MSc IT

1
MSc IT Computing Object Oriented Programming

• Java
• Lecture 5

2
Topics Covered

• Data Structures
• Stack
• Queue
• Vector
• Enumeration
• Hashtable

3
Data Structures
4
Data Storage

• In programming it is necessary to store and
  process collections of data
• If only a small amount of data is involved then
  unique variable names maybe used to reference
  each individual item
• rarely the case
• Usually it is necessary to create a data store to
  hold the data
• The subject of data structures studies different
  ways that data may be stored and how this
  influences the algorithms that are required to
  process it

5
Requisite Operations

• The methods or operations of a data structure
  include some or all of the following
• adding an element to the store
• removing an element
• retrieving an element
• searching for an element
• processing each element in turn
• These methods define the behaviour of the data
  structure
• There are many different implementations to
  achieve same behaviour
• The next sections identify a few common data
  structure
• and study the use of some general purpose data
  structures (container classes) provided with JDK

6
Review of Arrays
7
Review of Arrays

• An array eightnums of eight elements

first numberhas offset 0eightnums0
eighth numberhas offset 7 eightnums 7

• To declare an array specify type, name and size
• Appendix A

8
Common Data Structures

• Each of these common data structures organises
  stored data in a specific way and has a unique
  behaviour
• Stack - Last-In, First-Out behaviour (LIFO)
• Queue - First-In, First-Out behaviour (FIFO)
• Set - An unordered collection without duplicates
• Bag - A collection with duplicates
• List - Data is stored and processed in a linear
  or sequential manner
• Ordered List - Data is in a sorted order
• Tree - Hierarchical structure
• Graph - A network of objects
• Mapping - An set of associated pairs

9
OO Classes Data structures

• An object oriented class may also be considered a
  data structure
• it stores data (its attributes)
• has a behaviour (defined by its methods)
• To create the data structures above we define a
  class which specifies the required behaviour
• Then the challenge is to implement it efficiently

10
Stack

• The stack is a data structure with Last-In,
  First-Out behaviour (LIFO)
• The methods of a stack restrict access and
  processing to the top element
• place an element on the top of the stack (push)
• remove the top element (pop)
• There are no methods to process or access other
  elements
• Exercise
• think of some physical examples of stacks

11
Stack Class Specification

• Below we define the behaviour of a stack by
  specifying the methods of a Stack object
• Appendix B
• Note that the parameters and function results are
  Objects
• as every Java class inherits from Object a class
  that deals with Objects can store any class
• e.g String, Integer
• the classes that we will examine later work in
  this way
• The Stack class definition includes some extra
  methods to make it useful to a client object
• The isEmpty()method may be used to avoid
  accessing the top of an empty stack

12
Queue

• A queue is a list which restricts access to the
  ends
• The queue is a data structure with First-In,
  First-Out behaviour (FIFO)
• The methods are
• Place an element on the end of the queue (join)
• Remove the top element (leave)
• Return the length of the queue (length)
• A queue is used in many applications, especially
  where jobs must be scheduled for a particular
  process
• Variations of simple queues are priority queues
  and buffers

13
Queue Class Specification

• The class definition includes some extra methods
  to make it useful to a client object
• The isEmpty() method is particularly useful to
  prevent a run time error generated by an attempt
  to access the first element of an empty queue
• Appendix C

14
Vector

• The Vector class from java.util provides a
  dynamic array-like structure
• grows (and shrinks) as required
• If we then add the Enumeration interface (next
  section) our program can iterate through the
  elements in the container
• At any time a Vector contains a number of
  elements less than or equal to its capacity
• capacity is the space currently available, which
  will be increased as required
• Vector default capacity is 10 and it will double
  in size each time it runs out of space
• an adequate behavior in most cases

15
More Vectors

• A Vector stores references to Objects
• it is a general purpose container
• When retrieving elements they are cast to the
  type required
• no way of storing the type of object stored
• The following program demonstrates some of the
  methods of the Vector class
• Appendix D
• What method would be used to find the position of
  the String "second" ?
• Appendix G for documentation

16
Enumeration
17
Enumeration

• An Enumeration is a class that may be used to
  step through the elements of a data structure
• Two methods provided are
• hasMoreElements()
• returns true as long as there are elements left
• nextElement()
• which returns the next one as an Object
• The program demonstrates use of an Enumeration
• The Vector method elements returns an Enumeration
  of all the elements as its result
• Appendix E
• Appendix H for documentation

18
Hashtable
19
Hashtable

• A hash table is a data structure that stores
  pairs of objects
• one object is considered the key
• other object is the data associated with that key

20
Mapping

• A hash function is a mapping from keys to index
• The properties of this function are that
• for the a random set of keys, the function should
  produce a random set of keys
• the result of the function is an index within the
  range 0.. table size
• the hash function computes the same value for the
  same key each time it is applied

21
Storing Values

• The process to store an associated pair of
  objects in a hashtable is as follows
• Use the hash function to compute the index from
  the key
• Check that the table entry is free for that index
• Store the key and data in the table
• note the table is like an array

22
Collision Handling

• If at step 2 the cell is not free, a second hash
  function may be applied
• This process is repeated until an empty cell is
  found
• If no empty cell is found after several
  applications, the table is considered to be full
• Usually, we design a hashtable to be only half
  occupied at most to minimise rehashing
• This means that the chance of hitting an empty
  cell at random is high

23
Retrieving Data

• To retrieve an associated pair of objects from a
  hashtable
• use the hash function to compute the index from
  the target key
• if the cell is occupied, check that the key
  matches the target
• otherwise rehash and repeat the check
• if the key in the cell matches the target then
  the pair is found and maybe retrieved.
• if the result of hashing or rehashing finds an
  empty cell then the key is not in the table

24
Advantage

• The advantage of a hashtable over other data
  structures is that insertion and retrieval of
  data has complexity 0(1)
• That is, no matter how large the hash table,
  storing and searching for data takes
  (approximately) the same number of operations
• it is very efficient

25
Hashtable Constraints

• The constraints are
• keys must be unique, duplicate entries are not
  allowed
• the hash function must be designed to generate
  'random indices
• the table loading must not exceed 0.7 full
• otherwise efficiency drops...
• the order of the data is not important
• Hashtables are often used for look up
  applications
• e.g. Symbol table in a compiler

26
Using Hashtable

• The HashDem program demonstrates use of
  java.util.Hashtable
• Appendix F
• Appendix I for documentation
• How would you override the default size and
  loading in a Hashtable?

27
Laboratory Work

• Look at the Vector demonstration files on my
  web-site
• Make sure that you understand how they work
• Read Java Gently, p 244 254
• Attempt problem 6.6

28
Summary

• We have examined the data structures stack and
  queue
• The JDK classes Vector, Hashtable and Enumeration
  have been introduced