Java 自定义队列Queue:
队列的抽象数据类型就是一个容器,其中的对象排成一个序列,我们只能访问和取出排在最前端( Front)的对象,只能在队列的尾部( Rear)插入新对象。正是按照这一规则,才能保证最先被插入的对象首先被删除( FIFO)。java本身是有自带Queue类包,为了达到学习目的已经更好深入了解Queue队列,自己动手自建java Queue类是个很好的学习开始:
基于数组的实现
„ 顺序数组
借助一个定长数组 Q 来存放对象,即可简单地实现队列。那么,为了符合 FIFO 准则,应该如何表示和记录队列中各对象的次序呢?
一种自然的办法就是仿照栈的实现,以 Q[0]作为队首,其它对象顺序往后存放。然而如此一来,每次首元素出队之后,都需要将后续的所有元素向前顺移一个单元若队长为 n,这项工作需要O(n)时间,因此效率很低。
„ 循环数组
为了避免数组的整体移动,可以引入如下两个变量 f 和 r:
f:始终等于 Q 的首元素在数组中的下标,即指向下次出队元素的位置
r:始终等于 Q 的末元素的下标加一,即指向下次入队元素的位置
一开始, f = r = 0,此时队空。每次有对象入队时,将其存放于 Q[r],然后 r 加一,以指向下一单元。对称地,每次有对象出队之后,也将 f 加一,指向新的队首元素。这样,对 front()、 enqueue()和 dequeue()方法的每一次调用都只需常数时间。
然而,这还不够。细心的读者或许已经注意到,按照上述约定,在队列的生命期内, f 和 r 始终在单调增加。因此,若队列数组的容量为 N,则在经过 N 次入队操作后, r 所指向的单元必然超出数组的范围;在经过 N 次出队操作后, f 所指向的单元也会出现类似的问题。
解决上述问题的一种简便方法,就是在每次 f 或 r 加一后,都要以数组的长度做取模运算,以保证其所指单元的合法性。就其效果而言,这就相当于把数组的头和尾相联,构成一个环状结构。
基于上述构想,可以得到如 下所示java代码:
Queue类:
package com.queue;
import java.util.Arrays;
/**
*
* @author gannyee
*
*/
public class Queue {
// Define capacity constant: CAPACITY
private static final int CAPACITY = 1024;
// Define capacity of queue
private static int capacity;
// Front of queue
private static int front;
// Tail of queue
private static int tail;
// Array for queue
private static Object[] array;
// Constructor of Queue class
public Queue() {
this.capacity = CAPACITY;
array = new Object[capacity];
front = tail = 0;
}
// Get size of queue
public static int getSize() {
if (isEmpty())
return 0;
else
return (capacity + tail - front) % capacity;
}
// Whether is empty
public static boolean isEmpty() {
return (front == tail);
}
// put element into the end of queue
public static void enqueue(Object element) throws ExceptionQueueFull {
if (getSize() == capacity - 1)
throw new ExceptionQueueFull("Queue is full");
array[tail] = element;
tail = (tail + 1) % capacity;
}
// get element from queue
public static Object dequeue() throws ExceptionQueueEmpty {
Object element;
if (isEmpty())
throw new ExceptionQueueEmpty("Queue is empty");
element = array[front];
front = (front + 1) % capacity;
return element;
}
// Get the first element for queue
public static Object frontElement() throws ExceptionQueueEmpty {
if (isEmpty())
throw new ExceptionQueueEmpty("Queue is empty");
return array[front];
}
// Travel all elements of queue
public static void getAllElements() {
Object[] arrayList = new Object[getSize()];
for (int i = front,j = 0; j < getSize(); i ++,j ++) {
arrayList[j] = array[i];
}
System.out.println("All elements of queue: "
+ Arrays.toString(arrayList));
}
}
自定义ExceptionStackEmpty异常类:
package com.queue;
public class ExceptionQueueEmpty extends Exception {
// Constructor
public ExceptionQueueEmpty() {
}
// Constructor with parameters
public ExceptionQueueEmpty(String mag) {
System.out.println(mag);
}
}
自定义ExceptionStackFull异常类
package com.queue;
public class ExceptionQueueFull extends Exception {
// Constructor
public ExceptionQueueFull() {
}
// Constructor with parameters
public ExceptionQueueFull(String mag) {
System.out.println(mag);
}
}
测试类:
package com.queue;
/**
* QueueTest
* @author gannyee
*
*/
public class QueueTest {
public static void main(String[] args) {
// TODO Auto-generated method stub
Queue queue = new Queue();
System.out.println("The size of queue is: " + queue.getSize());
System.out.println("Is empty: " + queue.isEmpty());
try {
queue.enqueue(8);
queue.enqueue(3);
queue.enqueue(5);
queue.enqueue(7);
queue.enqueue(9);
queue.getAllElements();
System.out.println("The size of queue is: " + queue.getSize());
System.out.println("Is empty: " + queue.isEmpty());
System.out.println("The front element of queue: "
+ queue.frontElement());
System.out.println(queue.dequeue());
System.out.println(queue.dequeue());
System.out.println(queue.dequeue());
System.out.println(queue.dequeue());
System.out.println(queue.dequeue());
System.out.println("The size of queue is: " + queue.getSize());
System.out.println("Is empty: " + queue.isEmpty());
} catch (ExceptionQueueFull e) {
// TODO Auto-generated catch block
e.printStackTrace();
} catch (ExceptionQueueEmpty e) {
// TODO Auto-generated catch block
e.printStackTrace();
}
}
}
测试结果:
The size of queue is: 0
Is empty: true
All elements of queue: [8, 3, 5, 7, 9]
The size of queue is: 5
Is empty: false
The front element of queue: 8
8
3
5
7
9
The size of queue is: 0
Is empty: true
队列的应用:
孩提时的你是否玩过“烫手山芋”游戏:一群小孩围成一圈,有一个刚出锅的山芋在他们之间传递。其中一个孩子负责数数,每数一次,拿着山芋的孩子就把山芋转交给右边的邻居。一旦数到某个特定的数,拿着山芋的孩子就必须退出,然后重新数数。如此不断,最后剩下的那个孩子就是幸运者。通常,数数的规则总是从 1 开始,数到 k 时让拿着山芋的孩子出列,然后重新从 1 开始。Josephus问题可以表述为: n 个孩子玩这个游戏,最后的幸运者是谁?
