Deitel® Series Page

Deitel® Series Page
How To Program Series
Android How to Program
C++ How to Program, 8/E
C How to Program, 7/E
Java™ How to Program, 9/E
Java™ How to Program, Late Objects Version, 8/E
Internet & World Wide Web How to Program, 5/E
Visual C++® 2008 How to Program, 2/E
Visual Basic® 2010 How to Program
Visual C#® 2010 How to Program, 3/E
Simply Series
Simply C++: An App-Driven Tutorial Approach
Simply Java™ Programming: An App-Driven
Tutorial Approach
Simply C#: An App-Driven Tutorial Approach
Simply Visual Basic® 2010: An App-Driven
Approach, 4/E
CourseSmart Web Books
www.deitel.com/books/CourseSmart/
C++ How to Program, 5/E, 6/E, 7/E & 8/E
Simply C++: An App-Driven Tutorial Approach
Java™ How to Program, 6/E, 7/E, 8/E & 9/E
Simply Visual Basic 2010: An App-Driven
Approach, 4/E
(continued from previous column)
Visual Basic® 2010 How to Program
Visual Basic® 2008 How to Program
Visual C#® 2010 How to Program, 4/E
Visual C#® 2008 How to Program, 3/E
Deitel® Developer Series
AJAX, Rich Internet Applications and Web
Development for Programmers
Android for Programmers: An App-Driven
Approach
C++ for Programmers
C# 2010 for Programmers
iPhone® for Programmers: An App-Driven Approach
Java™ for Programmers, 2/e
JavaScript for Programmers
LiveLessons Video Learning Products
www.deitel.com/books/LiveLessons/
Android App Development Fundamentals
C++ Fundamentals
Java™ Fundamentals
C# 2010 Fundamentals
iPhone® App Development Fundamentals
JavaScript Fundamentals
Visual Basic Fundamentals
To receive updates on Deitel publications, Resource Centers, training courses, partner offers and more,
please register for the free Deitel ® Buzz Online e-mail newsletter at:
www.deitel.com/newsletter/subscribe.html
and join the Deitel communities on Twitter®
@deitel
Facebook®
facebook.com/DeitelFan
and Google+
gplus.to/deitel
To communicate with the authors, send e-mail to:
deitel@deitel.com
For information on government and corporate Dive-Into® Series on-site seminars offered by Deitel &
Associates, Inc. worldwide, visit:
www.deitel.com/training/
or write to
deitel@deitel.com
For continuing updates on Prentice Hall/Deitel publications visit:
www.deitel.com
www.pearsonhighered.com/deitel/
Visit the Deitel Resource Centers that will help you master programming languages, software development, Android and iPhone/iPad app development, and Internet- and web-related topics:
www.deitel.com/ResourceCenters.html
Paul Deitel
Deitel & Associates, Inc.
Harvey Deitel
Deitel & Associates, Inc.
Editorial Director: Marcia J. Horton
Editor-in-Chief: Michael Hirsch
Associate Editor: Carole Snyder
Vice President, Marketing: Patrice Jones
Marketing Manager: Yezan Alayan
Marketing Coordinator: Kathryn Ferranti
Vice President, Production: Vince O’Brien
Managing Editor: Jeff Holcomb
Associate Managing Editor: Robert Engelhardt
Operations Specialist: Lisa McDowell
Art Director: Anthony Gemmellaro
Cover Design: Paul Deitel, Harvey Deitel, Abbey Deitel, Anthony Gemmellaro
Cover Photo Credit: Excellent backgrounds/Shutterstock.com
Media Editor: Daniel Sandin
Credits and acknowledgments borrowed from other sources and reproduced, with permission, in this textbook appear
on page vi.
The authors and publisher of this book have used their best efforts in preparing this book. These efforts include the
development, research, and testing of the theories and programs to determine their effectiveness. The authors and publisher make no warranty of any kind, expressed or implied, with regard to these programs or to the documentation
contained in this book. The authors and publisher shall not be liable in any event for incidental or consequential damages in connection with, or arising out of, the furnishing, performance, or use of these programs.
Copyright © 2013, 2010, 2007, 2004, 2001 Pearson Education, Inc., publishing as Prentice Hall. All rights reserved.
