【JAVA集合框架一 】java集合框架官方介绍 Collections Framework Overview 集合框架总览 翻译 javase8 集合官方文档中文版

原文链接:

https://docs.oracle.com/javase/8/docs/technotes/guides/collections/overview.html

原文内容也一并附加在本文最后.

简介:

Java平台包含一个集合框架。

集合是表示一组对象的对象(如经典的Vector类)。

集合框架是用于表示和操作集合的统一体系结构,使集合可以独立于实现细节而被操纵。

集合框架的主要有点在于:

  1. 通过提供数据结构和算法减少编程工作量,因此您不必亲自编写它们。
  2. 通过提供数据结构和算法的高性能实现来提高性能。由于每个接口的各种实现都是可以互换的,因此可以通过切换实现来调整程序。
  3. 通过建立通用语言来回传递集合,提供不相关的API之间的互操作性。--可以部分理解为向上转型 多态
  4. 通过学习多个特定的集合API来减少学习API所需的工作量。  --通过自顶而下的API设计风格
  5. 不要求您实现特定的集合API来减少设计和实现API所需的工作量。--通过提供抽象类提供了很多默认实现
  6. 通过为集合和算法提供标准接口来操纵它们,从而促进软件重用。

集合框架包括:

  1. 集合接口。表示不同类型的集合,例如Set,List和Map。这些接口构成了框架的基础。
  2. 通用实现。集合接口的主要实现。
  3. 遗留实现。早期版本Vector和Hashtable中的集合类进行了改进,以实现集合接口。
  4. 特殊用途的实现。设计用于特殊情况的实现。这些实现展示非标准的性能特征,使用限制或行为。
  5. 并发实现。为高并发使用而设计的实现。
  6. 包装实现。将功能(如同步)添加到其他实现。
  7. 更简化便利的实现.  集合接口的高性能低复杂实现.
  8. 抽象实现。集合接口的部分实现,以促进自定义实现。
  9. 算法。collections中非常有用的静态方法,比如例如排序列表。
  10. 基础架构。为集合即可欧提供重要支持的接口。--就是一些其他辅助的,比如Iterable
  11. 数组工具类。基本类型和引用对象数组的实用函数。严格来说,并不是集合框架的一部分,这个特性与集合框架同时添加到Java平台,并依赖于相同的基础结构。

集合接口:

集合接口分为两大阵营,最基础的接口java.util.Collection,有下面这些后代:

java.util.Set java.util.SortedSet java.util.NavigableSet java.util.Queue java.util.concurrent.BlockingQueue java.util.concurrent.TransferQueue java.util.Deque java.util.concurrent.BlockingDeque

另外的一些集合接口派生于 java.util.Map  不过他们并是真正的集合。

但是,这些接口包含集合视角的操作,这些操作可以将它们作为集合进行操作。Map有以下后代:

java.util.SortedMap java.util.NavigableMap java.util.concurrent.ConcurrentMap java.util.concurrent.ConcurrentNavigableMap

集合中的许多修改方法是可选的.

也就是允许实现类并不必须要实现一个或者多个这种操作.

如果你尝试使用,将会抛出运行时异常UnsupportedOperationException

每个实现的文档必须指定支持哪些可选操作。

引入了几个术语来帮助本规范:

  1. 不支持修改操作的集合(如添加,删除和清除)称为不可修改,非不可修改的集合是可修改的。;
  2. 如果集合额外保证集合对象中没有任何可见的修改那么称之为为不可变,非不可变的就是可变的;
  3. 保证集合大小不变的的集合,即使元素可以变化更改,称之为固定大小,非固定大小的列表称之为可变大小;
  4. 支持快速(通常是固定时间)索引元素访问的List是随机访问(RandomAccess) List,比如arrayList,不支持快速索引访问的称之为顺序访问 (sequential access)List ,RandomAccess标记接口指明一个list具有支持随机访问的特性,这一点保证了你可以编写通用的算法代码,通过选择不同的随机访问还是顺序访问List,以达到更好的性能.

一些实现可以做到限制哪些元素(或者在Map场景下的键和值)可以被存储。可能的限制包括要求元素:

  1. 特定类型
  2. 不能为null 
  3. 遵从一些自定的断言判断式(Obey some arbitrary predicate)

如果尝试往有限制的接口实现中添加不符合的元素,会出现运行时异常,比如 ClassCastException, IllegalArgumentException, or a NullPointerException.

