Principles of programming

Principles of Programming and

Software Engineering

1-1

Coding without a solution design increases debugging time

A team of programmers for a large software development project involvesAn overall planOrganizationCommunication

Software engineeringProvides techniques to facilitate the

development of computer programs

1-2

Problem Solving and Software Engineering

Problem solvingThe process of taking the statement of a

problem and developing a computer program that solves that problem

A solution consists of:Algorithms

Algorithm: a step-by-step specification of a function to solve a problem within a finite amount of time

Ways to store data

1-3

What is Problem Solving?

1-4

The Life Cycle of Software

Figure 1.1The life cycle of software as a waterwheel that can rotate from one phase to any other phase

Phase 1: SpecificationAspects of the problem which must be

specified:What is the input data?What data is valid and what data is invalid?Who will use the software, and what user interface

should be used?What error detection and error messages are

desirable?What assumptions are possible?Are there special cases?What is the form of the output?What documentation is necessary?What enhancements to the program are likely in the

future?

1-5


Phase 1: Specification (Continued)Prototype program

A program that simulates the behavior of portions of the desired software product

Phase 2: DesignIncludes:

Dividing the program into modulesSpecifying the purpose of each moduleSpecifying the data flow among modules

1-6


Phase 2: Design (Continued)Modules

Self-contained units of codeShould be designed to be:

Loosely coupledHighly cohesive

InterfacesCommunication mechanisms among modules

1-7


Phase 2: Design (Continued)Specifications of a function

A contract between the function and the module that calls it

Should not commit the function to a particular way of performing its task

Include the function’s preconditions and postconditions

Phase 3: Risk AnalysisBuilding software entails risksTechniques exist to identify, assess, and

manage the risks of creating a software product

1-8


Phase 4: VerificationFormal methods can be used to prove that an

algorithm is correctAssertion

A statement about a particular condition at a certain point in an algorithm

InvariantA condition that is always true at a particular point

in an algorithm

1-9


Phase 4: Verification (Continued)Loop invariant

A condition that is true before and after each execution of an algorithm’s loop

Can be used to detect errors before coding is startedThe invariant for a correct loop is true:

Initially, after any initialization steps, but before the loop begins execution

Before every iteration of the loopAfter every iteration of the loopAfter the loop terminates

1-10


Phase 5: Coding (Continued)Involves:

Translating the design into a particular programming language

Removing syntax errorsPhase 6: Testing

Involves:Removing the logical errors

1-11


Phase 6: Testing (Continued)Test data should include:

Valid data that leads to a known resultInvalid dataRandom dataActual data

Phase 7: Refining the SolutionMore sophisticated input and output routinesAdditional featuresMore error checks

1-12


Phase 8: ProductionInvolves:

Distribution to the intended usersUse by the users

Phase 9: MaintenanceInvolves

Correcting user-detected errorsAdding more featuresModifying existing portions to suit the users better

1-13


A solution is good if:The total cost it incurs over all phases of its life

cycle is minimal

1-14

What is a Good Solution?

The cost of a solution includes:Computer resources that the program

consumesDifficulties encountered by usersConsequences of a program that does not

behave correctlyPrograms must be well structured and

documentedEfficiency is one aspect of a solution’s cost

1-15

What is a Good Solution?

AbstractionSeparates the purpose of a module from its

implementationSpecifications for each module are written

before implementationFunctional abstraction

Separates the purpose of a function from its implementation

1-16

Achieving a Modular Design: Abstraction

Data abstractionFocuses of the operations of data, not on the

implementation of the operationsAbstract data type (ADT)

A collection of data and operations on the dataAn ADT’s operations can be used without knowing

how the operations are implemented, if the operations’ specifications are known

1-17

Achieving a Modular Design

Data structureA construct that can be defined within a

programming language to store a collection of data

1-18


Information hidingHide details within a moduleEnsure that no other module can tamper with

these hidden detailsPublic view of a module

Described by its specificationsPrivate view of a module

Consists of details which should not be described by the specifications

1-19


Principles of object-oriented programming (OOP)Encapsulation

Objects combine data and operationsInheritance

Classes can inherit properties from other classesPolymorphism

Objects can determine appropriate operations at execution time

1-20

Object-Oriented Design

Advantages of an object-oriented approachExisting classes can be reusedProgram maintenance and verification are

easier

1-21

Object-Oriented Design

Object-oriented design (OOD)Produces modular solutions for problems that

primarily involve dataIdentifies objects by focusing on the nouns in

the problem statement

1-22

Top-Down Design

Top-down design (TDD)Produces modular solutions for problems in

which the emphasis is on the algorithmsIdentifies actions by focusing on the verbs in

the problem statementA task is addressed at successively lower levels

of detail

1-23

Top-Down Design

1-24

Top-Down Design

Figure 1.4

A structure chart showing the hierarchy of modules

Use OOD and TDD togetherUse OOD for problems that primarily involve

dataUse TDD to design algorithms for an object’s

operations

1-25

General Design Guidelines

Consider TDD to design solutions to problems that emphasize algorithms over data

Focus on what, not how, when designing both ADTs and algorithms

Consider incorporating previously written software components into your design

1-26

General Design Guidelines

The Unified Modeling Language (UML) is a modeling language used to express object-oriented designsClass diagrams specify the name of the class,

the data members of the class, and the operations

1-27

Modeling Object-Oriented Design Using UML

UML (Continued)Relationships between classes can be shown by

connecting classes with linesInheritance is shown with a line and an open

triangle to the parent classContainment is shown with an arrow to the

containing class UML provides notation to specify visibility,

type, parameter, and default value information

1-28

Modeling Object-Oriented Design Using UML

1-29

UML DiagramsFigure 1.6

UML diagram for

a banking system

Modularity has a favorable impact onConstructing programsDebugging programsReading programsModifying programsEliminating redundant code

1-30

A Summary of Key Issues in Programming

Modifiability is possible through the use ofFunctionNamed constantsThe typedef statement

1-31


Ease of use In an interactive environment, the program

should prompt the user for input in a clear manner

A program should always echo its inputThe output should be well labeled and easy to

read

1-32


Fail-Safe ProgrammingFail-safe programs will perform reasonably no

matter how anyone uses itTest for invalid input data and program logic

errorsHandle errors through exception handling

1-33


Eight issues of personal style in programmingExtensive use of functionsUse of private data fieldsAvoidance of global variables in functionsProper use of reference arguments

1-34


Eight issues of personal style in programming (Continued)Proper use of functionsError handlingReadabilityDocumentation

1-35


DebuggingProgrammer must systematically check a

program’s logic to determine where an error occurs

Tools to use while debugging:WatchesBreakpointscout statementsDump functions

1-36


Software engineering: techniques to facilitate development of programs

Software life cycle consists of nine phasesLoop invariant: property that is true before

and after each iteration of a loopEvaluating the quality of a solution must

consideration a variety of factors

1-37

Summary

A combination of object-oriented and top-down design techniques leads to a modular solution

The final solution should be as easy to modify as possible

A function should be as independent as possible and perform one well-defined task

1-38

Summary

Functions should include a comment: purpose, precondition, and postcondition

A program should be as fail-safe as possibleEffective use of available diagnostic aids is

one of the keys to debuggingUse “dump functions” to help to examine and

debug the contents of data structures

1-39

Summary