Did you have fun with Livedata? StateFlow makes you feel better

This is the 19th day of my participation in the More text Challenge. For details, see more text Challenge

Follow my official account “Ananzhuo” to learn more

StateFlow and Livedata

Both similarities and differences between

There are a lot of similarities between StateFlow and Livedata if you look closely. Both are designed to look at data and are ideal for responsive development.

But there are some differences:

• The StateFlow construct has a default value, but Livedata does not
• Livedata can receive data updates only when its life cycle is in the started or resumed state, while StateFlow can receive data even when it is back to the background.

Cold flow of heat flow

Cold flow: The producer starts emitting data when the consumer calls the collect method. The following code example is a cold flow.

(2.. 9).asFlow().collect{ println(it) }Copy the code

Heat flow: StateFlow is heat flow in this article, and the producer will transmit data without calling the collect method.

Livedata use

Check out my last post about Livedata

Juejin. Cn/post / 697568…

Juejin. Cn/post / 697831…

StateFlow use

Implement data listening using StateFlow

StateFlow can change the data state with setValue

Here’s a brief introduction to usage:

1. Add the Activity and Lifecycle extension libraries

implementation 'androidx. Activity: activity - KTX: 1.2.2'
    implementation "Androidx. Lifecycle: lifecycle - livedata - KTX: 2.2.0." "
Copy the code

2. Add StateFlow code to the ViewModel

Private val mutableStateFlow = mutableStateFlow (" safe "); StateFlow<String> = mutableStateFlow <String> = mutableStateFlow Fun changeData(data: String) {mutableStateFlow.value = data// Send data}Copy the code

3. Data monitoring in the activity

When we click the button, we call the viewModel’s changeData method to update the flow’s data.

BtnSendData. SetOnClickListener {viewModel. ChangeData (" public no. : In peace in zhuo ")} lifecycleScope. Launch {/ / here is run in the main thread, so the following code can update the UI viewModel. StateFlow. Collect {tvShowData. Text = it}}Copy the code

4. Implementation effect

This example achieves the following effect: The first time you enter the page StateFlow sends the default value: “Safe “, click the button and call the changeData method of ViewModel to update the data, and display it in TextView.

The usage here is basically the same as that of Livedata, but in the next example you’ll find that StateFlow has much less template code and is much cleaner

Implement the effects of MediatorLiveData using StateFlow

You can check out my post about MediatorLiveData:

Juejin. Cn/post / 697568…

How can the same effect be achieved with StateFlow?

Code for StateFlow implementation:

1. Code declaration in ViewModel

class StateFlowMediatorViewModel : ViewModel() {private var count1 = 0 private var count2 = 0 private val flow1 = 0 MutableStateFlow(0) private val flow2 = MutableStateFlow(0) val flow = flow1.combine(flow2) { data1, Data2 -> data1 + data2 } fun Flow1pp () {flow1.value = ++count1// Add 1 to the number of clicks on the first button} fun Flow2pp () {flow2.value = ++count2// Add 1 to the number of clicks on the second button} }Copy the code

2. Code declaration in the activity

findViewById<Button>(R.id.btn_mediator1).setOnClickListener { model.flow1pp() } findViewById<Button>(R.id.btn_mediator2).setOnClickListener { model.flow2pp() } lifecycleScope.launch { model.flow.collect { textView? .text = if (it > 10) "$it" else "$it"}}Copy the code

3. Implementation effect

The demo in the previous Livadata article was implemented with a very small amount of code, so StateFlow is really good, and it’s definitely better than that

Some features of StateFlow

2021-07-03 New content in this section (follow public account: Ananzhuo learn more knowledge)

Multiple StateFlows are registered simultaneously in the ViewModel

Conclusion: If multiple StateFlows are registered in the ViewModel, only the first registered StateFlow can receive a callback for data updates

Here’s the code to verify:

The activity of the code

LifecycleScope. Launch {multipleViewModel. State11. Collect {logEE (it) count++ TV. The text = "received StateFlow1 data update $count time}" MultipleViewModel. State22. Collect {logEE (it) count++ TV. The text = "received $count StateFlow2 data update time"}} The findViewById < Button > (R.i db tn_state1_senddata) setOnClickListener {multipleViewModel. SendData1 () / / click on the Button to update the first StateFlow value } findViewById<Button>(R.id.btn_state2_senddata).setOnClickListener { MultipleViewModel. SendData2 () / / click on the button to update the second StateFlow value}Copy the code

ViewModel:

Private val state1 = MutableStateFlow<String>(" hahaha 1") private Val State2 = MutableStateFlow<String>(" hahaha 2") private var  count = 0 val state11 = state1.asStateFlow() val state22 = state2.asStateFlow() fun sendData1() { count++ state1.value = "state1 changed $count"} fun sendData2() {count++ state2. Value = "state2 changed $count"}Copy the code

2. Achieve results

Through our implementation, we found that:

The first registered State1 calls back a collect on each click to update the TextView display value

The second registered state2 although we also called set state2. Value =”??” “, but never a pullback

This proves the conclusion we stated at the beginning of this section

StateFlow and Livedata are similar and different

2021-07-03 New content in this section (follow public account: Ananzhuo learn more knowledge)

StateFlow receives only one callback for multiple updates to the same data, while Livedata receives multiple callbacks

For example, if we set a button that resets the value of Livedata or StateFlow every time we click it (the value does not change, such as setting Value =4), StateFlow will receive only one callback, but Livedata will receive an unlimited callback. The following two sections demonstrate this with code and effects, respectively

This is not a good or bad feature for StateFlow or Livedata, but a good or bad one

StateFlow sets the same value to value multiple times

1. code

Registration code in activity:

LifecycleScope. Launch {multipleViewModel. State11. Collect {logEE (it) count++ TV. The text = "received $count data update time"} multipleViewModel.state22.collect { logEE(it) } }Copy the code

Modify value code in StateFlow:

Fun sendData1() {state1.value = "state1 changed "// We set this value}Copy the code

2. Running effect diagram

We found that StateFlow’s Collect only called back the data once, even though we clicked send multiple times

Livedata sets the same value to value multiple times

1. code

Code in an Activity

Btn_updatewith_simdata. SetOnClickListener {model. The setData () / / click on the button to send data} model. Livedata. Observe (this) {logEE (the "receiving data is: ${it}") count++ tv_beupdate.text=" ${it}" ${it}") count++ tv_beupdate.text=" ${it}"Copy the code

Code in ViewModel:

val livedata = MutableLiveData<Int>()
    fun setData(){
        livedata.value=124
    }
Copy the code

2. The effect

We found that we had sent the same data multiple times, and each time Livedata received a callback to update the data

Follow my official account “Ananzhuo” to learn more