如果尝试在有限制的实现中移除一个不符合条件的元素或者测试是否存在,会导致异常

不过一些受限制的集合支持这种用法.

集合实现:

集合的实现类一般遵从这样子的形式 < 实现特性>加上<接口名称>的形式

下表中列出了通用实现

Interface

Hash Table

Resizable Array

Balanced Tree

Linked List

Hash Table + Linked List

Set

HashSet

TreeSet

LinkedHashSet

List

ArrayList

LinkedList

Deque

ArrayDeque

LinkedList

Map

HashMap

TreeMap

LinkedHashMap

通用实现类支持集合框架中所有的可选操作,并且对于元素没有任何的限制

他们都是非同步的

不过Collections类中包含了很多静态的工厂方法-synchronization wrappers 同步包装器

可以在很多非同的集合中提供同步的功能.

所有新的实现都具有快速失败迭代器,它可以检测到非法的并发修改,并且快速而干净地(而不是报错异常)失败。

ps: fail-fast   java集合框架中的一种机制,检测并发修改

AbstractCollection, AbstractSet, AbstractList, AbstractSequentialList 和 AbstractMap

为了减少实现接口所需代价,提供了核心集合框架的基本实现。

这些类的API文档准确描述了每种方法的实现方式,

因此实现者知道哪些方法必须被重写,考虑到具体实现类的基本操作的性能。

并发集合:

应用程序中使用多个线程的集合的必须仔细编程,这个一般叫做并发编程

Java平台包含对并发编程的广泛支持。有关详细信息,请参阅Java Concurrency Utilities

由于集合经常被使用,各种支持并发编程的接口和集合的实现都被包含在这些API中

这些类型,超出了前面提到过的同步包装,提供了并发编程中经常需要用到的特性

并发接口

BlockingQueue TransferQueue BlockingDeque ConcurrentMap ConcurrentNavigableMap

并发实现类

LinkedBlockingQueue ArrayBlockingQueue PriorityBlockingQueue DelayQueue SynchronousQueue LinkedBlockingDeque LinkedTransferQueue CopyOnWriteArrayList CopyOnWriteArraySet ConcurrentSkipListSet ConcurrentHashMap ConcurrentSkipListMap

设计目标:

主要的设计目标是编写一个体积小概念强大的API

新功能与当前程序差别不大,这一点非常重要.而且,需要在现有的基础上增加

而不是替代他们。

而且,新的API必须足够强大,才能够提供前面所述的所有优势。

为了使核心接口的数量保持较小,

接口不会尝试捕捉诸如易变性,可修改性和可调节性之类的细微区别

相反,在核心接口中一些调用是可选的

使得实现类可以抛出一个UnsupportedOperationException异常指示出他们不支持指定的可选操作

集合实现者必须清楚地在文档中表明实现支持哪些可选操作。

为了使每个核心接口中的方法数量保持较小,

只有在以下情况下,接口才包含方法:

这是一项真正的基本操作:就其他方面可以被合理定义而言, 它是一个基本的操作,

有一个不可抗拒的性能因素导致重要的实现类将会想要去重写它  

所有关于集合的表现形式的互操作性至关重要。

这包括数组,在不改变语言的情况下,不能直接实现Collection接口。

因此,

该框架包含的方法可以将集合移入数组,将数组视为集合,将Map视为集合。

原文内容:

Skip to Content

Collections Framework Overview

Introduction

The Java platform includes a collections framework. A collection is an object that represents a group of objects (such as the classic Vector class). A collections framework is a unified architecture for representing and manipulating collections, enabling collections to be manipulated independently of implementation details.

The primary advantages of a collections framework are that it:

  • Reduces programming effort by providing data structures and algorithms so you don't have to write them yourself.
  • Increases performance by providing high-performance implementations of data structures and algorithms. Because the various implementations of each interface are interchangeable, programs can be tuned by switching implementations.
  • Provides interoperability between unrelated APIs by establishing a common language to pass collections back and forth.
  • Reduces the effort required to learn APIs by requiring you to learn multiple ad hoc collection APIs.
  • Reduces the effort required to design and implement APIs by not requiring you to produce ad hoc collections APIs.
  • Fosters software reuse by providing a standard interface for collections and algorithms with which to manipulate them.

