public abstract class

JobIntentService

extends Service

 java.lang.Object

↳Service

↳androidx.core.app.JobIntentService

Gradle dependencies

compile group: 'androidx.core', name: 'core', version: '1.9.0-alpha04'

  • groupId: androidx.core
  • artifactId: core
  • version: 1.9.0-alpha04

Artifact androidx.core:core:1.9.0-alpha04 it located at Google repository (https://maven.google.com/)

Androidx artifact mapping:

androidx.core:core com.android.support:support-compat

Androidx class mapping:

androidx.core.app.JobIntentService android.support.v4.app.JobIntentService

Overview

Helper for processing work that has been enqueued for a job/service. When running on or later, the work will be dispatched as a job via . When running on older versions of the platform, it will use .

You must publish your subclass in your manifest for the system to interact with. This should be published as a , as described for that class, since on O and later platforms it will be executed that way.

Use JobIntentService.enqueueWork(Context, Class, int, Intent) to enqueue new work to be dispatched to and handled by your service. It will be executed in JobIntentService.onHandleWork(Intent).

You do not need to use WakefulBroadcastReceiver when using this class. When running on , the JobScheduler will take care of wake locks for you (holding a wake lock from the time you enqueue work until the job has been dispatched and while it is running). When running on previous versions of the platform, this wake lock handling is emulated in the class here by directly calling the PowerManager; this means the application must request the permission.

There are a few important differences in behavior when running on or later as a Job vs. pre-O:

  • When running as a pre-O service, the act of enqueueing work will generally start the service immediately, regardless of whether the device is dozing or in other conditions. When running as a Job, it will be subject to standard JobScheduler policies for a Job with a of 0: the job will not run while the device is dozing, it may get delayed more than a service if the device is under strong memory pressure with lots of demand to run jobs.

  • When running as a pre-O service, the normal service execution semantics apply: the service can run indefinitely, though the longer it runs the more likely the system will be to outright kill its process, and under memory pressure one should expect the process to be killed even of recently started services. When running as a Job, the typical execution time limit will apply, after which the job will be stopped (cleanly, not by killing the process) and rescheduled to continue its execution later. Job are generally not killed when the system is under memory pressure, since the number of concurrent jobs is adjusted based on the memory state of the device.

Here is an example implementation of this class:

frameworks/support/samples/Support4Demos/src/main/java/com/example/android/supportv4/app/SimpleJobIntentService.java complete

Summary

Constructors
publicJobIntentService()

Default empty constructor.

Methods
public static voidenqueueWork(Context context, java.lang.Class<java.lang.Object> cls, int jobId, Intent work)

Call this to enqueue work for your subclass of JobIntentService.

public static voidenqueueWork(Context context, ComponentName component, int jobId, Intent work)

Like JobIntentService.enqueueWork(Context, Class, int, Intent), but supplies a ComponentName for the service to interact with instead of its class.

public booleanisStopped()

Returns true if JobIntentService.onStopCurrentWork() has been called.

public IBinderonBind(Intent intent)

Returns the IBinder for the when running as a JobService on O and later platforms.

public voidonCreate()

public voidonDestroy()

protected abstract voidonHandleWork(Intent intent)

Called serially for each work dispatched to and processed by the service.

public intonStartCommand(Intent intent, int flags, int startId)

Processes start commands when running as a pre-O service, enqueueing them to be later dispatched in JobIntentService.onHandleWork(Intent).

public booleanonStopCurrentWork()

This will be called if the JobScheduler has decided to stop this job.

public voidsetInterruptIfStopped(boolean interruptIfStopped)

Control whether code executing in JobIntentService.onHandleWork(Intent) will be interrupted if the job is stopped.

from java.lang.Objectclone, equals, finalize, getClass, hashCode, notify, notifyAll, toString, wait, wait, wait

Constructors

public JobIntentService()

Default empty constructor.

