概要

Association, Aggregation and Composition are terms that represent relationships among objects. They are very basic stuff of Object Oriented Programming.

关联

Association is a relationship among the objects. Association is "*a*" relationship among objects. In Association, the relationship among the objects determine what an object instance can cause another to perform an action on its behalf. We can also say that an association defines the multiplicity among the objects. We can define a one-to-one, one-to-many, many-to-one and many-to-many relationship among objects. Association is a more general term to define a relationship among objects. Association means that an object "uses" another object.
For example Managers and Employees, multiple employees may be associated with a single manager and a single employee may be associated with multiple managers.

聚合

Aggregation is a special type of Association. Aggregation is "*the*" relationship among objects. We can say it is a direct association among the objects. In Aggregation, the direction specifies which object contains the other object. There are mutual dependencies among objects.
For example, departments and employees, a department has many employees but a single employee is not associated with multiple departments.
UML Representation of Aggregation (white diamond):

The UML representation of the example above (relation between employee and department):

Here, the lives of both objects are independent of each other. That means that in this Association (Aggregation) the object has their own life cycle. Employees may exist without a department. Here, department can be called an owner object and the employee can be called a child object. The owner and child objects cannot belong to a different parent object.

组合(复合)

Composition is special type of Aggregation. It is a strong type of Aggregation. In this type of Aggregation the child object does not have their own life cycle. The child object's life depends on the parent's life cycle. Only the parent object has an independent life cycle. If we delete the parent object then the child object(s) will also be deleted. We can define the Composition as a "Part of" relationship.
For example, the company and company location, a single company has multiple locations. If we delete the company then all the company locations are automatically deleted. The company location does not have their independent life cycle, it depends on the company object's life (parent object).
UML Representation of Composition (black diamond):

UML representation of the example above (relation between Company and Company Location):

Here, the lives of both objects are not independent. The life of the company location object can be determined by the life of the company object. The company object is responsible for creating and destroying company location objects.

关联,聚合和组合三者之间的关系图


Aggregation and Composition are a special type of Association. Composition is again a special type of Aggregation. We can define Aggregation and Composition as "has a" relationships. Composition is more restrictive or more specific. In Composition, composed objects cannot exist without the other object. This type of restriction does not exist in Aggregations. In Aggregation, the existence of a composed object is optional. In Aggregation, the child object can exist beyond the life cycle of its parent whereas in Composition the child object cannot exist beyond the life cycle of its parent.
Combined example of Aggregation and Composition:


聚合VS组合

Aggregation Composition
Aggregation is a special type of Association. Composition is a special type of Aggregation.
All objects have their own life cycle. In Composition, the child object does not have their own life cycle and it depends on the parent's life cycle.
A parent class is not responsible for creating or destroying the child class. The parent class is responsible for creating or destroying the child class.
Aggregation can be described as a "Has-a" relationship. Composition can be described as a "Has-a" relationship as well as a "Part of" relationship, but here the difference is the length of the relationship among the objects.
Aggregation is a weak Association. Composition is a strong Association.
Aggregation means one object is the owner of another object. Composition means one object is contained in another object.
The direction of a relation is a requirement in both Composition and Aggregation. The direction specifies which object contains the other one.
Both have a single direction of association.
Both have a single owner.

结论

These three terms are more important in the object oriented world. They denote or represent the relations among objects. If you are confused or unable to decide whether a specific relation best describes an Association, Aggregation or Composition then it can be decribed as an Association.

引用:http://www.c-sharpcorner.com/UploadFile/ff2f08/association-aggregation-and-composition/

关联,聚合和组合(复合)--Association, Aggregation and Composition的更多相关文章

  1. UML类关系:依赖,关联,聚合和组合

    UML图示例:(可使用StartUML来画图,小巧^_^) http://www.blogjava.net/lukangping/archive/2010/08/01/327693.html 聚合:表 ...

  2. Inheritance, Association, Aggregation, and Composition 类的继承,关联,聚合和组合的区别

    在C++中,类与类之间的关系大概有四种,分别为继承,关联,聚合,和组合.其中继承我们大家应该都比较熟悉,因为是C++的三大特性继承Inheritance,封装Encapsulation,和多态Poly ...

  3. UML类关系(依赖,关联,聚合,组合,泛化,实现)

    转自https://blog.csdn.net/k346k346/article/details/59582926   在学习面向对象设计时,类关系涉及依赖.关联.聚合.组合和泛化这五种关系,耦合度依 ...

