(chapter 8) A Concise and Practical Introduction to Programming Algorithms in Java

1A Concise and Practical Introduction to Programming Algorithms in Java © 2009 Frank Nielsen
Frank NIELSEN
nielsen@lix.polytechnique.fr
A Concise and
Practical
Introduction to
Programming
Algorithms in Java
Chapter 8: Data-structures and object methods

● FIFO data-structures
(= First In First Out)
● Heap data-structures
● Non-static methods
(= object methods)
● Revisiting lists (OO-style)
Agenda

FIFOs: Mastering queues
● Objects are considered in turn, one by one
● Process each object according to their arrival time
● While objects are processed, others are queued
● First object should be first served!
Basic examples:
● Waiting in a queue at the post office.
●Printing « jobs> on a printer.
FIFO = First In First Out !

A basic solution
● Stack objects in an array as soon as they arrive
● To stack an incoming object, should know the index
of the last location
● Should also know the index of the last processed object
(so that we can process the next one)
While processing an object, others can come in the array
(= queued)

A basic solution
● An array: container array for storing objects
● Two indices: lastProcessed and freePlace
● To add an object x, we do array[freePlace]=x and
we then increment: freePlace++
● To process an object, we increment lastProcessed and
we process array[lastProcessed]

Visual depiction of a FIFO
lastProcessed
freePlace
0 1 n-1-1

lastProcessed
freePlace
O1
FIFO: Queuing objects
array[freePlace++]=O1;

O1
Queuing another object
O2
lastProcessed
freePlace

O1
Processing and queuing
O2 O3
lastProcessed
freePlace
Process(array[lastProcessed++]);
Processing and queuing can be done in parallel using threads

Programming queues
static int lastProcessed=-1;
static int freePlace=0;
static double[] container=new double[1000];
static void add(double Object)
{
if (freePlace<1000)
{container[freePlace]=Object;
freePlace++;}
}
static double process()
{
if (freePlace-lastProcessed>1)
{ // Do something here
lastProcessed++;
return container[lastProcessed];
}
else
return -1.0; // special return code: no process
}

class QueueDouble
{
static int lastProcessed=-1;
static int freePlace=0;
// Max objects is set to 1000
static double[] container=new double[1000];
// Stack in FIFO order
static void add(double a)
{...}
// Process in FIFO order
static double process()
{...}
public static void main(String[] arg)
{
System.out.println("Queue demo:");
add(3.0);
add(5.0);
add(7.0);
System.out.println(process());
}
FIFO:
First In First Out

Exercise: FIFO in action!
Let A be a set of integers such that:
● 1 belongs to A, and
● If a belongs to A, then 2*a+1 and 3*a belongs to A
Question:
For a given n, display all integers less or equal to n
that belong to A.

Programming queues
Start with a FIFO initialized with element 1
Use a boolean array to store whether a belong to A
(= marks, tag elements)
For each element a of the FIFO do:
● Compute 2*a+1 and 3*a
● Add them to the FIFO if they are less than n
...and not yet encountered (=marked)
Terminate when the FIFO is empty
Display all marked elements (=result)

final static int n=1000; static int lastProcessed=-1;
static int freePlace=0; static int[] container=new int[n];
static boolean[] mark=new boolean[n];
static void add(int a)
{if (freePlace<n) {container[freePlace]=a;freePlace++;}}
static boolean Empty()
{ return ((freePlace-lastProcessed)==1); }
static void process()
{int a;
lastProcessed++; a=container[lastProcessed];
if (a<n) mark[a]=true;
if (2*a+1<n) add(2*a+1);
if (3*a<n) add(3*a);
}
public static void main(String[] arg)
{int i;
for(i=0;i<n;i++) mark[i]=false;
add(1);
while(!Empty())
process();
for(i=0;i<n;i++)
{if (mark[i])
System.out.print(i+" ");}
System.out.println("");
}

A few remarks on FIFOs
● Set beforehand the size of the array?
● Can wrap the array using mod MAX_SIZE
(=circular ring, extend arrays, etc.)
...But how to check whether the queue is empty
... or full with circular arrrays?