Methods

public void onCreate()

public int onStartCommand(Intent intent, int flags, int startId)

Processes start commands when running as a pre-O service, enqueueing them to be later dispatched in JobIntentService.onHandleWork(Intent).

public IBinder onBind(Intent intent)

Returns the IBinder for the when running as a JobService on O and later platforms.

public void onDestroy()

public static void enqueueWork(Context context, java.lang.Class<java.lang.Object> cls, int jobId, Intent work)

Call this to enqueue work for your subclass of JobIntentService. This will either directly start the service (when running on pre-O platforms) or enqueue work for it as a job (when running on O and later). In either case, a wake lock will be held for you to ensure you continue running. The work you enqueue will ultimately appear at JobIntentService.onHandleWork(Intent).

Parameters:

context: Context this is being called from.
cls: The concrete class the work should be dispatched to (this is the class that is published in your manifest).
jobId: A unique job ID for scheduling; must be the same value for all work enqueued for the same class.
work: The Intent of work to enqueue.

public static void enqueueWork(Context context, ComponentName component, int jobId, Intent work)

Like JobIntentService.enqueueWork(Context, Class, int, Intent), but supplies a ComponentName for the service to interact with instead of its class.

Parameters:

context: Context this is being called from.
component: The published ComponentName of the class this work should be dispatched to.
jobId: A unique job ID for scheduling; must be the same value for all work enqueued for the same class.
work: The Intent of work to enqueue.

protected abstract void onHandleWork(Intent intent)

Called serially for each work dispatched to and processed by the service. This method is called on a background thread, so you can do long blocking operations here. Upon returning, that work will be considered complete and either the next pending work dispatched here or the overall service destroyed now that it has nothing else to do.

Be aware that when running as a job, you are limited by the maximum job execution time and any single or total sequential items of work that exceeds that limit will cause the service to be stopped while in progress and later restarted with the last unfinished work. (There is currently no limit on execution duration when running as a pre-O plain Service.)

Parameters:

intent: The intent describing the work to now be processed.

public void setInterruptIfStopped(boolean interruptIfStopped)

Control whether code executing in JobIntentService.onHandleWork(Intent) will be interrupted if the job is stopped. By default this is false. If called and set to true, any time JobIntentService.onStopCurrentWork() is called, the class will first call to interrupt the running task.

Parameters:

interruptIfStopped: Set to true to allow the system to interrupt actively running work.

public boolean isStopped()

Returns true if JobIntentService.onStopCurrentWork() has been called. You can use this, while executing your work, to see if it should be stopped.

public boolean onStopCurrentWork()

This will be called if the JobScheduler has decided to stop this job. The job for this service does not have any constraints specified, so this will only generally happen if the service exceeds the job's maximum execution time.

Returns:

True to indicate to the JobManager whether you'd like to reschedule this work, false to drop this and all following work. Regardless of the value returned, your service must stop executing or the system will ultimately kill it. The default implementation returns true, and that is most likely what you want to return as well (so no work gets lost).

Source

/*
 * Copyright (C) 2017 The Android Open Source Project
 *
 * Licensed under the Apache License, Version 2.0 (the "License");
 * you may not use this file except in compliance with the License.
 * You may obtain a copy of the License at
 *
 *      http://www.apache.org/licenses/LICENSE-2.0
 *
 * Unless required by applicable law or agreed to in writing, software
 * distributed under the License is distributed on an "AS IS" BASIS,
 * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
 * See the License for the specific language governing permissions and
 * limitations under the License.
 */

package androidx.core.app;

import android.app.Service;
import android.app.job.JobInfo;
import android.app.job.JobParameters;
import android.app.job.JobScheduler;
import android.app.job.JobServiceEngine;
import android.app.job.JobWorkItem;
import android.content.ComponentName;
import android.content.Context;
import android.content.Intent;
import android.os.Build;
import android.os.IBinder;
import android.os.PowerManager;
import android.util.Log;

import androidx.annotation.NonNull;
import androidx.annotation.Nullable;
import androidx.annotation.RequiresApi;

import java.util.ArrayList;
import java.util.HashMap;

