Android学习的第三周笔记

1. Handler

  • 可以定时执行Message&MessageQueue
  • 在不同线程中执行

A Handler allows you to send and process Message and Runnable objects associated with a thread's MessageQueue. Each Handler instance is associated with a single thread and that thread's message queue. When you create a new Handler, it is bound to the thread / message queue of the thread that is creating it -- from that point on, it will deliver messages and runnables to that message queue and execute them as they come out of the message queue.

There are two main uses for a Handler: (1) to schedule messages and runnables to be executed as some point in the future; and (2) to enqueue an action to be performed on a different thread than your own.

Scheduling messages is accomplished with the post(Runnable)postAtTime(Runnable, long)postDelayed(Runnable, long)sendEmptyMessage(int),sendMessage(Message)sendMessageAtTime(Message, long), and sendMessageDelayed(Message, long) methods. The post versions allow you to enqueue Runnable objects to be called by the message queue when they are received; the sendMessage versions allow you to enqueue a Message object containing a bundle of data that will be processed by the Handler's handleMessage(Message) method (requiring that you implement a subclass of Handler).

When posting or sending to a Handler, you can either allow the item to be processed as soon as the message queue is ready to do so, or specify a delay before it gets processed or absolute time for it to be processed. The latter two allow you to implement timeouts, ticks, and other timing-based behavior.

When a process is created for your application, its main thread is dedicated to running a message queue that takes care of managing the top-level application objects (activities, broadcast receivers, etc) and any windows they create. You can create your own threads, and communicate back with the main application thread through a Handler. This is done by calling the same post or sendMessage methods as before, but from your new thread. The given Runnable or Message will then be scheduled in the Handler's message queue and processed when appropriate.

(1). Message

Defines a message containing a description and arbitrary data object that can be sent to a Handler. This object contains two extra int fields and an extra object field that allow you to not do allocations in many cases.

(2).MessageQueue

Low-level class holding the list of messages to be dispatched by a Looper. Messages are not added directly to a MessageQueue, but rather throughHandler objects associated with the Looper.

You can retrieve the MessageQueue for the current thread with Looper.myQueue().

  • Message的队列
  • 每个线程最多只能拥有一个
  • Thread-->Looper-->MessageQueue

(3).Looper

Class used to run a message loop for a thread. Threads by default do not have a message loop associated with them; to create one, call prepare() in the thread that is to run the loop, and then loop() to have it process messages until the loop is stopped.

Most interaction with a message loop is through the Handler class.

  • Looper从MessageQueue中取出Message
  • 交由Handler的HandlerMessage处理
  • 调用Message.recycle将其放入Message Pool中

2.Service

A Service is an application component representing either an application's desire to perform a longer-running operation while not interacting with the user or to supply functionality for other applications to use. Each service class must have a corresponding <service> declaration in its package'sAndroidManifest.xml. Services can be started with Context.startService() and Context.bindService().

Note that services, like other application objects, run in the main thread of their hosting process. This means that, if your service is going to do any CPU intensive (such as MP3 playback) or blocking (such as networking) operations, it should spawn its own thread in which to do that work. More information on this can be found in Processes and Threads. The IntentService class is available as a standard implementation of Service that has its own thread where it schedules its work to be done.

  • A Service is not a separate process. The Service object itself does not imply it is running in its own process; unless otherwise specified, it runs in the same process as the application it is part of.
  • A Service is not a thread. It is not a means itself to do work off of the main thread (to avoid Application Not Responding errors).

Thus a Service itself is actually very simple, providing two main features:

  • A facility for the application to tell the system about something it wants to be doing in the background (even when the user is not directly interacting with the application). This corresponds to calls to Context.startService(), which ask the system to schedule work for the service, to be run until the service or someone else explicitly stop it.
  • A facility for an application to expose some of its functionality to other applications. This corresponds to calls to Context.bindService(), which allows a long-standing connection to be made to the service in order to interact with it.

(1).Service Lifecycle

There are two reasons that a service can be run by the system. If someone calls Context.startService() then the system will retrieve the service (creating it and calling its onCreate() method if needed) and then call its onStartCommand(Intent, int, int) method with the arguments supplied by the client. The service will at this point continue running until Context.stopService() or stopSelf() is called. Note that multiple calls to Context.startService() do not nest (though they do result in multiple corresponding calls to onStartCommand()), so no matter how many times it is started a service will be stopped once Context.stopService() or stopSelf() is called; however, services can use their stopSelf(int) method to ensure the service is not stopped until started intents have been processed.

For started services, there are two additional major modes of operation they can decide to run in, depending on the value they return from onStartCommand(): START_STICKY is used for services that are explicitly started and stopped as needed, while START_NOT_STICKY orSTART_REDELIVER_INTENT are used for services that should only remain running while processing any commands sent to them. See the linked documentation for more detail on the semantics.

