简体繁体 English

实现数据结构-Flex词典

[英]Implementing data structure - flex dictionary

原文 2014-05-27 18:21:26 1 1 java/ dictionary/ hashtable/ skip-lists

I've been asked to devise a data structure that represents a dictionary. 我被要求设计一个代表字典的数据结构。 The dictionary holds items with distinct key number. 该词典包含具有不同键号的项目。 The data structure should support the following operations in O(1) time: insert(x), delete(x), findMin(), findMax(), successor(x), predecessor(x). 数据结构应在O（1）时间内支持以下操作：insert（x），delete（x），findMin（），findMax（），successor（x），predecessor（x）。 Also search(x) operation should be in O(log n) time. search（x）操作也应在O（log n）时间内进行。

The assignment was given on the following subjects: skip list, hash tables and heaps. 分配是在以下主题上进行的：跳过列表，哈希表和堆。 I guess the best structure will be a skip list, but I couldn't find a way implementing insert and delete in O(1) tine. 我想最好的结构将是一个跳过列表，但是我找不到在O（1）例行中实现插入和删除的方法。 Any suggestions? 有什么建议么？

1 个解决方案

You should use both hash table and skip list data structures. 您应该同时使用哈希表和跳过列表数据结构。 You can update min and max values each time you insert an element. 您可以在每次插入元素时更新min和max 。 Hence, findMin() and findMax() are also O(1) . 因此， findMin()和findMax()也是O(1) 。
insert() and delete() are also in constant time. insert()和delete()也处于恒定时间。

By the way, you cannot delete an item before searching for it. 顺便说一句，您无法在搜索之前删除项目。 If you do search in O(logn) , delete() will be O(logn) automatically. 如果确实在O(logn)搜索，则delete()将自动为O(logn) 。

Hash table provides (if implemented correctly), search, delete and insert in O(1) . 哈希表提供（如果正确实现），搜索，删除和插入O(1) 。 The rest of the operations ( successor() , predecessor() ) are available in skip list. 其余操作（ successor() ， predecessor() ）在跳过列表中可用。