Exception Handling - PowerPoint Presentation

Slide1

Exception Handling

Andy Wang

Object Oriented Programming in C++

COP 3330Slide2

Exception Handling

A type of error checking, available in many programming languages

Exception

Some sort of problem or error that occurs during a program’s execution

Many error situations could occur besides those that we usually check for (usually related to things like user input)

Exception handler

A piece of code that resolves an exception situation

Typical error check often intermixed with the tasks of a program (if statements,

etc

)

Exception handlers are intended to be separate from the main tasksSlide3

Why?

Intermixing program logic with the error-checking code can sometimes make programs hard to read, debug, etc.

Many potential problems are very infrequent

Exception handlers are separate from main tasks of a program

Can improve clarity and modifiability

Exception handling can improve a program’s fault toleranceSlide4

When?

This does not mean that exception handlers should be used in all cases

Sometimes conventional error checking is more appropriate

Exception handling best for problems that occur infrequently

Errors that will result in termination of the program

Not for user input checking

Good for setting up uniform techniques for error handling when many programmers and multiple modules are involvedSlide5

How?

Reserved words in C++: try throw, catch

t

ry

blocks

Consists of keyword try and a block of code inside { }

Encloses the statements that might cause exceptions

catch

blocks

1+ catch blocks follow a try block (also in { })

Each catch block is an exception handler

A cache block has a single parameter (with type listed)Slide6

How?

If an exception occurs in a try block

The try block immediately ends

Program attempts to match the exception to one of the catch handlers (based on type of item thrown)

If a match is found, the code in the catch block executes

Only one catch block will be matched, if any

Program control resumes after the last catch block

If no exception occur in a try block, the catch blocks are skipped

A point where an exception occurs is the throw point

Keyword throw used to throw an exception to be caughtSlide7

How?

In C++, there is a standard library with pre-built exception classes

#include <exception>

u

sing

std

::exception;

When a function intends to throw an exception back to the caller (not handled internally), it’s good to tell the caller what to expect via a

throw list

when declaring a function

v

oid

someFunction

throw (

DivideByZero

,

OtherException

);

This can be used to

limit the type of exceptions that a function is allowed to throwSlide8

How?

v

oid someFunction1() throw(); // empty throw list

v

oid someFunction2(); // no throw list

The first function can throw no exceptions to the outside

The second can throw exceptions of any kindSlide9

Simple Example

http://www.cs.fsu.edu/~myers/cop3330/examples/exceptions/except.cppSlide10

Main.cpp

#include <

iostream

>

u

sing namespace

std

;

i

nt

main() {

int

cookies, people;

double

cpp

;

try {

cout

<< “Enter number of people: “;

cin

>> people;

cout

<< “Enter number of cookies: “;

cin

>> cookies

if (cookies == 0) throw people;

if (cookies < 0) throw

static_cast

<double>(people);Slide11

Main.cpp

cpp

= cookies/

static_cast

<double>(people);

cout

<< cookies << “ cookies.\n” << people

<< “ people.\n” << “You have “ <<

cpp

<< “ cookies per person.\n”;

}

catch(

int

e) {

cout

<< e << “ people, and no cookies!\

nGo

buy some cookies!\n”;

}

catch(double t) {

cout

<< “Second catch block type double – do we reach it?\n”;

}

cout

<< “End of program.\n”;

return 0;

}Slide12

Negative Number Example

http://www.cs.fsu.edu/~myers/savitch3c++/Ch18/18-04.cppSlide13

18-04.cpp

#include <

iostream

>

#include <string>

u

sing namespace

std

;

c

lass

NegativeNumber

{

public:

NegativeNumber

() {}

NegativeNumber

(string m): message(m) {}

string

getMessage

()

const

( return message; }

private:

string message;

};

c

lass

DivideByZero

{};Slide14

18-04.cpp

i

nt

main() {

int

pencils, erasers;

double

ppe

;

try {

cout

<< “How many pencils do you have?\n”;

cin

>> pencils;

if (pencils < 0) throw

NegativeNumber

(“pencils”);

cout

<< “How many erasers do you have?\n”;

cin

>> erasers;

if (erasers < 0) throw

NegativeNumber

(“erasers”);

if (erasers != 0)

ppe

=

pensils

/

static_cast

<double>(erasers);

else

throw

DivideByZero

();Slide15

18-04.cpp

cout

<< “Each eraser must last through “ <<

ppe

<< pencils.\n”;

}

catch(

NegativeNumber

e) {

cout

<< “Cannot have a negative number of “

<<

e.getMessage

() <<

endl

;

}

catch(

DivideByZero

) {

cout

<< “Do not make any mistakes.\n”;

}

cout

<< “End of program.\n”;

return 0;

}Slide16

Safe Divide Example

http://www.cs.fsu.edu/~myers/savitch3c++/Ch18/18-05.cppSlide17

18-05.cpp

#include <

iostream

>

#include <

cstdlib

>

u

sing namespace

std

;

c

lass

DivideByZero

{};

double

safeDevide

(

int

top,

int

bottom) throw (

DivideByZero

) {

if (bottom == 0)

thorw

DivideByZero

();

return top/

static_cast

<double>(bottom);

}Slide18

18-05.cpp

i

nt

main() {

int

numerator, denominator;

double quotient;

cout

<< “Enter numerator:\n”;

cin

>> numerator;

cout

<< “Enter denominator:\n”;

cin

>> denominator;

try {

quotient =

safeDivide

(numerator, denominator);

}

catch(

DivideByZero

) {

cout

<< “Error: Division by zero!\n”;

<< “Program aborting.\n”;

exit(0);

}

cout

<< numerator << “/” << denominator << “ = “

<< quotient <<

endl

;

return 0;

}