Manufactured in the United States of America. This publication is protected by Copyright, and permission should be
obtained from the publisher prior to any prohibited reproduction, storage in a retrieval system, or transmission in any
form or by any means, electronic, mechanical, photocopying, recording, or likewise. To obtain permission(s) to use
material from this work, please submit a written request to Pearson Education, Inc., Permissions Department, One
Lake Street, Upper Saddle River, New Jersey 07458, or you may fax your request to 201-236-3290.
Many of the designations by manufacturers and sellers to distinguish their products are claimed as trademarks. Where
those designations appear in this book, and the publisher was aware of a trademark claim, the designations have been
printed in initial caps or all caps.
Library of Congress Cataloging-in-Publication Data
Deitel, Paul J.
C : how to program / Paul Deitel, Deitel & Associates, Inc., Harvey Deitel, Deitel & Associates,
Inc., Abbey Deitel, Deitel & Associates, Inc. -- Seventh edition.
pages cm -- (How to program series)
ISBN 978-0-13-299044-8
1. C (Computer program language) 2. C++ (Computer program language) 3. Java (Computer program
language) I. Deitel, Harvey M., II. Deitel, Abbey. III. Title.
QA76.73.C15D44 2012
005.13'3--dc23
2011051087
10 9 8 7 6 5 4 3 2 1
ISBN-10: 0-13-299044-X
ISBN-13: 978-0-13-299044-8
In Memory of Dennis Ritchie,
creator of the C programming language
and co-creator of the UNIX operating system.
Paul and Harvey Deitel
Trademarks
DEITEL, the double-thumbs-up bug and DIVE INTO are registered trademarks of Deitel and Associates,
Inc.
MICROSOFT AND/OR ITS RESPECTIVE SUPPLIERS MAKE NO REPRESENTATIONS
ABOUT THE SUITABILITY OF THE INFORMATION CONTAINED IN THE DOCUMENTS
AND RELATED GRAPHICS PUBLISHED AS PART OF THE SERVICES FOR ANY PURPOSE.
ALL SUCH DOCUMENTS AND RELATED GRAPHICS ARE PROVIDED "AS IS" WITHOUT
WARRANTY OF ANY KIND. MICROSOFT AND/OR ITS RESPECTIVE SUPPLIERS HEREBY
DISCLAIM ALL WARRANTIES AND CONDITIONS WITH REGARD TO THIS INFORMATION, INCLUDING ALL WARRANTIES AND CONDITIONS OF MERCHANTABILITY,
WHETHER EXPRESS, IMPLIED OR STATUTORY, FITNESS FOR A PARTICULAR PURPOSE,
TITLE AND NON-INFRINGEMENT. IN NO EVENT SHALL MICROSOFT AND/OR ITS
RESPECTIVE SUPPLIERS BE LIABLE FOR ANY SPECIAL, INDIRECT OR CONSEQUENTIAL
DAMAGES OR ANY DAMAGES WHATSOEVER RESULTING FROM LOSS OF USE, DATA
OR PROFITS, WHETHER IN AN ACTION OF CONTRACT, NEGLIGENCE OR OTHER
TORTIOUS ACTION, ARISING OUT OF OR IN CONNECTION WITH THE USE OR PERFORMANCE OF INFORMATION AVAILABLE FROM THE SERVICES.
THE DOCUMENTS AND RELATED GRAPHICS CONTAINED HEREIN COULD INCLUDE
TECHNICAL INACCURACIES OR TYPOGRAPHICAL ERRORS. CHANGES ARE PERIODICALLY ADDED TO THE INFORMATION HEREIN. MICROSOFT AND/OR ITS RESPECTIVE
SUPPLIERS MAY MAKE IMPROVEMENTS AND/OR CHANGES IN THE PRODUCT(S) AND/
OR THE PROGRAM(S) DESCRIBED HEREIN AT ANY TIME. PARTIAL SCREEN SHOTS
MAY BE VIEWED IN FULL WITHIN THE SOFTWARE VERSION SPECIFIED.
Contents
Appendices E through H are PDF documents posted online at the book’s Companion
Website (located at www.pearsonhighered.com/deitel).