/**
 * Helper for processing work that has been enqueued for a job/service.  When running on
 * {@link android.os.Build.VERSION_CODES#O Android O} or later, the work will be dispatched
 * as a job via {@link android.app.job.JobScheduler#enqueue JobScheduler.enqueue}.  When running
 * on older versions of the platform, it will use
 * {@link android.content.Context#startService Context.startService}.
 *
 * <p>You must publish your subclass in your manifest for the system to interact with.  This
 * should be published as a {@link android.app.job.JobService}, as described for that class,
 * since on O and later platforms it will be executed that way.</p>
 *
 * <p>Use {@link #enqueueWork(Context, Class, int, Intent)} to enqueue new work to be
 * dispatched to and handled by your service.  It will be executed in
 * {@link #onHandleWork(Intent)}.</p>
 *
 * <p>You do not need to use {@link androidx.legacy.content.WakefulBroadcastReceiver}
 * when using this class.  When running on {@link android.os.Build.VERSION_CODES#O Android O},
 * the JobScheduler will take care of wake locks for you (holding a wake lock from the time
 * you enqueue work until the job has been dispatched and while it is running).  When running
 * on previous versions of the platform, this wake lock handling is emulated in the class here
 * by directly calling the PowerManager; this means the application must request the
 * {@link android.Manifest.permission#WAKE_LOCK} permission.</p>
 *
 * <p>There are a few important differences in behavior when running on
 * {@link android.os.Build.VERSION_CODES#O Android O} or later as a Job vs. pre-O:</p>
 *
 * <ul>
 *     <li><p>When running as a pre-O service, the act of enqueueing work will generally start
 *     the service immediately, regardless of whether the device is dozing or in other
 *     conditions.  When running as a Job, it will be subject to standard JobScheduler
 *     policies for a Job with a {@link android.app.job.JobInfo.Builder#setOverrideDeadline(long)}
 *     of 0: the job will not run while the device is dozing, it may get delayed more than
 *     a service if the device is under strong memory pressure with lots of demand to run
 *     jobs.</p></li>
 *     <li><p>When running as a pre-O service, the normal service execution semantics apply:
 *     the service can run indefinitely, though the longer it runs the more likely the system
 *     will be to outright kill its process, and under memory pressure one should expect
 *     the process to be killed even of recently started services.  When running as a Job,
 *     the typical {@link android.app.job.JobService} execution time limit will apply, after
 *     which the job will be stopped (cleanly, not by killing the process) and rescheduled
 *     to continue its execution later.  Job are generally not killed when the system is
 *     under memory pressure, since the number of concurrent jobs is adjusted based on the
 *     memory state of the device.</p></li>
 * </ul>
 *
 * <p>Here is an example implementation of this class:</p>
 *
 * {@sample frameworks/support/samples/Support4Demos/src/main/java/com/example/android/supportv4/app/SimpleJobIntentService.java
 *      complete}
 * @deprecated This class has been deprecated in favor of the Android Jetpack
 * <a href="https://developer.android.com/topic/libraries/architecture/workmanager>WorkManager</a>
 * library, which makes it easy to schedule deferrable, asynchronous tasks that are expected to run
 * even if the app exits or the device restarts.
 */
@Deprecated
public abstract class JobIntentService extends Service {
    static final String TAG = "JobIntentService";

    static final boolean DEBUG = false;

    CompatJobEngine mJobImpl;
    WorkEnqueuer mCompatWorkEnqueuer;
    CommandProcessor mCurProcessor;
    boolean mInterruptIfStopped = false;
    boolean mStopped = false;
    boolean mDestroyed = false;

    final ArrayList<CompatWorkItem> mCompatQueue;

    static final Object sLock = new Object();
    static final HashMap<ComponentName, WorkEnqueuer> sClassWorkEnqueuer = new HashMap<>();

    /**
     * Base class for the target service we can deliver work to and the implementation of
     * how to deliver that work.
     */
    abstract static class WorkEnqueuer {
        final ComponentName mComponentName;

        boolean mHasJobId;
        int mJobId;

        WorkEnqueuer(ComponentName cn) {
            mComponentName = cn;
        }

        void ensureJobId(int jobId) {
            if (!mHasJobId) {
                mHasJobId = true;
                mJobId = jobId;
            } else if (mJobId != jobId) {
                throw new IllegalArgumentException("Given job ID " + jobId
                        + " is different than previous " + mJobId);
            }
        }

        abstract void enqueueWork(Intent work);

        public void serviceStartReceived() {
        }

        public void serviceProcessingStarted() {
        }

        public void serviceProcessingFinished() {
        }
    }

    /**
     * Get rid of lint warnings about API levels.
     */
    interface CompatJobEngine {
        IBinder compatGetBinder();
        GenericWorkItem dequeueWork();
    }