The collections framework consists of:

  • Collection interfaces. Represent different types of collections, such as sets, lists, and maps. These interfaces form the basis of the framework.
  • General-purpose implementations. Primary implementations of the collection interfaces.
  • Legacy implementations. The collection classes from earlier releases, Vector and Hashtable, were retrofitted to implement the collection interfaces.
  • Special-purpose implementations. Implementations designed for use in special situations. These implementations display nonstandard performance characteristics, usage restrictions, or behavior.
  • Concurrent implementations. Implementations designed for highly concurrent use.
  • Wrapper implementations. Add functionality, such as synchronization, to other implementations.
  • Convenience implementations. High-performance "mini-implementations" of the collection interfaces.
  • Abstract implementations. Partial implementations of the collection interfaces to facilitate custom implementations.
  • Algorithms. Static methods that perform useful functions on collections, such as sorting a list.
  • Infrastructure. Interfaces that provide essential support for the collection interfaces.
  • Array Utilities. Utility functions for arrays of primitive types and reference objects. Not, strictly speaking, a part of the collections framework, this feature was added to the Java platform at the same time as the collections framework and relies on some of the same infrastructure.

Collection Interfaces

The collection interfaces are divided into two groups. The most basic interface, java.util.Collection, has the following descendants:

The other collection interfaces are based on java.util.Map and are not true collections. However, these interfaces contain collection-view operations, which enable them to be manipulated as collections. Map has the following offspring:

Many of the modification methods in the collection interfaces are labeled optional. Implementations are permitted to not perform one or more of these operations, throwing a runtime exception (UnsupportedOperationException) if they are attempted. The documentation for each implementation must specify which optional operations are supported. Several terms are introduced to aid in this specification:

  • Collections that do not support modification operations (such as add, remove and clear) are referred to as unmodifiable. Collections that are not unmodifiable are modifiable.
  • Collections that additionally guarantee that no change in the Collection object will be visible are referred to as immutable. Collections that are not immutable are mutable.
  • Lists that guarantee that their size remains constant even though the elements can change are referred to as fixed-size. Lists that are not fixed-size are referred to as variable-size.
  • Lists that support fast (generally constant time) indexed element access are known as random access lists. Lists that do not support fast indexed element access are known as sequential access lists. The RandomAccess marker interface enables lists to advertise the fact that they support random access. This enables generic algorithms to change their behavior to provide good performance when applied to either random or sequential access lists.

Some implementations restrict what elements (or in the case of Maps, keys and values) can be stored. Possible restrictions include requiring elements to:

  • Be of a particular type.
  • Be not null.
  • Obey some arbitrary predicate.

Attempting to add an element that violates an implementation's restrictions results in a runtime exception, typically a ClassCastException, an IllegalArgumentException, or a NullPointerException. Attempting to remove or test for the presence of an element that violates an implementation's restrictions can result in an exception. Some restricted collections permit this usage.


Collection Implementations

Classes that implement the collection interfaces typically have names in the form of <Implementation-style><Interface>. The general purpose implementations are summarized in the following table:

Interface

Hash Table

Resizable Array

Balanced Tree

Linked List

Hash Table + Linked List

Set

HashSet

TreeSet

LinkedHashSet

List

ArrayList

LinkedList

Deque

ArrayDeque

LinkedList

Map

HashMap

TreeMap

LinkedHashMap

The general-purpose implementations support all of the optional operations in the collection interfaces and have no restrictions on the elements they may contain. They are unsynchronized, but the Collections class contains static factories called synchronization wrappers that can be used to add synchronization to many unsynchronized collections. All of the new implementations have fail-fast iterators, which detect invalid concurrent modification, and fail quickly and cleanly (rather than behaving erratically).

The AbstractCollection, AbstractSet, AbstractList, AbstractSequentialList and AbstractMap classes provide basic implementations of the core collection interfaces, to minimize the effort required to implement them. The API documentation for these classes describes precisely how each method is implemented so the implementer knows which methods must be overridden, given the performance of the basic operations of a specific implementation.


