Arrays , Link Lists, Stacks and Queues

Deliverables

Arrays
Link Lists
Stacks
Queues
6/5/2012 7:14 PM Copyright@ gdeepak.com 2
 Arrays
Arrays are stored in contiguous memory locations and contain similar
data. Position of any array element can be accessed from its index
position with help of array’s starting address.

An element can be accessed, inserted or removed by specifying its

position (number of elements preceding it)

Vector is abstraction of array data structure, where we can access by

rank.

6/5/2012 7:14 PM Copyright@ gdeepak.com 3

 Array operations
• Deletion: In operation removeAtPosition(r), we need to
fill the hole left by the removed element by shifting
backward n-r-1 elements A[r + 1], …, A[n - 1]. In the
worst case (r = 0), this takes O(n) time. In the average
case also it is O(n).
• A[0] A[1] A[2] A[3] A[4] A[5] A[6] A[7] A[8] A[9]

3 10 0 -135 2 21 41 6 87 11

• Is it necessary to shift the array elements after deletion?

6/5/2012 7:14 PM Copyright@ gdeepak.com 4

 Deletion of -34
5
23
-34
1
81
9 After deletion of -34
123
45
2
-31

6/5/2012 7:14 PM Copyright@ gdeepak.com 5

 Insertion
Insertion in the array again requires shifting all the
elements one index ahead to make room for insertion.

Need to understand shifting different in deletion and

insertion, otherwise data may be lost.

In Insertion shifting is to start from the last element of

the array and in deletion shifting is to start from the next
array index from where the element is to be deleted.

6/5/2012 7:14 PM Copyright@ gdeepak.com 6

 Inserting 100 after -34
5
23
-34
1
81 100
9
123
45
2

6/5/2012 7:14 PM Copyright@ gdeepak.com 7

 Array complexity
Accessing an array: O(1) time

Copying an array: O(n) time

Merging two arrays: O(m+n) time

Splitting an array: O(n) time

Intersection of two arrays: O(n2)

Union of two arrays: O(n2)

6/5/2012 7:14 PM Copyright@ gdeepak.com 8

 Array incrementing by C or doubling
Let n be the number of elements in the array. The space
used is O(n) and each operation runs in time O(1)

The maximum size of the array must be defined apriori

and cannot be changed.

Incremental strategy: increase the size by a constant C

Doubling strategy: double the size

6/5/2012 7:14 PM Copyright@ gdeepak.com 9

 Array incrementing by C or Doubling
If increment by C every time then
Creation copying insertion (iteration wise)
ci c(i-1) c
Total = ci+ci-c+c=2ci

If there are n elements in end then total iterations will be

n/c , So it is 2c(1+2+3+……..n/c) = n n
2c( )(  1) 2
c c n
  O(n 2 )
2 c
6/5/2012 7:14 PM Copyright@ gdeepak.com 10
 Array incrementing by C or Doubling
If we double the array every time then the creation,
copying, insertion cost will be as follows

Creation copying insertion (iteration wise)

2i 2i-1 2i-1
Total cost in every iteration is 2i+( 2i-1+2i-1) =2i+2i=2i+1

if there are n elements to be inserted then there will be lgn

iterations , total becomes (21 +22+23+…+ 2lgn+1)=4n-1=O(n)

6/5/2012 7:14 PM Copyright@ gdeepak.com 11

 Link List
Singly linked list is a data structure
consisting of a sequence of nodes

Each node stores element and link to the

next node

There may be a header link and trailer

link

6/5/2012 7:14 PM Copyright@ gdeepak.com 12

 Link list operations

Access (): Accessing an element from a list

Worst case: O (n)

Average Case: 1+2+3+…n = n(n-1)/2 divided by n=n-1/2 = O(n)

Delete (): O(1) complexity, but it is preceded by accessing that

position, effectively taking O(n) time. Delete first takes O(1) time

6/5/2012 7:14 PM Copyright@ gdeepak.com 13

 Link list complexity
Insert : O(1) but inserting before or after an element will
require accessing element effectively taking O(n) time.
Inserting at first place is O(1).
Merging : O(1) If the trailer node information is not
available then it will be equal to the length of the smaller
list to reach at the end of the list for merging with the
second list
Size: O(n) time
Intersection: O(n2)
Union: O(n2)

6/5/2012 7:14 PM Copyright@ gdeepak.com 14

 Doubly link list

A doubly linked list provides a natural implementation of the

List

Nodes implement Position and stores element, link to the

previous node and link to the next node

Special trailer and header nodes. Deletion an Insertion

becomes easy as there is no need to remember the address of
the previous node.

6/5/2012 7:14 PM Copyright@ gdeepak.com 15

 Insertion

6/5/2012 7:14 PM Copyright@ gdeepak.com 16

 Deletion

6/5/2012 7:14 PM Copyright@ gdeepak.com 17

 Circular link list

There will be no difference in the time

complexity of various operations

Circular Link List and circular doubly link list

Input output buffers use circular queue for
buffering.

6/5/2012 7:14 PM Copyright@ gdeepak.com 18

 Comparison link list vs. array
It is easier to delete in the link list

it is easier to insert in the link list

It is easier to access in the array

We have to predefine the array so there can be wastage of

memory

6/5/2012 7:14 PM Copyright@ gdeepak.com 19

 Comparison link list vs. array
Array Size has to be defined and is limited to the defined size while link
list can grow to limit of memory

Continuous allocation is required for array, while this is not with link
list

Arrays compute address of data elements by arithmetic operations

while in linked list addresses are in structure itself.