    /**
     * An implementation of WorkEnqueuer that works for pre-O (raw Service-based).
     */
    static final class CompatWorkEnqueuer extends WorkEnqueuer {
        private final Context mContext;
        private final PowerManager.WakeLock mLaunchWakeLock;
        private final PowerManager.WakeLock mRunWakeLock;
        boolean mLaunchingService;
        boolean mServiceProcessing;

        CompatWorkEnqueuer(Context context, ComponentName cn) {
            super(cn);
            mContext = context.getApplicationContext();
            // Make wake locks.  We need two, because the launch wake lock wants to have
            // a timeout, and the system does not do the right thing if you mix timeout and
            // non timeout (or even changing the timeout duration) in one wake lock.
            PowerManager pm = ((PowerManager) context.getSystemService(Context.POWER_SERVICE));
            mLaunchWakeLock = pm.newWakeLock(PowerManager.PARTIAL_WAKE_LOCK,
                    cn.getClassName() + ":launch");
            mLaunchWakeLock.setReferenceCounted(false);
            mRunWakeLock = pm.newWakeLock(PowerManager.PARTIAL_WAKE_LOCK,
                    cn.getClassName() + ":run");
            mRunWakeLock.setReferenceCounted(false);
        }

        @Override
        void enqueueWork(Intent work) {
            Intent intent = new Intent(work);
            intent.setComponent(mComponentName);
            if (DEBUG) Log.d(TAG, "Starting service for work: " + work);
            if (mContext.startService(intent) != null) {
                synchronized (this) {
                    if (!mLaunchingService) {
                        mLaunchingService = true;
                        if (!mServiceProcessing) {
                            // If the service is not already holding the wake lock for
                            // itself, acquire it now to keep the system running until
                            // we get this work dispatched.  We use a timeout here to
                            // protect against whatever problem may cause it to not get
                            // the work.
                            mLaunchWakeLock.acquire(60 * 1000);
                        }
                    }
                }
            }
        }

        @Override
        public void serviceStartReceived() {
            synchronized (this) {
                // Once we have started processing work, we can count whatever last
                // enqueueWork() that happened as handled.
                mLaunchingService = false;
            }
        }

        @Override
        public void serviceProcessingStarted() {
            synchronized (this) {
                // We hold the wake lock as long as the service is processing commands.
                if (!mServiceProcessing) {
                    mServiceProcessing = true;
                    // Keep the device awake, but only for at most 10 minutes at a time
                    // (Similar to JobScheduler.)
                    mRunWakeLock.acquire(10 * 60 * 1000L);
                    mLaunchWakeLock.release();
                }
            }
        }

        @Override
        public void serviceProcessingFinished() {
            synchronized (this) {
                if (mServiceProcessing) {
                    // If we are transitioning back to a wakelock with a timeout, do the same
                    // as if we had enqueued work without the service running.
                    if (mLaunchingService) {
                        mLaunchWakeLock.acquire(60 * 1000);
                    }
                    mServiceProcessing = false;
                    mRunWakeLock.release();
                }
            }
        }
    }

    /**
     * Implementation of a JobServiceEngine for interaction with JobIntentService.
     */
    @RequiresApi(26)
    static final class JobServiceEngineImpl extends JobServiceEngine
            implements JobIntentService.CompatJobEngine {
        static final String TAG = "JobServiceEngineImpl";

        static final boolean DEBUG = false;

        final JobIntentService mService;
        final Object mLock = new Object();
        JobParameters mParams;

        final class WrapperWorkItem implements JobIntentService.GenericWorkItem {
            final JobWorkItem mJobWork;

            WrapperWorkItem(JobWorkItem jobWork) {
                mJobWork = jobWork;
            }

            @Override
            public Intent getIntent() {
                return mJobWork.getIntent();
            }

            @Override
            public void complete() {
                synchronized (mLock) {
                    if (mParams != null) {
                        mParams.completeWork(mJobWork);
                    }
                }
            }
        }