  4. UML类图关系(泛化 、继承、实现、依赖、关联、聚合、组合)

    UML类图关系(泛化 .继承.实现.依赖.关联.聚合.组合) 继承.实现.依赖.关联.聚合.组合的联系与区别 分别介绍这几种关系: 继承 指的是一个类(称为子类.子接口)继承另外的一个类(称为父类.父 ...

  5. java--依赖、关联、聚合和组合之间区别的理解

    在学习面向对象设计对象关系时,依赖.关联.聚合和组合这四种关系之间区别比较容易混淆.特别是后三种,仅仅是在语义上有所区别,所谓语义就是指上下文环境.特定情景等. 依赖(Dependency)关系是类与 ...

  6. java_UML:继承/泛化、实现、依赖、关联、聚合、组合的联系与区别 (2016-07-12)

    分别介绍这几种关系: UML关系:继承(泛化).实现.依赖.关联.聚合.组合的联系与区别 一.表示符号上的区别 二.具体区别与联系 1. 继承/泛化(Generalization) [泛化关系]:是一 ...

  7. UML[1] UML类图关系(泛化 、继承、实现、依赖、关联、聚合、组合)(转)

    转自:http://blog.csdn.net/zhaoxu0312/article/details/7212152 继承.实现.依赖.关联.聚合.组合的联系与区别 分别介绍这几种关系: 继承 指的是 ...

  8. UML图中类之间的关系:依赖,泛化,关联,聚合,组合,实现

    UML图中类之间的关系:依赖,泛化,关联,聚合,组合,实现 类与类图 1) 类(Class)封装了数据和行为,是面向对象的重要组成部分,它是具有相同属性.操作.关系的对象集合的总称. 2) 在系统中, ...

  9. UML类图及依赖,泛化,关联,聚合,组合,实现

    UML图中类之间的关系:依赖,泛化,关联,聚合,组合,实现 类与类图 1) 类(Class)封装了数据和行为,是面向对象的重要组成部分,它是具有相同属性.操作.关系的对象集合的总称. 2) 在系统中, ...

随机推荐

  1. spark总结5 RDD

    创建RDD 有两种方式 1 通过hdfs支持的文件系统创建 RDD, RDD里面没有真正要计算的数据,只记录了一下元数据 2 从过scala集合或者数组以并行化的方式创建RDD collect 把结果 ...

  2. mongodb的使用(入门)

    1.登录mongodb ./bin/mongo 2.查看所有数据库 show dbs  ##默认有admin  和  local两个库 3.创建数据库 use test #创建数据库后,如果不写入数据 ...

  3. window7 共享wifi(不通过wifi软件)

    1.新建共享网络账号 管理员登录cmd输入:netsh wlan set hostednetwork mode=allow ssid=4Gtest key=12345678 ssid是无线网络名称.k ...

  4. Codeforces Round #412 div2 ABCD

    A 按rank给出每个人的赛前分数和赛后分数 如果一个人打败了比他分数高的人 他的分数必然升高 问比赛rated吗 如果一个人的分数改变了肯定rate 如果全都没改的话 也可能是rated 这时候ch ...

  5. JVM内存管理中的垃圾回收策略

    JVM垃圾回收策略 1.静态内存分配和回收 编译时已经确定了内存空间大小,程序被加载后则一次性分配好内存空间.程序结束后,则对应栈帧撤销,分配的静态内存空间则被回收. 2.动态内存分配和回收 程序运行 ...

  6. numpy常用函数之randn

    numpy中有一些常用的用来产生随机数的函数,randn就是其中一个,randn函数位于numpy.random中,函数原型如下: numpy.random.randn(d0, d1, ..., dn ...

  7. js 读取json数据 挖坑

    一般遇到后台给的json数据格式不对 比如key和value都是单引号. 但真正的json 的key和value都是双引号,必须双引号才能取值. 再来一个例子看看 var test = [{ &quo ...

  8. LeetCode OJ:Unique Binary Search Trees(唯一二叉搜索树)

    Given n, how many structurally unique BST's (binary search trees) that store values 1...n? For examp ...

  9. Python3.6.0安装

    1.安装 具体详情请参考下图: 双击安装包:   勾选“add  python 3.6 to PATH”这样可以自动生成环境变量,选择“Customize installation”自定义安装. 2. ...

  10. 2018.7.2 AK22 不良品分析

    a 电路 b 软件 STM32-工业炉温控制器程序 c . layout 不良分析: 1测各模块电压 12V  ok 5V  ok 3.3V  ok 2跟换MCU 给MCU烧程序 ok 3测量MCU晶 ...