为了解答这个问题,我们可以利用队列结构来表示围成一圈的n个孩子。一开始,假定对应于队列首节点的那个孩子拿着山芋。然后,按照游戏的规则,把“土豆”向后传递到第k个孩子(交替进行k次dequeue()和k次enqueue()操作),并让她出队( dequeue())。如此不断迭代,直到队长(getSize())为 1。
Java代码如下:
package com.queue;
public class QueueBuilder {
// Building string type array for queue
public static Queue queueBuild(String[] str) {
Queue queue = new Queue();
for (int i = 0; i < str.length; i++) {
try {
queue.enqueue(str[i]);
} catch (ExceptionQueueFull e) {
// TODO Auto-generated catch block
e.printStackTrace();
}
}
return queue;
}
// Weed out the kth kid who get the sweet potato
public static String gameWiner(Queue queue, int k)
throws ExceptionQueueFull, ExceptionQueueEmpty {
if (queue.isEmpty())
return null;
while (queue.getSize() > 1) {
queue.getAllElements();// Output recently queue
for (int i = 0; i < k - 1; i++)
queue.enqueue(queue.dequeue());
System.out.println("\n\t" + queue.dequeue() + ": Weep out");
}
return (String) queue.dequeue();
}
}
package com.queue;
public class QueueGame {
public static void main(String[] args) throws ExceptionQueueFull, ExceptionQueueEmpty {
// TODO Auto-generated method stub
String[] kid = {"Alice", "Bob", "Cindy", "Doug", "Ed",
"Fred", "Gene", "Hope", "Irene", "Jack",
"Kim", "Lance", "Mike", "Nancy", "Ollie"};
QueueBuilder qb = new QueueBuilder();
System.out.println("The luck dog is: " + qb.gameWiner(qb.queueBuild(kid), 5));
}
}
测试结果:
All elements of queue: [Alice, Bob, Cindy, Doug, Ed, Fred, Gene, Hope, Irene, Jack, Kim, Lance, Mike, Nancy, Ollie]
Ed: weep out
All elements of queue: [Fred, Gene, Hope, Irene, Jack, Kim, Lance, Mike, Nancy, Ollie, Alice, Bob, Cindy, Doug]
Jack: weep out
All elements of queue: [Kim, Lance, Mike, Nancy, Ollie, Alice, Bob, Cindy, Doug, Fred, Gene, Hope, Irene]
Ollie: weep out
All elements of queue: [Alice, Bob, Cindy, Doug, Fred, Gene, Hope, Irene, Kim, Lance, Mike, Nancy]
Fred: weep out
All elements of queue: [Gene, Hope, Irene, Kim, Lance, Mike, Nancy, Alice, Bob, Cindy, Doug]
Lance: weep out
All elements of queue: [Mike, Nancy, Alice, Bob, Cindy, Doug, Gene, Hope, Irene, Kim]
Cindy: weep out
All elements of queue: [Doug, Gene, Hope, Irene, Kim, Mike, Nancy, Alice, Bob]
Kim: weep out
All elements of queue: [Mike, Nancy, Alice, Bob, Doug, Gene, Hope, Irene]
Doug: weep out
All elements of queue: [Gene, Hope, Irene, Mike, Nancy, Alice, Bob]
Nancy: weep out
All elements of queue: [Alice, Bob, Gene, Hope, Irene, Mike]
Irene: weep out
All elements of queue: [Mike, Alice, Bob, Gene, Hope]
Hope: weep out
All elements of queue: [Mike, Alice, Bob, Gene]
Mike: weep out
All elements of queue: [Alice, Bob, Gene]
Bob: weep out
All elements of queue: [Gene, Alice]
Gene: weep out
The luck dog is: Alice
文章参考:数据结构与算法( Java 描述)邓俊辉 著
转载请注明出处,谢谢!
http://blog.csdn.net/github_27609763/article/details/46434301