1. Stacks
황승원
Fall 2010
CSE, POSTECH

2. Stack Of Cups bottom top bottom top Remove a cup from new stack.D E F
bottom
top C A B D E
Picture is really a stack of cups and saucers.
LIFO = last in first out. The first cup that is removed from a stack of cups is the
Last one that was added to the stack.
Other examples of LIFO lists in real life: stack of trays in a cafeteria; paper stack in a printer or copy machine; newspaper stack at a news stand.
Add a cup to the stack.
Remove a cup from new stack.
A stack is a LIFO list.

3. The Interface Stack public interface Stack { public boolean empty();
public void push(Object theObject);
public Object pop();
public Object peek(); }
Choice of method names is the same as used in Java’s class java.util.Stack. Notice the difference in names of the method to check if an instance is empty—isEmpty for linear lists and empty for a stack.

4. Derive From A Linear List Class
ArrayLinearList
Chain

5. Derive From ArrayLinearList1 2 3 4 5 6 a b c d e
stack top is either left end or right end of linear list
empty() => isEmpty()
O(1) time
peek() => get(0)
The decision whether to use the left or right end of the linear list as the stack top is made on the basis of efficiency of resulting stack methods. The complexity of empty() and peek() are independent of the choice.

6. Derive From ArrayLinearList1 2 3 4 5 6 a b c d e
top=left end
push(theObject) => add(0, theObject)
 O(size) time
pop() => remove(0)

7. Derive From ArrayLinearList1 2 3 4 5 6 a b c d e
top=right end
push(theObject) => add(size(), theObject)
 O(1) time
pop() => remove(size()-1)

8. Derive From Chain a b c d e
null
firstNode
stack top is either left end or right end of linear list
empty() => isEmpty()
O(1) time
The complexity of empty() is independent of which end of the chain is used as the top of the stack. The complexity of the remaining methods depends on which end is the stack top.

9. Derive From Chain (top=left)b c d e
null
firstNode
peek() => get(0)
 O(1) time
push(theObject) => add(0, theObject)
pop() => remove(0)

10. Derive From Chain (top=right)b c d e
null
firstNode
peek() => get(size()-1)
 O(size) time
push(theObject) => add(size(), theObject)
pop() => remove(size()-1)

11. Derive From ArrayLinearList
package dataStructures;
import java.util.*; // has stack exception
public class DerivedArrayStack
extends ArrayLinearList
implements Stack {
// constructors come here
// Stack interface methods come here }

12. Constructors /** create a stack with the given initial * capacity */
public DerivedArrayStack(int initialCapacity)
{super(initialCapacity);}
/** create a stack with initial capacity 10 */
public DerivedArrayStack()
{this(10);}

13. empty() And peek() 1 2 3 4 5 6 a b c d e public boolean empty()1 2 3 4 5 6 a b c d e
public boolean empty()
{return isEmpty();}
public Object peek() {
if (empty())
throw new EmptyStackException();
return get(size() - 1) }

14. push(theObject) And pop()1 2 3 4 5 6 a b c d e
public void push(Object theElement)
{add(size(), theElement);}
public Object pop() {
if (empty())
throw new EmptyStackException();
return remove(size() - 1); }

15. Evaluation Merits of deriving from ArrayLinearList
Code for derived class is quite simple and easy to develop.
Code is expected to require little debugging.
Code for other stack implementations such as a linked implementation are easily obtained.
Just replace extends ArrayLinearList with extends Chain
For efficiency reasons we must also make changes to use the left end of the list as the stack top rather than the right end.

16. Demerits
All public methods of ArrayLinearList may be performed on a stack.
get(0) … get bottom element
remove(5)
add(3, x)
So we do not have a true stack implementation.
Must override undesired methods.
public Object get(int theIndex)
{throw new UnsupportedOperationException();}
Change earlier use of get(i) to super.get(i).

17. Evaluation
Code developed from scratch will run faster but will take more time (cost) to develop.
Tradeoff between software development cost and performance.
Tradeoff between time to market and performance.
Could develop easy code first and later refine it to improve performance.

