Run Headless JS task with WorkManager
Headless JS is a way to run tasks in JavaScript while your app is in the background. It can be used, for example, to sync fresh data, handle push notifications, or play music.
Why not use existing RN API?
From Android 8.0(API Level 26), the system imposes limitations on running services in the background. As React Native's stock headless task need to be executed with service, these limitations prevent to execute Headless JS in some cases.
To resolve this issue, this library provide a way to execute Headless JS task with WorkManager.
npm install react-native-headless-task-worker
You need to set your project's compileSdkVersion
and targetSdkVersion
to 31
to use WorkManager
.
Then, you need to add following dependency to your app level build.gradle
.
implementation 'androidx.work:work-runtime:2.7.1'
A task is an async function that you register on AppRegistry
, similar to registering React applications:
import { AppRegistry } from 'react-native';
AppRegistry.registerHeadlessTask('SomeTaskName', () =>
require('SomeTaskName')
);
Then, in SomeTaskName.js
:
module.exports = async (taskData) => {
// do stuff
};
You can do anything in your task such as network requests, timers and so on, as long as it doesn't touch UI. Once your task completes (i.e. the promise is resolved), React Native will go into "paused" mode (unless there are other tasks running, or there is a foreground app).
Yes, this does still require some native code, but it's pretty thin. You need to extend HeadlessJsTaskWorker
and override getTaskConfig
, e.g.:
package com.your_application_name;
import android.content.Context;
import androidx.annotation.NonNull;
import androidx.annotation.Nullable;
import androidx.work.Data;
import androidx.work.WorkerParameters;
import com.facebook.react.bridge.Arguments;
import com.facebook.react.jstasks.HeadlessJsTaskConfig;
import io.wjaykim.rnheadlesstaskworker.HeadlessJsTaskWorker;
public class MyTaskWorker extends HeadlessJsTaskWorker {
public MyTaskWorker(@NonNull Context context, @NonNull WorkerParameters params) {
super(context, params);
}
@Nullable
@Override
protected HeadlessJsTaskConfig getTaskConfig(Data data) {
if (data != null) {
return new HeadlessJsTaskConfig(
"SomeTaskName",
Arguments.makeNativeMap(data.getKeyValueMap()),
5000, // timeout for the task
false // optional: defines whether or not the task is allowed in foreground. Default is false
);
}
return null
}
}
Now, whenever your work is executed, JS will spin up, run your task, then spin down. You can schedule your work by enqueuing WorkRequest to WorkManager.
Example:
Data inputData = new Data.Builder()
.putString("message", "Task has executed")
.build();
WorkRequest headlessJsTaskWorkRequest =
new OneTimeWorkRequest.Builder(MyTaskWorker.class)
.setInputData(inputData)
.build();
WorkManager
.getInstance(this)
.enqueue(headlessJsTaskWorkRequest);
By default, the headless JS task will not perform any retries. In order to do so, you need to create a HeadlessJsRetryPolicy
and throw a specific Error
.
LinearCountingRetryPolicy
is an implementation of HeadlessJsRetryPolicy
that allows you to specify a maximum number of retries with a fixed delay between each attempt. If that does not suit your needs then you can implement your own HeadlessJsRetryPolicy
. These policies can be passed as an extra argument to the HeadlessJsTaskConfig
constructor, e.g.
HeadlessJsRetryPolicy retryPolicy = new LinearCountingRetryPolicy(
3, // Max number of retry attempts
1000 // Delay between each retry attempt
);
return new HeadlessJsTaskConfig(
'SomeTaskName',
Arguments.makeNativeMap(data.getKeyValueMap()),
5000,
false,
retryPolicy
);
A retry attempt will only be made when a specific Error
is thrown. Inside a headless JS task, you can import the error and throw it when a retry attempt is required.
Example:
import {HeadlessJsTaskError} from 'HeadlessJsTask';
module.exports = async (taskData) => {
const condition = ...;
if (!condition) {
throw new HeadlessJsTaskError();
}
};
If you wish all errors to cause a retry attempt, you will need to catch them and throw the above error.
- The function passed to
setTimeout
does not always behave as expected. Instead the function is called only when the application is launched again. If you only need to wait, use the retry functionality. - By default, your app will crash if you try to run a task while the app is in the foreground. This is to prevent developers from shooting themselves in the foot by doing a lot of work in a task and slowing the UI. You can pass a fourth
boolean
argument to control this behaviour.
Work can be started from Java API. First you need to decide when the work should be started and implement your solution accordingly. Here is an example that reacts to power connection.
Following lines shows registering broadcast receiver listening for power connection.
registerReceiver(new PowerConnectReceiver(), new IntentFilter(Intent.ACTION_POWER_CONNECTED));
Broadcast receiver then handles intent that was broadcasted in onReceive function. This is a great place to check whether your app is on foreground or not. If app is not on foreground we can build our WorkRequest
, with no data or additional data by assigning input data. Finally we will submit our work request.
public class PowerConnectReceiver extends BroadcastReceiver {
@Override
public void onReceive(final Context context, final Intent intent) {
/*
This part will be called every time power is connected
*/
if (!isAppOnForeground((context))) {
/*
We will submit work request with extra message
*/
Data inputData = new Data.Builder()
.putString("message", "Power has connected")
.build();
WorkRequest headlessJsTaskWorkRequest =
new OneTimeWorkRequest.Builder(MyTaskWorker.class)
.setInputData(inputData)
.build();
WorkManager
.getInstance(context)
.enqueue(headlessJsTaskWorkRequest);
}
}
private boolean isAppOnForeground(Context context) {
/*
We need to check if app is in foreground otherwise the app will crash.
http://stackoverflow.com/questions/8489993/check-android-application-is-in-foreground-or-not
*/
ActivityManager activityManager = (ActivityManager) context.getSystemService(Context.ACTIVITY_SERVICE);
List<ActivityManager.RunningAppProcessInfo> appProcesses =
activityManager.getRunningAppProcesses();
if (appProcesses == null) {
return false;
}
final String packageName = context.getPackageName();
for (ActivityManager.RunningAppProcessInfo appProcess : appProcesses) {
if (appProcess.importance ==
ActivityManager.RunningAppProcessInfo.IMPORTANCE_FOREGROUND &&
appProcess.processName.equals(packageName)) {
return true;
}
}
return false;
}
}
MIT