background_fetch 0.5.2

Periodic callbacks in the background for both iOS and Android. Includes Android Headless implementation.

0.5.2 - 2020-02-20

• [Android] Fix Android NPE in hasTaskId for case where plugin is installed first time in had previous version of plugin

0.5.1 - 2020-02-19

Minor Breaking Change for iOS Setup

• [iOS] It's no longer necessary to registerBGProcessingTask in AppDelegate.m for tasks registered for use with #scheduleTask. The SDK now reads the App .plist and automatically registers those tasks found in "Permitted background task scheduler identifiers". Remove all code in your AppDelegate.m that references TSBackgroundFetch.

0.5.0 - 2020-02-03

• [Added] [Android] New option forceAlarmManager for bypassing JobScheduler mechanism in favour of AlarmManager for more precise scheduling task execution.
• [Changed] Migrate iOS deprecated "background-fetch" API to new BGTaskScheduler. See new required steps in iOS Setup.
• [Added] Added new BackgroundFetch.scheduleTask method for scheduling custom "onehot" and periodic tasks in addition to the default fetch-task.

BackgroundFetch.configure(BackgroundFetchConfig(
  minimumFetchInterval: 15,
  stopOnTerminate: false
), (String taskId) {  // <-- [NEW] taskId provided to Callback
  print("[BackgroundFetch] taskId: $taskId");
  switch(taskId) {
    case 'foo':
      // Handle scheduleTask 'foo'
      break;
    default:
      // Handle default fetch event.
      break;
  }
  BackgroundFetch.finish(taskId);  // <-- [NEW] Provided taskId to #finish method.
});

// This event will end up in Callback provided to #configure above.
BackgroundFetch.scheduleTask(TaskConfig(
  taskId: 'foo',  //<-- required
  delay: 60000,
  periodic: false  
));

Breaking Changes

• With the introduction of ability to execute custom tasks via #scheduleTask, all tasks are executed in the Callback provided to #configure. As a result, this Callback is now provided an argument String taskId. This taskId must now be provided to the #finish method, so that the SDK knows which task is being #finished.

BackgroundFetch.configure(BackgroundFetchConfig(
  minimumFetchInterval: 15,
  stopOnTerminate: false
), (String taskId) {  // <-- [NEW] taskId provided to Callback
  print("[BackgroundFetch] taskId: $taskId");
  BackgroundFetch.finish(taskId);  // <-- [NEW] Provided taskId to #finish method.
});

And with the Headless Task, as well:

/// This "Headless Task" is run when app is terminated.
void backgroundFetchHeadlessTask(String taskId) async {  // <-- 1.  Headless task receives String taskId
  print("[BackgroundFetch] Headless event received: $taskId");
  
  BackgroundFetch.finish(taskId);  // <-- 2.  #finish with taskId here as well.
}

void main() {
  // Enable integration testing with the Flutter Driver extension.
  // See https://flutter.io/testing/ for more info.
  runApp(new MyApp());

  // Register to receive BackgroundFetch events after app is terminated.
  // Requires {stopOnTerminate: false, enableHeadless: true}
  BackgroundFetch.registerHeadlessTask(backgroundFetchHeadlessTask);
}

0.4.0 - 2019-12-17

• [Changed] Upgrade to new Flutter Plugin API "V2". Requires flutter sdk version 1.12. See Upgrading pre 1.12 Android Projects

0.3.2 - 2019-10-06

• [Fixed] Resolve Android StrictMode violations; typically from accessing SharedPreferences on main-thread.

0.3.1 - 2019-09-20

• Fix error FlutterMain.findBundleAppPath(). The plugin modified a deprecated API for flutter 1.9.1, breaking those on previous flutter versions. Will use deprecated API for now.

0.3.0 - 2019-09-18

• Implement Android JobInfo constraints.
• Fix NSLog warnings casting to long
• Default startOnBoot: true in example

0.2.0 - 2019-03-15

• Use AndroidX.

0.1.2 - 2019-02-28

• Fixed bug with setting jobServiceClass using a reference to HeadlessJobService.class. This crashes devices < api 21, since Android's JobService wasn't available until then. Simply provide the class name as a String.

0.1.1 - 2018-11-21

• Fixed issue with Android headless config.

0.1.0

• First working implementation

0.0.1

• First working implementation