18. Code From Scratch Use a 1D array stack whose data type is Object.
same as using array element in ArrayLinearList
Use an int variable top.
Stack elements are in stack[0:top].
Top element is in stack[top].
Bottom element is in stack[0].
Stack is empty iff top = -1.
Number of elements in stack is top+1.

19. Code From Scratch package dataStructures;
import java.util.EmptyStackException;
import utilities.*; // ChangeArrayLength
public class ArrayStack implements Stack {
// data members
int top; // current top of stack
Object [] stack; // element array
// constructors come here
// Stack interface methods come here }

20. Constructors public ArrayStack(int initialCapacity) {
if (initialCapacity < 1)
throw new IllegalArgumentException
("initialCapacity must be >= 1");
stack = new Object [initialCapacity];
top = -1; }
public ArrayStack()
{this(10);}

21. push(…) top 1 2 3 4 a b c d e public void push(Object theElement) {1 2 3 4 a b c d e
top
public void push(Object theElement) {
// increase array size if necessary
if (top == stack.length - 1)
stack = ChangeArrayLength.changeLength1D
(stack, 2 * stack.length);
// put theElement at the top of the stack
stack[++top] = theElement; }

22. pop() top 1 2 3 4 a b c d e public Object pop() { if (empty())1 2 3 4 a b c d e
top
public Object pop() {
if (empty())
throw new EmptyStackException();
Object topElement = stack[top];
stack[top--] = null; // enable garbage collection
return topElement; }

23. Linked Stack From Scratch
See text.

24. java.util.Stack Derives from java.util.Vector.
java.util.Vector is an array implementation of a linear list.

25. Application

26. Towers Of Hanoi/BrahmaC 4 3 2
Could use animation of towers of hanoi from animations page on Web site.
Also known as Towers of Brahma.
According to legend, on the day of creation Buddhist monks began the task of moving disks from tower A to tower C. When they get done, the world will come to an end. 1
64 gold disks to be moved from tower A to tower C
each tower operates as a stack
one at a time
cannot place big disk on top of a smaller one!

27. Towers Of Hanoi/BrahmaC 3 2 1 A
3-disk Towers Of Hanoi/Brahma? Rather easy (7 moves)
What if there are 64 disks? 1.8 * 10^9 

28. Recursive Solution A B C 1
move top n-1 disks from A to B using C

29. Recursive Solution B C 1 A
move top disk from A to C

30. Recursive Solution B C 1 A
move top n-1 disks from B to C using A

31. Recursive Solution B C A moves(n) = 0 when n = 01 A
moves(n) = 0 when n = 0
moves(n) = 2*moves(n-1) + 1 = 2n-1 when n > 0

32. Simple Recursive Implementation
Public class Hanoi {
public static void tHanoi(int n){
tower=new ArrayStack[4] for (int i=1;i<=3; i++) tower[i]=new ArrayStack(); for (int d=n; d>0; d--) tower[1].push(new Integer(d)); move(n,1,2,3); //12 using 3 as intermediate }
public static void move (int n, int x, int y, int z){
move(n-1,x,z,y); // (n-1) xz d=tower[x].pop(); tower[y].push(d); // top xy
System.out.println(“move ”+d+” from ”+x+” to ”+y); move(n-1,z,y,x); // (n-1) zy

33. Towers Of Hanoi/Brahma
moves(64) = 1.8 * 1019 (approximately)
Performing 109 moves/second, a computer would take about 570 years to complete.
At 1 disk move/min, the monks will take about 3.4 * 1013 years.
At the rate of a billion moves/second it would take 570 years to complete the task of moving 64 disks. Of course, the Buddhist monks engaged in this activity are moving far fewer disks/second. At the rate of (say) 1 disk a minute (the disks are, after all, rather heavy), the monks will take about 3.4 * 1013 years. So there is quite a while to go before the world comes to an end.

34. Coming Up Next
READING: Ch 9
NEXT: Queue (FIFO) Ch 10