Preface
1
1.1
1.2
1.3
1.4
1.5
1.6
1.7
1.8
1.9
1.10
1.11
xix
Introduction to Computers, the Internet
and the Web
Introduction
Computers and the Internet in Industry and Research
Hardware and Software
1.3.1 Moore’s Law
1.3.2 Computer Organization
Data Hierarchy
Programming Languages
The C Programming Language
C Standard Library
C++ and Other C-Based Languages
Object Technology
Typical C Program Development Environment
1.10.1 Phase 1: Creating a Program
1.10.2 Phases 2 and 3: Preprocessing and Compiling a C Program
1.10.3 Phase 4: Linking
1.10.4 Phase 5: Loading
1.10.5 Phase 6: Execution
1.10.6 Problems That May Occur at Execution Time
1.10.7 Standard Input, Standard Output and Standard Error Streams
Test-Driving a C Application in Windows, Linux and Mac OS X
1.11.1 Running a C Application from the Windows
Command Prompt
1.12
1.11.2 Running a C Application Using GNU C with Linux
1.11.3 Running a C Application Using GNU C with Mac OS X
Operating Systems
1.12.1 Windows—A Proprietary Operating System
1.12.2 Linux—An Open-Source Operating System
1.12.3 Apple’s Mac OS X; Apple’s iOS for iPhone®, iPad® and
iPod Touch® Devices
1.12.4 Google’s Android
1
2
2
5
6
6
7
9
10
12
13
14
16
16
16
18
18
18
18
18
19
20
22
25
27
28
28
29
29
viii
Contents
1.13
1.14
1.15
1.16
The Internet and World Wide Web
Some Key Software Development Terminology
Keeping Up-to-Date with Information Technologies
Web Resources
2
Introduction to C Programming
2.1
2.2
2.3
2.4
2.5
2.6
2.7
Introduction
A Simple C Program: Printing a Line of Text
Another Simple C Program: Adding Two Integers
Memory Concepts
Arithmetic in C
Decision Making: Equality and Relational Operators
Secure C Programming
3
Structured Program Development in C
3.1
3.2
3.3
3.4
3.5
3.6
3.7
3.8
3.9
3.11
3.12
3.13
Introduction
Algorithms
Pseudocode
Control Structures
The if Selection Statement
The if…else Selection Statement
The while Repetition Statement
Formulating Algorithms Case Study 1: Counter-Controlled Repetition
Formulating Algorithms with Top-Down, Stepwise Refinement
Case Study 2: Sentinel-Controlled Repetition
Formulating Algorithms with Top-Down, Stepwise Refinement
Case Study 3: Nested Control Statements
Assignment Operators
Increment and Decrement Operators
Secure C Programming
4
C Program Control
4.1
4.2
4.3
4.4
4.5
4.6
4.7
4.8
4.9
4.10
4.11
Introduction
Repetition Essentials
Counter-Controlled Repetition
for Repetition Statement
for Statement: Notes and Observations
Examples Using the for Statement
switch Multiple-Selection Statement
do…while Repetition Statement
break and continue Statements
Logical Operators
Confusing Equality (==) and Assignment (=) Operators
3.10
30
31
33
34
40
41
41
45
49
50
54
58
70
71
71
71
72
74
75
79
80
82
89
93
93
96
114
115
115
116
117
120
121
124
130
132
134
137
Contents
4.12
4.13
Structured Programming Summary
Secure C Programming
5
C Functions
5.1
5.2
5.3
5.4
5.5
5.6
5.7
5.8
5.9
5.10
5.11
5.12
5.13
5.14
5.15
5.16
5.17
Introduction
Program Modules in C
Math Library Functions
Functions
Function Definitions
Function Prototypes: A Deeper Look
Function Call Stack and Stack Frames
Headers
Passing Arguments By Value and By Reference
Random Number Generation
Example: A Game of Chance
Storage Classes
Scope Rules
Recursion
Example Using Recursion: Fibonacci Series
Recursion vs. Iteration
Secure C Programming
6
C Arrays
6.1
6.2
6.3
6.4
6.5
6.6
6.7
6.8
6.9
6.10
6.11
Introduction
Arrays
Defining Arrays
Array Examples
Passing Arrays to Functions
Sorting Arrays
Case Study: Computing Mean, Median and Mode Using Arrays
Searching Arrays