Clients can also use Context.bindService() to obtain a persistent connection to a service. This likewise creates the service if it is not already running (calling onCreate() while doing so), but does not call onStartCommand(). The client will receive the IBinder object that the service returns from itsonBind(Intent) method, allowing the client to then make calls back to the service. The service will remain running as long as the connection is established (whether or not the client retains a reference on the service's IBinder). Usually the IBinder returned is for a complex interface that has beenwritten in aidl.

A service can be both started and have connections bound to it. In such a case, the system will keep the service running as long as either it is started orthere are one or more connections to it with the Context.BIND_AUTO_CREATE flag. Once neither of these situations hold, the service's onDestroy() method is called and the service is effectively terminated. All cleanup (stopping threads, unregistering receivers) should be complete upon returning from onDestroy().

(2).Permissions

Global access to a service can be enforced when it is declared in its manifest's <service> tag. By doing so, other applications will need to declare a corresponding <uses-permission> element in their own manifest to be able to start, stop, or bind to the service.

As of GINGERBREAD, when using Context.startService(Intent), you can also set Intent.FLAG_GRANT_READ_URI_PERMISSION and/orIntent.FLAG_GRANT_WRITE_URI_PERMISSION on the Intent. This will grant the Service temporary access to the specific URIs in the Intent. Access will remain until the Service has called stopSelf(int) for that start command or a later one, or until the Service has been completely stopped. This works for granting access to the other apps that have not requested the permission protecting the Service, or even when the Service is not exported at all.

In addition, a service can protect individual IPC calls into it with permissions, by calling the checkCallingPermission(String) method before executing the implementation of that call.



3.BroadCastReceiver

Base class for code that will receive intents sent by sendBroadcast().

If you don't need to send broadcasts across applications, consider using this class with LocalBroadcastManager instead of the more general facilities described below. This will give you a much more efficient implementation (no cross-process communication needed) and allow you to avoid thinking about any security issues related to other applications being able to receive or send your broadcasts.

You can either dynamically register an instance of this class with Context.registerReceiver() or statically publish an implementation through the<receiver> tag in your AndroidManifest.xml.

Note:    If registering a receiver in your Activity.onResume() implementation, you should unregister it in Activity.onPause(). (You won't receive intents when paused, and this will cut down on unnecessary system overhead). Do not unregister in Activity.onSaveInstanceState(), because this won't be called if the user moves back in the history stack.

There are two major classes of broadcasts that can be received:

  • Normal broadcasts (sent with Context.sendBroadcast) are completely asynchronous. All receivers of the broadcast are run in an undefined order, often at the same time. This is more efficient, but means that receivers cannot use the result or abort APIs included here.
  • Ordered broadcasts (sent with Context.sendOrderedBroadcast) are delivered to one receiver at a time. As each receiver executes in turn, it can propagate a result to the next receiver, or it can completely abort the broadcast so that it won't be passed to other receivers. The order receivers run in can be controlled with the android:priority attribute of the matching intent-filter; receivers with the same priority will be run in an arbitrary order.

Even in the case of normal broadcasts, the system may in some situations revert to delivering the broadcast one receiver at a time. In particular, for receivers that may require the creation of a process, only one will be run at a time to avoid overloading the system with new processes. In this situation, however, the non-ordered semantics hold: these receivers still cannot return results or abort their broadcast.

Note that, although the Intent class is used for sending and receiving these broadcasts, the Intent broadcast mechanism here is completely separate from Intents that are used to start Activities with Context.startActivity(). There is no way for a BroadcastReceiver to see or capture Intents used with startActivity(); likewise, when you broadcast an Intent, you will never find or start an Activity. These two operations are semantically very different: starting an Activity with an Intent is a foreground operation that modifies what the user is currently interacting with; broadcasting an Intent is a background operation that the user is not normally aware of.

(1).Security

Receivers used with the Context APIs are by their nature a cross-application facility, so you must consider how other applications may be able to abuse your use of them. Some things to consider are:

The Intent namespace is global. Make sure that Intent action names and other strings are written in a namespace you own, or else you may inadvertently conflict with other applications.

When you use registerReceiver(BroadcastReceiver, IntentFilter)any application may send broadcasts to that registered receiver. You can control who can send broadcasts to it through permissions described below.

When you publish a receiver in your application's manifest and specify intent-filters for it, any other application can send broadcasts to it regardless of the filters you specify. To prevent others from sending to it, make it unavailable to them with android:exported="false".

When you use sendBroadcast(Intent) or related methods, normally any other application can receive these broadcasts. You can control who can receive such broadcasts through permissions described below. Alternatively, starting with ICE_CREAM_SANDWICH, you can also safely restrict the broadcast to a single application with Intent.setPackage

None of these issues exist when using LocalBroadcastManager, since intents broadcast it never go outside of the current process.