Priority queues: Heaps (=tas)
● Objects are considered in turn
● Need to process them according to their priorities
● While processing an objects, other may arrive
(= are being queued)
● Serve the object with the highest priority first
Examples:
● Ressource request
● Operating system tasks

Defining mathematically heaps
A heap is a sequence of integers:
stored compactly in an array such that:
For example:
37, 22, 3131, 16, 17, 2, 2323, 12, 6, 9
( heap of 10 elements)
i=7
j=(int)(7/2)=3

Drawing a heap
● Draw a heap as a tree (=special graph Vertex/Edge)
● Each node i contains a value t[i] and has
0, 1 or 2 siblings that contain nodes of values
less than its parent
● Node i has children 2i and 2i+1 t[i]
37, 22, 31, 16, 17, 2, 23, 12, 6, 9:
Read layer by layer,
from the root til the leaves

Storing and manipulating heaps
Easy to code with a linear arraypublic class Heap
{
int size;
int [] label;
static final int MAX_SIZE=10000;
Heap()
{
this.size=0;
this.label=new int[MAX_SIZE];
}
public static void main(String[] args)
{}
}

Fundamental property of heaps
Largest value is stored at the root of the tree, namely
at the first element of the array.
static int maxHeap(Heap h)
{
return h.label[0];
}

Adding an element in a heap
●
Add the new element in position n (=n+1th
element)...
● But the condition that the array is a heap is violated...
● So that we swap the element until...
. ...it satisfies the heap constraint

Example: Add element 25
Not a heap anymore!!! 25>17
Swap with your parent!

Add element 25... and swap!!!
It is a heap now!

Adding an element in the heap
static void addHeap(int element, Heap h)
{
h.label[h.size]=element;
h.size++;
int i=h.size;
int j=i/2;
while (i>1 && h.label[i]>h.label[j])
{
int tmp=h.label[i];
h.label[i]=h.label[j];
h.label[j]=tmp;
i=j; // swap
j=i/2;
}
}

Removing the largest element of a heap
● We move the element at position (n-1) and
put it at the root (position 0)
● But it is not anymore a heap...
● So we swap to the bottom until...
...the heap condition is satisfied

37

17

Then Swap parent-child...

static int removeHeap(int element, Heap h)
{
h.label[0]=h.label[h.size-1];
h.size--;
int i=0,j,k,tmp;
while(2*i<=h.size)
{
j=2*i;
if (j<h.size && h.label[j+1]>h.label[j])
j++;
if (h.label[i]<h.label[j])
{tmp=h.label[i];
h.label[i]=h.label[j];
h.label[j]=tmp;
i=j;}
else break;
}
return h.label[h.size-1];
}

Non-static methods and objects
● Do not write static in front of the method
● Method is thus attached to the object for which it applies for
● For example, we prefer:
u.display() rather than display(u)
u.addElement(a) instead of addElement(a,u)
● To reference the object on which the method is called upon
use this
Object-oriented programming paradigm (OO)

Object-oriented programming paradigm (OO)
● Design a software as a set of objects and methods
applying on these objects
● Ask yourself first:
- What are the objects?
- What are the methods?
● Usually, a method often modifies the object (=fields)
on which it applies for.
(But not always, for example: Obj.Display())

OO style:
object methods
versus
static functions
class Box
{
double width, height, depth;
Box(double w, double h, double d)
{
this.width=w; this.height=h; this.depth=d;
}
double Volume()
{return this.width*this.height*this.depth;}
}
class OOstyle
{
static double Volume(Box box)
{return box.width*box.height*box.depth;}
public static void main(String[] s)
{
Box myBox=new Box(5,2,1);
System.out.println("Volume by static method:"+Volume(myBox));
System.out.println("Volume by object method:"+myBox.Volume());
}
} Object (non-static) method

class Toolkit
{
static final double PI=3.14;
static double Square(double x)
{return x*x;}
static double Cubic(double x)
{return x*x*x;}
}
class StaticFuncStyle
{
public static void main(String[] s)
{
double radius=0.5;
double volSphere=(4/3.0)*Toolkit.PI*Toolkit.Cubic(radius);
double areaDisk=Toolkit.PI*Toolkit.Square(radius);
}
}
Static methods are
useful for defining:
- Constants
- Basic functions
....in a library.

Heaps revisited in Object-Oriented style
int maxHeap()
{
return this.label[0];
}
void add(int element)
{
...
}
void removeTop()
{
...
}
Observe that the keyword static has disappeared

List in object-oriented style
● A cell stores information (say, an integer) and
point/refer to the next one.
● Pointing to another cell means storing a reference
to the corresponding cell.