Multidimensional Arrays
Variable-Length Arrays
Secure C Programming
7
C Pointers
7.1
7.2
7.3
7.4
7.5
Introduction
Pointer Variable Definitions and Initialization
Pointer Operators
Passing Arguments to Functions by Reference
Using the const Qualifier with Pointers
7.5.1 Converting a String to Uppercase Using a Non-Constant
Pointer to Non-Constant Data
ix
138
143
158
159
159
160
162
162
166
169
172
173
174
179
182
184
187
191
194
197
216
217
217
218
219
232
236
239
244
249
256
259
277
278
278
279
282
284
287
x
Contents
7.5.2
7.6
7.7
7.8
7.9
7.10
7.11
7.12
7.13
Printing a String One Character at a Time Using a
Non-Constant Pointer to Constant Data
7.5.3 Attempting to Modify a Constant Pointer to Non-Constant Data
7.5.4 Attempting to Modify a Constant Pointer to Constant Data
Bubble Sort Using Pass-by-Reference
sizeof Operator
Pointer Expressions and Pointer Arithmetic
Relationship between Pointers and Arrays
Arrays of Pointers
Case Study: Card Shuffling and Dealing Simulation
Pointers to Functions
Secure C Programming
8
C Characters and Strings
8.1
8.2
8.3
Introduction
Fundamentals of Strings and Characters
Character-Handling Library
8.3.1 Functions isdigit, isalpha, isalnum and isxdigit
8.3.2 Functions islower, isupper, tolower and toupper
8.3.3 Functions isspace, iscntrl, ispunct, isprint and isgraph
String-Conversion Functions
8.4.1 Function strtod
8.4.2 Function strtol
8.4.3 Function strtoul
Standard Input/Output Library Functions
8.5.1 Functions fgets and putchar
8.5.2 Function getchar
8.5.3 Function sprintf
8.5.4 Function sscanf
String-Manipulation Functions of the String-Handling Library
8.6.1 Functions strcpy and strncpy
8.6.2 Functions strcat and strncat
Comparison Functions of the String-Handling Library
Search Functions of the String-Handling Library
8.8.1 Function strchr
8.8.2 Function strcspn
8.8.3 Function strpbrk
8.8.4 Function strrchr
8.8.5 Function strspn
8.8.6 Function strstr
8.8.7 Function strtok
Memory Functions of the String-Handling Library
8.9.1 Function memcpy
8.9.2 Function memmove
8.9.3 Function memcmp
8.4
8.5
8.6
8.7
8.8
8.9
288
290
291
291
294
297
299
303
304
309
314
334
335
335
337
338
340
341
342
343
344
345
346
346
348
349
349
350
351
352
353
354
355
356
357
357
358
358
359
360
361
362
363
Contents
8.11
8.9.4 Function memchr
8.9.5 Function memset
Other Functions of the String-Handling Library
8.10.1 Function strerror
8.10.2 Function strlen
Secure C Programming
9
C Formatted Input/Output
9.1
9.2
9.3
9.4
9.5
9.6
9.7
9.8
9.9
9.10
9.11
9.12
Introduction
Streams
Formatting Output with printf
Printing Integers
Printing Floating-Point Numbers
Printing Strings and Characters
Other Conversion Specifiers
Printing with Field Widths and Precision
Using Flags in the printf Format Control String
Printing Literals and Escape Sequences
Reading Formatted Input with scanf
Secure C Programming
10
C Structures, Unions, Bit Manipulation and
Enumerations
8.10
10.1
10.2
10.3
10.4
10.5
10.6
10.7
10.8
10.9
Introduction
Structure Definitions
10.2.1 Self-Referential Structures
10.2.2 Defining Variables of Structure Types
10.2.3 Structure Tag Names
10.2.4 Operations That Can Be Performed on Structures
Initializing Structures
Accessing Structure Members
Using Structures with Functions
typedef
Example: High-Performance Card Shuffling and Dealing Simulation
Unions
10.8.1 Union Declarations
10.8.2 Operations That Can Be Performed on Unions
10.8.3 Initializing Unions in Declarations
10.8.4 Demonstrating Unions
Bitwise Operators
10.9.1 Displaying an Unsigned Integer in Bits
