CSCE 210 Data Structures and Algorithms

1
CSCE 210Data Structures and Algorithms

• Prof. Amr Goneid
• AUC
• Part 1. Introduction to S/W Design

2
Introduction to S/W Design

• Software for Problem Solving
• Software Development Process
• Top-Down Design
• Elements of Module Design
• Example of a Design Document

3
1.Software for Problem Solving

• Use of S/W

Problem Domain
S/W
TO
Decide, Learn, View Interact, Play, .
4
2. Software Development Process

• Stages of S/W Production

Specify Requirements
Solution Strategy
S/W Production
S/W Testing
Problem Definition
Maintenance Upgrade
5
Software Production

Implementation (Programs)
Debugging
S/W Design
Performance Analysis / Benchmarks
6
3. Top-Down Design

• Divide and Conquer Strategy.
• Given problem X (Level 0). Divide it into
  sub-problems X1, X2, , Xn (Level 1) such that
  solving X1, then X2, then solves X completely.
• Repeat for each sub-problem until we reach basic
  or trivial implementation level.

7
Top-Down Design

• Level

0
X
1
X1
X2
X3
Xn
2
X21
X22
. .
8
Modules

• In C

Function Xi
Sub-Problem Xi
Implemented as
Main Function X
Main Problem X
Implemented as
9
4. Elements of Module Design

With What Resources? (Data)
What? Purpose
How? (Algorithm)
Module Design
10
Elements of Module Design

Functional Specs
Data Specs
Algorithm Specs
Pseudo- code
11
Elements of Module Design

• S/W Design for Each Module
• Functional Specifications the purpose of the
  module and what it is supposed to do (What to do)
• Data Specifications the data resources needed by
  the module to achieve it functionality (with
  what)
• Algorithm Specification the algorithm or
  methodology used by the module (How to do it)
• In addition, for Each Module we should specify
• Precondition the state of processing or data
  before the module is executed (state before)
• Postcondition the state of processing or data
  after the module is executed (state after)

12
Some Guidelines for Module Design

• Transparency of Purpose
• Correctness
• Completeness
• Ease of Use
• Efficiency
• Writability
• Maintainability

13
5. Example of a Design Document

• Problem Definition
• To find the median of (n) integers.
• Requirement Specification
• - Input(s) n, and list of integers in random
  order.
• - Output(s) The median.

14
Example(continued)

• Solution Strategy
• The median of a list of integers is an integer
  below which and above which there is an equal
  number of integers. To find the median of a list
  of (n) integers
• 1. Input random list of (n) integers
• 2. Sort list in ascending order.
• 3. Median is the integer at (n/2)
• 3. Print median.

15
Example(continued)

• S/W Design
• Structured (Top-Down) Design
• Yields a main module, a sorting module, and a
  median module.

main
sort
median
16
Example(continued)

• Sorting Module sort( )
• Functional Specification
• Receives list, returns sorted list.
• Input (s) n and list
• Output (s) sorted list
• Precondition none
• Postcondition the list is sorted
• Data Specification
• No extra data needed
• Algorithm Specification
• Uses the method of Selection Sort.

17
Example(continued)

• How selection sort works (optional)
• Assume list occupies locations 0.. n-1
• Repeat for list items k from 0 to n-2
• Assume smallest item to be at location (k)
• Examine items k1 to n-1 and find location of
  smallest
• Exchange that element with that at (k)

18
Example(continued)

• Median Module median( )
• Functional Specification
• Receives sorted list, returns median.
• Input (s) n and list
• Output (s) median
• Precondition list is sorted
• Postcondition none
• Data Specification
• No extra data needed
• Algorithm Specification
• median is the element at (n/2) in the sorted
  list

19
Example(continued)

• Main Module main( )
• Functional Specification
• Inputs list, prints median.
• Input (s) n and list (form keyboard)
• Output (s) median (on screen)
• Precondition none
• Postcondition median is printed
• Data Specification
• - MAX 100 to represent maximum list size
• - An Array (a) of size MAX to store list.
• - n is the actual list size.

20
Example(continued)

• main( ) (continued)
• Algorithm Specification
• Input n.
• Check that n does not exceed MAX.
• Input random list into array (a).
• Call sorting module.
• Call median module
• Display median

21
Example (implementation)

• include ltiostreamgt
• using namespace std
• typedef int itemtype
• // Function Prototypes
• void sort (itemtype a , int n)
• itemtype median (itemtype a , int n)
• //________________________________________________
  __________
• int main( ) // Start of main function
• const int MAX 100 // Data Declarations
• itemtype aMAX, m
• int k, n
• cin gtgt n // Read size of data
• if (n gt MAX) n MAX // to prevent array
  overflow

22
Example (implementation)

• for (k 0 k lt n k) cin gtgt ak // Read
  data into array
• sort(a,n) // Call sorting module
• m median(a,n) // Call median module
• cout ltlt m // Display median
• return 0
• //________________________________________________
  ___

23
Example (implementation)

• void sort (itemtype a , int n) // Sort module
• // Selection Sort Algorithm
• // Precondition None
• // Postcondition Array a is sorted
• int k, j, min // Module Local Data
• itemtype temp
• for (k 0 k lt n-1 k )
• min k
• for ( j k1 j lt n j) if (aj lt amin )
  min j
• temp amin amin ak ak temp
• //________________________________________________
  ____

24
Example (implementation)

• itemtype median (itemtype a , int n) //median
  module
• // Computes median
• // Precondition Array a is sorted
• // Postcondition None
• return (an/2)
• //________________________________________________
  ____

25
Learn on your own about

• Waterfall and Spiral Models
• UML
• S/W system Life Cycle