Concurrent Collections

Applications that use collections from more than one thread must be carefully programmed. In general, this is known as concurrent programming. The Java platform includes extensive support for concurrent programming. See Java Concurrency Utilities for details.

Collections are so frequently used that various concurrent friendly interfaces and implementations of collections are included in the APIs. These types go beyond the synchronization wrappers discussed previously to provide features that are frequently needed in concurrent programming.

These concurrent-aware interfaces are available:

The following concurrent-aware implementation classes are available. See the API documentation for the correct usage of these implementations.


Design Goals

The main design goal was to produce an API that was small in size and, more importantly, in "conceptual weight." It was critical that the new functionality not seem too different to current Java programmers; it had to augment current facilities, rather than replace them. At the same time, the new API had to be powerful enough to provide all the advantages described previously.

To keep the number of core interfaces small, the interfaces do not attempt to capture such subtle distinctions as mutability, modifiability, and resizability. Instead, certain calls in the core interfaces are optional, enabling implementations to throw an UnsupportedOperationException to indicate that they do not support a specified optional operation. Collection implementers must clearly document which optional operations are supported by an implementation.

To keep the number of methods in each core interface small, an interface contains a method only if either:

  • It is a truly fundamental operation: a basic operations in terms of which others could be reasonably defined,
  • There is a compelling performance reason why an important implementation would want to override it.

It was critical that all reasonable representations of collections interoperate well. This included arrays, which cannot be made to implement the Collection interface directly without changing the language. Thus, the framework includes methods to enable collections to be moved into arrays, arrays to be viewed as collections, and maps to be viewed as collections.

Copyright © 1993, 2018, Oracle and/or its affiliates. All rights reserved.

Contact Us

本文参与腾讯云自媒体分享计划,欢迎正在阅读的你也加入,一起分享。

发表于

我来说两句

0 条评论
登录 后参与评论

相关文章

来自专栏Albert陈凯

2018-06-13 关于Java集合的小抄

15730
来自专栏刘望舒

JNI的实现原理

JNI是Java Native Interface的缩写,它为java提供了调用C和C++代码的能力。java.lang包下的很多类都用到了native方法,比...

62220
来自专栏编程之旅

线性表的顺序存储结构

举个简单的例子,蔺老师在给九班学生安排座位之前,会让学生们从矮到高按照身高的高矮升序排列,假如蔺老师的班上只有十个学生,而全班共有50个座位,那蔺老师会把这10...

20620
来自专栏Python小屋

使用Python编写程序求解数独游戏答案

问题描述:数独盘面是个九宫,每一宫又分为九个小格。在这八十一格中给出一定的已知数字和解题条件,利用逻辑和推理,在其他的空格上填入1-9的数字。使1-9每个数字在...

30730
来自专栏java达人

ConcurrentHashMap使用示例

作者:mononite 链接:https://my.oschina.net/mononite/blog/144329(点击文末阅读原文前往) Concurren...

40880
来自专栏Golang语言社区

用Golang写一个搜索引擎

本篇较长较枯燥,请保持耐心看完。 前面两章介绍了一下倒排索引以及倒排索引字典的两种存储结构,分别是 跳跃表 和 哈希表 ,本篇我们介绍另一种数据结构,他也被大量...

48970
来自专栏光变

你所不知道的Java之Switch

??? Enum,String,Character,Byte,Short,Integer

22500
来自专栏Java技术栈

跟我学 Java 8 新特性之 Stream 流(二)关键知识点

我们的第一篇文章,主要是通过一个Demo,让大家体验了一下使用流API的那种酣畅淋漓的感觉。如果你没有实践,我还是再次呼吁你动手敲一敲,自己实实在跑一遍上一篇的...

15140
来自专栏CSDN技术头条

【问底】静行:FastJSON实现详解

还记得电影《功夫》中火云邪神的一句话:天下功夫,无坚不破,唯快不破。在程序员的世界中,“快”一直是大家苦苦修炼,竞相追逐的终极目标之一,甚至到了“不择手段”、“...

32270
来自专栏JAVA烂猪皮

轻松理解 Java HashMap 和 ConcurrentHashMap

Map 这样的 Key Value 在软件开发中是非常经典的结构,常用于在内存中存放数据。

8910

扫码关注云+社区

领取腾讯云代金券