10.9.2 Making Function displayBits More Scalable and Portable
10.9.3 Using the Bitwise AND, Inclusive OR, Exclusive OR and
Complement Operators
xi
363
364
365
365
365
366
379
380
380
380
381
382
384
385
386
388
391
391
398
405
406
406
407
407
408
408
409
409
411
411
412
415
415
415
416
416
417
418
420
420
xii
Contents
10.9.4 Using the Bitwise Left- and Right-Shift Operators
10.9.5 Bitwise Assignment Operators
10.10 Bit Fields
10.11 Enumeration Constants
10.12 Secure C Programming
11
C File Processing
11.1
11.2
11.3
11.4
11.5
11.6
11.7
11.8
11.9
11.10
Introduction
Files and Streams
Creating a Sequential-Access File
Reading Data from a Sequential-Access File
Random-Access Files
Creating a Random-Access File
Writing Data Randomly to a Random-Access File
Reading Data from a Random-Access File
Case Study: Transaction-Processing Program
Secure C Programming
12
C Data Structures
12.1
12.2
12.3
12.4
12.8
Introduction
Self-Referential Structures
Dynamic Memory Allocation
Linked Lists
12.4.1 Function insert
12.4.2 Function delete
12.4.3 Function printList
Stacks
12.5.1 Function push
12.5.2 Function pop
12.5.3 Applications of Stacks
Queues
12.6.1 Function enqueue
12.6.2 Function dequeue
Trees
12.7.1 Function insertNode
12.7.2 Traversals: Functions inOrder, preOrder and postOrder
12.7.3 Duplicate Elimination
12.7.4 Binary Tree Search
12.7.5 Other Binary Tree Operations
Secure C Programming
13
C Preprocessor
13.1
13.2
Introduction
#include Preprocessor Directive
12.5
12.6
12.7
423
425
426
429
431
441
442
442
443
448
452
453
455
458
459
465
476
477
478
478
479
485
487
488
488
492
492
493
494
498
499
500
504
504
505
505
505
506
517
518
518
Contents
13.3
13.4
13.5
13.6
13.7
13.8
13.9
13.10
13.11
Preprocessor Directive: Symbolic Constants
Preprocessor Directive: Macros
Conditional Compilation
#error and #pragma Preprocessor Directives
# and ## Operators
Line Numbers
Predefined Symbolic Constants
Assertions
Secure C Programming
14
Other C Topics
14.1
14.2
14.3
14.4
14.5
14.6
14.7
14.8
14.9
14.10
Introduction
Redirecting I/O
Variable-Length Argument Lists
Using Command-Line Arguments
Notes on Compiling Multiple-Source-File Programs
Program Termination with exit and atexit
Suffixes for Integer and Floating-Point Literals
Signal Handling
Dynamic Memory Allocation: Functions calloc and realloc
Unconditional Branching with goto
15
C++ as a Better C; Introducing Object
Technology
#define
#define
15.1
15.2
15.3
15.4
15.5
15.6
15.7
15.8
15.9
15.10
15.11
15.12
15.13
15.14
15.15
Introduction
C++
A Simple Program: Adding Two Integers
C++ Standard Library
Header Files
Inline Functions
References and Reference Parameters
Empty Parameter Lists
Default Arguments
Unary Scope Resolution Operator
Function Overloading
Function Templates
Introduction to C++ Standard Library Class Template vector
Introduction to Object Technology and the UML
Wrap-Up
16
Introduction to Classes, Objects and Strings
16.1
16.2
Introduction
Defining a Class with a Member Function
xiii
519
519
521
522
523
523
523
524
524
529
530
530
531
533
534
536
537
538
540
541
547
548
548
549
551
552
554
556
561
561
563
564
567
570
576
579
586
587
587
xiv
Contents
16.3
16.4
16.5
16.6
16.7
16.8
16.9
Defining a Member Function with a Parameter
Data Members, set Functions and get Functions
Initializing Objects with Constructors
Placing a Class in a Separate File for Reusability
Separating Interface from Implementation
Validating Data with set Functions
Wrap-Up
590
593
599
603
606
612
617
17
Classes: A Deeper Look, Part 1
623
17.1
17.2
17.3
17.4
17.5
17.6
17.7
17.8
17.9
Introduction
Time Class Case Study