        JobServiceEngineImpl(JobIntentService service) {
            super(service);
            mService = service;
        }

        @Override
        public IBinder compatGetBinder() {
            return getBinder();
        }

        @Override
        public boolean onStartJob(JobParameters params) {
            if (DEBUG) Log.d(TAG, "onStartJob: " + params);
            mParams = params;
            // We can now start dequeuing work!
            mService.ensureProcessorRunningLocked(false);
            return true;
        }

        @Override
        public boolean onStopJob(JobParameters params) {
            if (DEBUG) Log.d(TAG, "onStopJob: " + params);
            boolean result = mService.doStopCurrentWork();
            synchronized (mLock) {
                // Once we return, the job is stopped, so its JobParameters are no
                // longer valid and we should not be doing anything with them.
                mParams = null;
            }
            return result;
        }

        /**
         * Dequeue some work.
         */
        @Override
        public JobIntentService.GenericWorkItem dequeueWork() {
            JobWorkItem work;
            synchronized (mLock) {
                if (mParams == null) {
                    return null;
                }
                work = mParams.dequeueWork();
            }
            if (work != null) {
                work.getIntent().setExtrasClassLoader(mService.getClassLoader());
                return new WrapperWorkItem(work);
            } else {
                return null;
            }
        }
    }

    @RequiresApi(26)
    static final class JobWorkEnqueuer extends JobIntentService.WorkEnqueuer {
        private final JobInfo mJobInfo;
        private final JobScheduler mJobScheduler;

        JobWorkEnqueuer(Context context, ComponentName cn, int jobId) {
            super(cn);
            ensureJobId(jobId);
            JobInfo.Builder b = new JobInfo.Builder(jobId, mComponentName);
            mJobInfo = b.setOverrideDeadline(0).build();
            mJobScheduler = (JobScheduler) context.getApplicationContext().getSystemService(
                    Context.JOB_SCHEDULER_SERVICE);
        }

        @Override
        void enqueueWork(Intent work) {
            if (DEBUG) Log.d(TAG, "Enqueueing work: " + work);
            mJobScheduler.enqueue(mJobInfo, new JobWorkItem(work));
        }
    }

    /**
     * Abstract definition of an item of work that is being dispatched.
     */
    interface GenericWorkItem {
        Intent getIntent();
        void complete();
    }

    /**
     * An implementation of GenericWorkItem that dispatches work for pre-O platforms: intents
     * received through a raw service's onStartCommand.
     */
    final class CompatWorkItem implements GenericWorkItem {
        final Intent mIntent;
        final int mStartId;

        CompatWorkItem(Intent intent, int startId) {
            mIntent = intent;
            mStartId = startId;
        }

        @Override
        public Intent getIntent() {
            return mIntent;
        }

        @Override
        public void complete() {
            if (DEBUG) Log.d(TAG, "Stopping self: #" + mStartId);
            stopSelf(mStartId);
        }
    }

    /**
     * This is a task to dequeue and process work in the background.
     */
    final class CommandProcessor extends android.os.AsyncTask<Void, Void, Void> {
        @Override
        protected Void doInBackground(Void... params) {
            GenericWorkItem work;

            if (DEBUG) Log.d(TAG, "Starting to dequeue work...");

            while ((work = dequeueWork()) != null) {
                if (DEBUG) Log.d(TAG, "Processing next work: " + work);
                onHandleWork(work.getIntent());
                if (DEBUG) Log.d(TAG, "Completing work: " + work);
                work.complete();
            }

            if (DEBUG) Log.d(TAG, "Done processing work!");

            return null;
        }

        @Override
        protected void onCancelled(Void aVoid) {
            processorFinished();
        }

        @Override
        protected void onPostExecute(Void aVoid) {
            processorFinished();
        }
    }

    /**
     * Default empty constructor.
     */
    public JobIntentService() {
        if (Build.VERSION.SDK_INT >= 26) {
            mCompatQueue = null;
        } else {
            mCompatQueue = new ArrayList<>();
        }
    }