Access permissions can be enforced by either the sender or receiver of a broadcast.

To enforce a permission when sending, you supply a non-null permission argument to sendBroadcast(Intent, String) orsendOrderedBroadcast(Intent, String, BroadcastReceiver, android.os.Handler, int, String, Bundle). Only receivers who have been granted this permission (by requesting it with the <uses-permission> tag in their AndroidManifest.xml) will be able to receive the broadcast.

To enforce a permission when receiving, you supply a non-null permission when registering your receiver -- either when callingregisterReceiver(BroadcastReceiver, IntentFilter, String, android.os.Handler) or in the static <receiver> tag in your AndroidManifest.xml. Only broadcasters who have been granted this permission (by requesting it with the <uses-permission> tag in their AndroidManifest.xml) will be able to send an Intent to the receiver.

(2).Receiver Lifecycle

A BroadcastReceiver object is only valid for the duration of the call to onReceive(Context, Intent). Once your code returns from this function, the system considers the object to be finished and no longer active.

This has important repercussions to what you can do in an onReceive(Context, Intent) implementation: anything that requires asynchronous operation is not available, because you will need to return from the function to handle the asynchronous operation, but at that point the BroadcastReceiver is no longer active and thus the system is free to kill its process before the asynchronous operation completes.

In particular, you may not show a dialog or bind to a service from within a BroadcastReceiver. For the former, you should instead use theNotificationManager API. For the latter, you can use Context.startService() to send a command to the service.

(3).Process Lifecycle

A process that is currently executing a BroadcastReceiver (that is, currently running the code in its onReceive(Context, Intent) method) is considered to be a foreground process and will be kept running by the system except under cases of extreme memory pressure.

Once you return from onReceive(), the BroadcastReceiver is no longer active, and its hosting process is only as important as any other application components that are running in it. This is especially important because if that process was only hosting the BroadcastReceiver (a common case for applications that the user has never or not recently interacted with), then upon returning from onReceive() the system will consider its process to be empty and aggressively kill it so that resources are available for other more important processes.

This means that for longer-running operations you will often use a Service in conjunction with a BroadcastReceiver to keep the containing process active for the entire time of your operation.










  • 0
    点赞
  • 0
    收藏
    觉得还不错? 一键收藏
  • 0
    评论
目标检测(Object Detection)是计算机视觉领域的一个核心问题,其主要任务是找出图像中所有感兴趣的目标(物体),并确定它们的类别和位置。以下是对目标检测的详细阐述: 一、基本概念 目标检测的任务是解决“在哪里?是什么?”的问题,即定位出图像中目标的位置并识别出目标的类别。由于各类物体具有不同的外观、形状和姿态,加上成像时光照、遮挡等因素的干扰,目标检测一直是计算机视觉领域最具挑战性的任务之一。 二、核心问题 目标检测涉及以下几个核心问题: 分类问题:判断图像中的目标属于哪个类别。 定位问题:确定目标在图像中的具体位置。 大小问题:目标可能具有不同的大小。 形状问题:目标可能具有不同的形状。 三、算法分类 基于深度学习的目标检测算法主要分为两大类: Two-stage算法:先进行区域生成(Region Proposal),生成有可能包含待检物体的预选框(Region Proposal),再通过卷积神经网络进行样本分类。常见的Two-stage算法包括R-CNN、Fast R-CNN、Faster R-CNN等。 One-stage算法:不用生成区域提议,直接在网络中提取特征来预测物体分类和位置。常见的One-stage算法包括YOLO系列(YOLOv1、YOLOv2、YOLOv3、YOLOv4、YOLOv5等)、SSD和RetinaNet等。 四、算法原理 以YOLO系列为例,YOLO将目标检测视为回归问题,将输入图像一次性划分为多个区域,直接在输出层预测边界框和类别概率。YOLO采用卷积网络来提取特征,使用全连接层来得到预测值。其网络结构通常包含多个卷积层和全连接层,通过卷积层提取图像特征,通过全连接层输出预测结果。 五、应用领域 目标检测技术已经广泛应用于各个领域,为人们的生活带来了极大的便利。以下是一些主要的应用领域: 安全监控:在商场、银行
评论
添加红包

请填写红包祝福语或标题

红包个数最小为10个

红包金额最低5元

当前余额3.43前往充值 >
需支付:10.00
成就一亿技术人!
领取后你会自动成为博主和红包主的粉丝 规则
hope_wisdom
发出的红包
实付
使用余额支付
点击重新获取
扫码支付
钱包余额 0

抵扣说明:

1.余额是钱包充值的虚拟货币,按照1:1的比例进行支付金额的抵扣。
2.余额无法直接购买下载,可以购买VIP、付费专栏及课程。

余额充值