRepo #Inject constructor(private val retrofit: Retrofit){
var liveDataTime = MutableLiveData<TimeResponse>()
fun getServerTime(): LiveData<TimeResponse> {
val serverTimeService:ServerTimeService = retrofit.create(ServerTimeService::class.java)
val obs = serverTimeService.getServerTime()
obs.subscribeOn(Schedulers.io()).observeOn(AndroidSchedulers.mainThread()).unsubscribeOn(Schedulers.io())
.subscribe(object : Observer<Response<TimeResponse>> {
override fun onComplete() {
}
override fun onSubscribe(d: Disposable) {
}
override fun onNext(t: Response<TimeResponse>) {
val gson = Gson()
val json: String?
val code = t.code()
val cs = code.toString()
if (!cs.equals("200")) {
json = t.errorBody()!!.string()
val userError = gson.fromJson(json, Error::class.java)
} else {
liveDataTime.value = t.body()
}
}
override fun onError(e: Throwable) {
}
})
return liveDataTime
}
}
Then I have a viewmodel calling this repo like this
class ServerTimeViewModel #Inject constructor(private val serverTimeRepo: ServerTimeRepo):ViewModel() {
fun getServerTime(): LiveData<TimeResponse> {
return serverTimeRepo.getServerTime()
}
}
Then I have an activity where I have an onClickListener where I am observing the livedata, like this
tvPWStart.setOnClickListener {
val stlv= serverTimeViewModel.getServerTime()
stlv.observe(this#HomeScreenActivity, Observer {
//this is getting called multiple times??
})
}
I don't know what's wrong in this. Can anyone point me in the right direction? Thanks.
Issue is that every time your ClickListener gets fired, you observe LiveData again and again. So, you can solve that problem by following solution :
Take a MutableLiveData object inside your ViewModel privately & Observe it as LiveData.
class ServerTimeViewModel #Inject constructor(private val serverTimeRepo: ServerTimeRepo):ViewModel() {
private val serverTimeData = MutableLiveData<TimeResponse>() // We make private variable so that UI/View can't modify directly
fun getServerTime() {
serverTimeData.value = serverTimeRepo.getServerTime().value // Rather than returning LiveData, we set value to our local MutableLiveData
}
fun observeServerTime(): LiveData<TimeResponse> {
return serverTimeData //Here we expose our MutableLiveData as LiveData to avoid modification from UI/View
}
}
Now, we observe this LiveData directly outside of ClickListener and we just call API method from button click like below :
//Assuming that this code is inside onCreate() of your Activity/Fragment
//first we observe our LiveData
serverTimeViewModel.observeServerTime().observe(this#HomeScreenActivity, Observer {
//In such case, we won't observe multiple LiveData but one
})
//Then during our ClickListener, we just do API method call without any callback.
tvPWStart.setOnClickListener {
serverTimeViewModel.getServerTime()
}
I'm in the process of wrapping my head around Architecture Components / MVVM.
Let's say I have a repository, a ViewModel and a Fragment. I'm using a Resource class as a wrapper to expose network status, like suggested in the Guide to architecture components.
My repository currently looks something like this (simplified for brevity):
class MyRepository {
fun getLists(organizationId: String) {
var data = MutableLiveData<Resource<List<Something>>>()
data.value = Resource.loading()
ApolloClient().query(query)
.enqueue(object : ApolloCall.Callback<Data>() {
override fun onResponse(response: Response<Data>) {
response.data()?.let {
data.postValue(Resource.success(it))
}
}
override fun onFailure(exception: ApolloException) {
data.postValue(Resource.exception(exception))
}
})
}
Then in the ViewModel, I also declare a MutableLiveData:
var myLiveData = MutableLiveData<Resource<List<Something>>>()
fun getLists(organizationId: String, forceRefresh: Boolean = false) {
myLiveData = myRepository.getLists(organizationId)
}
Finally, the Fragment:
viewModel.getLists.observe(this, Observer {
it?.let {
if (it.status.isLoading()) showLoading() else hideLoading()
if (it.status == Status.SUCCESS) {
it.data?.let {
adapter.replaceData(it)
setupViews()
}
}
if (it.status == Status.ERROR) {
// Show error
}
}
})
As you see, there will be an issue with the observer not being triggered, since the LiveData variable will be reset in the process (the Repository creates a new instance).
I'm trying to figure out the best way to make sure that the same LiveData variable is used between the Repository and ViewModel.
I thought about passing the LiveData from the ViewModel to the getLists method, so that the Repository would be using the object from the ViewModel, but even if it works, it seems wrong to do that.
What I mean is something like that:
ViewModel
var myLiveData = MutableLiveData<Resource<List<Something>>>()
fun getLists(organizationId: String, forceRefresh: Boolean = false) {
myRepository.getLists(myLiveData, organizationId)
}
Repository
fun getLists(data: MutableLiveData<Resource<List<Something>>>, organizationId: String) {
...
}
I think I figured out how to do it, thanks to #NSimon for the cue.
My repository stayed the same, and my ViewModel looks like this:
class MyViewModel : ViewModel() {
private val myRepository = MyRepository()
private val organizationIdLiveData = MutableLiveData<String>()
private val lists = Transformations.switchMap(organizationIdLiveData) { organizationId -> myRepository.getLists(organizationId) }
fun getLists() : LiveData<Resource<MutableList<Something>>> {
return lists
}
fun fetchLists(organizationId: String, forceRefresh: Boolean = false) {
if (organizationIdLiveData.value == null || forceRefresh) {
organizationIdLiveData.value = organizationId
}
}
}
I observe getLists() in my fragment, and call viewModel.fetchLists(id) when I want the data. Seems legit?