    @Override
    public void onCreate() {
        super.onCreate();
        if (DEBUG) Log.d(TAG, "CREATING: " + this);
        if (Build.VERSION.SDK_INT >= 26) {
            mJobImpl = new JobServiceEngineImpl(this);
            mCompatWorkEnqueuer = null;
        } else {
            mJobImpl = null;
            ComponentName cn = new ComponentName(this, this.getClass());
            mCompatWorkEnqueuer = getWorkEnqueuer(this, cn, false, 0);
        }
    }

    /**
     * Processes start commands when running as a pre-O service, enqueueing them to be
     * later dispatched in {@link #onHandleWork(Intent)}.
     */
    @Override
    public int onStartCommand(@Nullable Intent intent, int flags, int startId) {
        if (mCompatQueue != null) {
            mCompatWorkEnqueuer.serviceStartReceived();
            if (DEBUG) Log.d(TAG, "Received compat start command #" + startId + ": " + intent);
            synchronized (mCompatQueue) {
                mCompatQueue.add(new CompatWorkItem(intent != null ? intent : new Intent(),
                        startId));
                ensureProcessorRunningLocked(true);
            }
            return START_REDELIVER_INTENT;
        } else {
            if (DEBUG) Log.d(TAG, "Ignoring start command: " + intent);
            return START_NOT_STICKY;
        }
    }

    /**
     * Returns the IBinder for the {@link android.app.job.JobServiceEngine} when
     * running as a JobService on O and later platforms.
     */
    @Override
    public IBinder onBind(@NonNull Intent intent) {
        if (mJobImpl != null) {
            IBinder engine = mJobImpl.compatGetBinder();
            if (DEBUG) Log.d(TAG, "Returning engine: " + engine);
            return engine;
        } else {
            return null;
        }
    }

    @Override
    public void onDestroy() {
        super.onDestroy();
        if (mCompatQueue != null) {
            synchronized (mCompatQueue) {
                mDestroyed = true;
                mCompatWorkEnqueuer.serviceProcessingFinished();
            }
        }
    }

    /**
     * Call this to enqueue work for your subclass of {@link JobIntentService}.  This will
     * either directly start the service (when running on pre-O platforms) or enqueue work
     * for it as a job (when running on O and later).  In either case, a wake lock will be
     * held for you to ensure you continue running.  The work you enqueue will ultimately
     * appear at {@link #onHandleWork(Intent)}.
     *
     * @param context Context this is being called from.
     * @param cls The concrete class the work should be dispatched to (this is the class that
     * is published in your manifest).
     * @param jobId A unique job ID for scheduling; must be the same value for all work
     * enqueued for the same class.
     * @param work The Intent of work to enqueue.
     */
    public static void enqueueWork(@NonNull Context context, @NonNull Class<?> cls, int jobId,
            @NonNull Intent work) {
        enqueueWork(context, new ComponentName(context, cls), jobId, work);
    }

    /**
     * Like {@link #enqueueWork(Context, Class, int, Intent)}, but supplies a ComponentName
     * for the service to interact with instead of its class.
     *
     * @param context Context this is being called from.
     * @param component The published ComponentName of the class this work should be
     * dispatched to.
     * @param jobId A unique job ID for scheduling; must be the same value for all work
     * enqueued for the same class.
     * @param work The Intent of work to enqueue.
     */
    public static void enqueueWork(@NonNull Context context, @NonNull ComponentName component,
            int jobId, @NonNull Intent work) {
        if (work == null) {
            throw new IllegalArgumentException("work must not be null");
        }
        synchronized (sLock) {
            WorkEnqueuer we = getWorkEnqueuer(context, component, true, jobId);
            we.ensureJobId(jobId);
            we.enqueueWork(work);
        }
    }

    static WorkEnqueuer getWorkEnqueuer(Context context, ComponentName cn, boolean hasJobId,
            int jobId) {
        WorkEnqueuer we = sClassWorkEnqueuer.get(cn);
        if (we == null) {
            if (Build.VERSION.SDK_INT >= 26) {
                if (!hasJobId) {
                    throw new IllegalArgumentException("Can't be here without a job id");
                }
                we = new JobWorkEnqueuer(context, cn, jobId);
            } else {
                we = new CompatWorkEnqueuer(context, cn);
            }
            sClassWorkEnqueuer.put(cn, we);
        }
        return we;
    }