Pay special attention to null !!!
● Remember that we cannot access fields of the null object
● Throw the exception nullPointerException
● Thus we need to check whether the current object is null
or not, before calling the method
● In the reminder, we consider that all lists (also the void list)
contain a first cell that stores no information.

public class List
{
int element;
List next;
List(int el, List l)
{
this.element=el;
this.next=l;
}
static List EmptyList()
{
return new List(0,null);
}
boolean isEmpty()
{
return (this.next==null);
}
}
Revisiting the linked list (OO style)

int length()
{
List u=this;
int res=0;
while(u!=null) {res++;u=u.next;}
return res-1;
}
boolean belongsTo(int el)
{
List u=this.next;
while(u!=null)
{
if (el==u.element) return true;
u=u.next;
}
return false;
}

void add(int el)
{List u=this.next;
this.next=new List(el,u);
}
void delete(int el)
{
List v=this;
List w=this.next;
while(w!=null && w.element !=el)
{
v=w;
w=w.next;
}
if (w!=null) v.next=w.next;
}

void display()
{
List u=this.next;
while(u!=null)
{System.out.print(u.element+"->");
u=u.next;}
System.out.println("null");
}
static List FromArray(int [] array)
{
List u=EmptyList();
for(int i=array.length-1; i>=0; i--)
u.add(array[i]);
return u;
}

public static void main(String[] args)
{
int [] array={2,3,5,7,11,13,17,19,23};
List u=FromArray(array);
u.add(1);
u.display();
u.delete(5);
u.display();
System.out.println(u.belongsTo(17));
System.out.println(u.belongsTo(24));
}

Stacks (LIFO): Last In First Out
Two basic operations for that data-structure:
● Push: Add an element on top of the stack
● Pull: Remove the topmost element

Stacks (LIFO) using arrays
class StackArray
{
int nbmax;
int index;
int [ ] array;
// Constructors
StackArray(int n)
{
this.nbmax=n;
array=new int[nbmax]; index=-1;
System.out.println("Succesfully created a stack array object...");
}
// Methods
void Push(int element)
{
if (index<nbmax-1)
array[++index]=element; }
int Pull()
{
if (index>=0 ) return array[index--];
else return -1;
}
}

class DemoStack{
public static void main(String [] args)
{
StackArray myStack=new StackArray(10);
int i;
for(i=0;i<10;i++)
myStack.Push(i);
for(i=0;i<15;i++)
System.out.println(myStack.Pull());
}

Stacks (LIFO) using linked lists
class List
{
int element;
List next;
// Constructor
List(int el, List tail)
{
this.element=el;
this.next=tail;
}
List insertHead(int el)
{
return new List(el,this);
}
}

class Stack
{
List list;
Stack()
{
list=null;
}
void Push(int el)
{
if (list!=null)
list=list.insertHead(el);
else
list=new List(el,null);
}
int Pull()
{int val;
if (list!=null)
{val=list.element;
list=list.next;}
else val=-1;
return val;
}
}

// Use a Java package here
import java.util.Stack;
public class MainClass {
public static void main (String args[]) {
Stack s = new Stack();
s.push("A");
s.push("B");
s.push("C");
System.out.println(s);
}
}
Stacks: API

class DemoStackList
{
public static void main(String [] args)
{
Stack myStack=new Stack();
int i;
for(i=0;i<10;i++)
myStack.Push(i);
for(i=0;i<15;i++)
System.out.println(myStack.Pull());
}
}
Stacks (LIFO) using linked lists
Notice: Same code as StackArray demo program.

Static functions versus methods
●
Static (class) functionsStatic (class) functions:
Access static/local variables only.
« class methods »
Potentially many arguments in functions
●
Object methodsObject methods:
Access object fields (using this)
Access (class) static variables too.
Objects are instances of classes
Data encapsulation
(=functions with limited number of arguments)
Constructor (= field initialization)

(chapter 8) A Concise and Practical Introduction to Programming Algorithms in Java

Recommended

More Related Content

What's hot (14)

Viewers also liked (20)

Similar to (chapter 8) A Concise and Practical Introduction to Programming Algorithms in Java (20)

Recently uploaded (20)

(chapter 8) A Concise and Practical Introduction to Programming Algorithms in Java