Use standard Collections

最新推荐文章于 2019-01-10 19:40:58 发布

转载最新推荐文章于 2019-01-10 19:40:58 发布 · 536 阅读

Java Practise 专栏收录该内容

13 篇文章

订阅专栏

本文探讨了集合框架中各种实现类的选择与性能优化策略，包括ArrayList、LinkedList、HashMap、HashSet、ArrayDeque等，强调了避免使用具有不确定迭代顺序的古老实现类如HashMap和HashSet，推荐使用TreeSet和TreeMap来保持元素的排序。文章还提到了在图形用户界面中使用这些类以保留项目操作的排序。

摘要生成于 C知道，由 DeepSeek-R1 满血版支持，前往体验 >

In the Collections Framework , there are many implementation classes to choose from. Most of the time, one of these "primary implementations" is the appropriate choice

ArrayList, for an implementation of the List interface
LinkedHashMap, for an implementation of the Map interface
LinkedHashSet, for an implementation of the Set interface
ArrayDeque, for an implementation of both the the Deque and Queue interfaces

Before JDK 1.5, HashMap and HashSet were the preferred implementations of Map and Set. However, the iterators returned by those classes have the somewhat bizarre property of having an undefined order. That is, iterating over a HashMap or HashSet can return elements in a different order at different times. Their iteration order is not guaranteed to be repeatable (and often isn't). Even though HashMap and HashSet have slightly better overall performance, for most business applications it is likely best to avoid such undefined behavior.

To retain the sorting of items being manipulated in a graphical user interface, TreeSet and TreeMap are useful. They force a collection to maintain a sort order when the user adds and deletes items.