Arrays can be accessed backward and forward but we have to use

special link lists like doubly linked list for backward access

6/5/2012 7:14 PM Copyright@ gdeepak.com 20

 Comparison link list vs. array
Arrays definition is part of the language construct but link
list we have to create

Merging two arrays is very difficult while merging link lists

is easy

There will be more cache misses in the Link list then in the
arrays

With large records, moving pointers is easier and faster

than moving the items themselves.

6/5/2012 7:14 PM Copyright@ gdeepak.com 21

 Comparison link list vs. array
There can be an array of pointers associated with a link list,
in that case address part of the link list is not required.

In link lists two different processors may work on two

different parts of link list.

Due to address part there is wastage of memory in link list,

it is not much if we look at practical size of record of name,
age etc.

6/5/2012 7:14 PM Copyright@ gdeepak.com 22

 Stack
Last in First out,
in and out only Page-visited
Undo sequence
from one end, history in a Web
in a text editor
other end is browser
closed

Chain of method
calls in the Java In other data
Recursion
Virtual Machine, Structures
Function Calls

6/5/2012 7:14 PM Copyright@ gdeepak.com 23

 Simulating Pop and Push

6
98
37
25
12
5
6/5/2012 7:14 PM Copyright@ gdeepak.com 24
 Stack Example

6/5/2012 7:14 PM Copyright@ gdeepak.com 25

 Stack
Creation of an stack: O(n)
push(object): inserts an element O(1)
pop(): removes and returns the last inserted element. O(1)
top(): returns last inserted element without removing O(1)
size(): returns number of elements stored. O(1)
isEmpty(): indicates whether no elements are stored. O(1)
isFull(): Indicates whether array limit is over. O(1)
6/5/2012 7:14 PM Copyright@ gdeepak.com 26
 Link List Vs. Array Implementation
Pop
Array: return element at “top” and decrement “top”
Linked list: return and remove at front of linked list

Push
Array: increment “top” and insert element. Must check for overflow!
Linked list: insert element at front of linked list

Array: more memory efficient

Linked list: don’t have to worry about “overflow”

6/5/2012 7:14 PM Copyright@ gdeepak.com 27

 Axioms

pop(push(S, v))=S

Top(push(S, v))=v

Top(pop(push(push(S,12),15))) = top(push(S,12))

6/5/2012 7:14 PM Copyright@ gdeepak.com 28

 Queues
Insertions and deletions follow the first-in first-out
scheme

Insertions are at the rear of the queue and removals are

at the front of the queue

Two variables keep track of the front and rear. Careful

implementation is necessary to avoid overlapping.

Operations complexity may vary depending upon the

implementation, whether front or rear is fixed or not

6/5/2012 7:14 PM Copyright@ gdeepak.com 29

 Queue Example

6/5/2012 7:14 PM Copyright@ gdeepak.com 30

 Simulating Enqueue and Dequeue

5 45 71

Perform Operations
Enqueue 99
Dequeue
Dequeue
99
6/5/2012 7:14 PM Copyright@ gdeepak.com 31
 Queues

enque(Object o): deque(): removes front(): returns

inserts an and returns front element at front
element o at end element of without
of queue. O(1) queue. O(1) removing it. O(1)

isEmpty():return
size(): returns
s Boolean
number of
indicating queue
elements stored.
is empty or not
O(1)
O(1)

6/5/2012 7:14 PM Copyright@ gdeepak.com 32

 Queue Applications
Waiting lists, bureaucracy

Access to shared resources (e.g. printer)

Multiprogramming

Tunnel

In other Data Structures

6/5/2012 7:14 PM Copyright@ gdeepak.com 33
 Axioms

Front(Enqueue(Enqueue(Q, w),v))=front(Enqueue(Q, w))

Deque(Enque(Enque(Q, w),v))=Enque(Deque(Enque(Q, w)),v)

In an Enqueue operation, when the array is full, instead of

throwing an exception, we can replace the array with a larger
one, Similar to what we did for an array based stack.

6/5/2012 7:14 PM Copyright@ gdeepak.com 34

 Circular queue

Circular Queue is
essentially implemented
as a linear array in a
wrapped around fashion
where even rear pointer
r f
of the queue can come
before the front pointer
6/5/2012 7:14 PM Copyright@ gdeepak.com 35
 Questions, Comments and Suggestions

6/5/2012 7:14 PM Copyright@ gdeepak.com 36

 Question 1
What is result of Last operation in given sequence :
Push(&);Push($);Push(@);Top();Pop();Push(#);
Pop();Pop();Push(%);Pop();

6/5/2012 7:14 PM Copyright@ gdeepak.com 37

 Question 2
What will be the result of last result of given priority queue
if number 1 is the highest priority
Enqueue(3), Enqueue(6), Enqueue(2),dequeue, dequeue,
Enqueue(5),Enqueue(1),dequeue, dequeue

6/5/2012 7:14 PM Copyright@ gdeepak.com 38

 Question 3
In a Singly circular linked list organization, insertion of a
record involves modification of
A. One pointer
B. Two pointers
C. Three pointers
D. Four pointer
E. No Pointer

6/5/2012 7:14 PM Copyright@ gdeepak.com 39