    /**
     * Called serially for each work dispatched to and processed by the service.  This
     * method is called on a background thread, so you can do long blocking operations
     * here.  Upon returning, that work will be considered complete and either the next
     * pending work dispatched here or the overall service destroyed now that it has
     * nothing else to do.
     *
     * <p>Be aware that when running as a job, you are limited by the maximum job execution
     * time and any single or total sequential items of work that exceeds that limit will
     * cause the service to be stopped while in progress and later restarted with the
     * last unfinished work.  (There is currently no limit on execution duration when
     * running as a pre-O plain Service.)</p>
     *
     * @param intent The intent describing the work to now be processed.
     */
    protected abstract void onHandleWork(@NonNull Intent intent);

    /**
     * Control whether code executing in {@link #onHandleWork(Intent)} will be interrupted
     * if the job is stopped.  By default this is false.  If called and set to true, any
     * time {@link #onStopCurrentWork()} is called, the class will first call
     * {@link android.os.AsyncTask#cancel(boolean) AsyncTask.cancel(true)} to interrupt the running
     * task.
     *
     * @param interruptIfStopped Set to true to allow the system to interrupt actively
     * running work.
     */
    public void setInterruptIfStopped(boolean interruptIfStopped) {
        mInterruptIfStopped = interruptIfStopped;
    }

    /**
     * Returns true if {@link #onStopCurrentWork()} has been called.  You can use this,
     * while executing your work, to see if it should be stopped.
     */
    public boolean isStopped() {
        return mStopped;
    }

    /**
     * This will be called if the JobScheduler has decided to stop this job.  The job for
     * this service does not have any constraints specified, so this will only generally happen
     * if the service exceeds the job's maximum execution time.
     *
     * @return True to indicate to the JobManager whether you'd like to reschedule this work,
     * false to drop this and all following work. Regardless of the value returned, your service
     * must stop executing or the system will ultimately kill it.  The default implementation
     * returns true, and that is most likely what you want to return as well (so no work gets
     * lost).
     */
    public boolean onStopCurrentWork() {
        return true;
    }

    boolean doStopCurrentWork() {
        if (mCurProcessor != null) {
            mCurProcessor.cancel(mInterruptIfStopped);
        }
        mStopped = true;
        return onStopCurrentWork();
    }

    @SuppressWarnings({"deprecation", "ObjectToString"}) /* AsyncTask */
    void ensureProcessorRunningLocked(boolean reportStarted) {
        if (mCurProcessor == null) {
            mCurProcessor = new CommandProcessor();
            if (mCompatWorkEnqueuer != null && reportStarted) {
                mCompatWorkEnqueuer.serviceProcessingStarted();
            }
            if (DEBUG) Log.d(TAG, "Starting processor: " + mCurProcessor);
            mCurProcessor.executeOnExecutor(android.os.AsyncTask.THREAD_POOL_EXECUTOR);
        }
    }

    void processorFinished() {
        if (mCompatQueue != null) {
            synchronized (mCompatQueue) {
                mCurProcessor = null;
                // The async task has finished, but we may have gotten more work scheduled in the
                // meantime.  If so, we need to restart the new processor to execute it.  If there
                // is no more work at this point, either the service is in the process of being
                // destroyed (because we called stopSelf on the last intent started for it), or
                // someone has already called startService with a new Intent that will be
                // arriving shortly.  In either case, we want to just leave the service
                // waiting -- either to get destroyed, or get a new onStartCommand() callback
                // which will then kick off a new processor.
                if (mCompatQueue != null && mCompatQueue.size() > 0) {
                    ensureProcessorRunningLocked(false);
                } else if (!mDestroyed) {
                    mCompatWorkEnqueuer.serviceProcessingFinished();
                }
            }
        }
    }

    GenericWorkItem dequeueWork() {
        if (mJobImpl != null) {
            return mJobImpl.dequeueWork();
        } else {
            synchronized (mCompatQueue) {
                if (mCompatQueue.size() > 0) {
                    return mCompatQueue.remove(0);
                } else {
                    return null;
                }
            }
        }
    }
}