Class Scope and Accessing Class Members
Separating Interface from Implementation
Access Functions and Utility Functions
Time Class Case Study: Constructors with Default Arguments
Destructors
When Constructors and Destructors Are Called
Time Class Case Study: A Subtle Trap—Returning a Reference to
a private Data Member
17.10 Default Memberwise Assignment
17.11 Wrap-Up
624
625
632
633
634
637
642
643
646
649
652
18
Classes: A Deeper Look, Part 2
18.1
18.2
18.3
18.4
18.5
18.6
18.7
18.8
Introduction
const (Constant) Objects and const Member Functions
Composition: Objects as Members of Classes
friend Functions and friend Classes
Using the this Pointer
static Class Members
Proxy Classes
Wrap-Up
659
659
667
673
675
680
685
689
19
Operator Overloading; Class string
695
19.1
19.2
19.3
19.4
19.5
19.6
19.7
19.8
19.9
Introduction
Using the Overloaded Operators of Standard Library Class string
Fundamentals of Operator Overloading
Overloading Binary Operators
Overloading the Binary Stream Insertion and Stream Extraction Operators
Overloading Unary Operators
Overloading the Unary Prefix and Postfix ++ and -- Operators
Case Study: A Date Class
Dynamic Memory Management
658
696
697
700
701
702
706
707
708
713
Contents
19.10 Case Study: Array Class
19.10.1 Using the Array Class
19.10.2 Array Class Definition
19.11 Operators as Member Functions vs. Non-Member Functions
19.12 Converting between Types
19.13 explicit Constructors
19.14 Building a String Class
19.15 Wrap-Up
xv
715
716
719
727
727
729
731
732
20
Object-Oriented Programming: Inheritance
20.1
20.2
20.3
20.4
20.5
20.6
20.7
20.8
Introduction
Base Classes and Derived Classes
protected Members
Relationship between Base Classes and Derived Classes
20.4.1 Creating and Using a CommissionEmployee Class
20.4.2 Creating a BasePlusCommissionEmployee Class Without
Using Inheritance
20.4.3 Creating a CommissionEmployee–BasePlusCommissionEmployee
Inheritance Hierarchy
20.4.4 CommissionEmployee–BasePlusCommissionEmployee Inheritance
Hierarchy Using protected Data
20.4.5 CommissionEmployee–BasePlusCommissionEmployee Inheritance
Hierarchy Using private Data
Constructors and Destructors in Derived Classes
public, protected and private Inheritance
Software Engineering with Inheritance
Wrap-Up
21
Object-Oriented Programming: Polymorphism
778
21.1
21.2
21.3
Introduction
Introduction to Polymorphism: Polymorphic Video Game
Relationships Among Objects in an Inheritance Hierarchy
21.3.1 Invoking Base-Class Functions from Derived-Class Objects
21.3.2 Aiming Derived-Class Pointers at Base-Class Objects
21.3.3 Derived-Class Member-Function Calls via Base-Class Pointers
21.3.4 Virtual Functions
Type Fields and switch Statements
Abstract Classes and Pure virtual Functions
Case Study: Payroll System Using Polymorphism
21.6.1 Creating Abstract Base Class Employee
21.6.2 Creating Concrete Derived Class SalariedEmployee
21.6.3 Creating Concrete Derived Class CommissionEmployee
21.6.4 Creating Indirect Concrete Derived Class
779
780
780
781
784
785
787
793
793
795
796
800
802
21.4
21.5
21.6
BasePlusCommissionEmployee
21.6.5 Demonstrating Polymorphic Processing
743
744
744
747
747
748
752
758
763
766
771
771
772
773
804
806
xvi
Contents
21.7
(Optional) Polymorphism, Virtual Functions and Dynamic Binding
“Under the Hood”
21.8 Case Study: Payroll System Using Polymorphism and Runtime Type
Information with Downcasting, dynamic_cast, typeid and type_info
21.9 Virtual Destructors
21.10 Wrap-Up
22
Templates
22.1
22.2
22.3
22.4
22.5
22.6
Introduction
Function Templates
Overloading Function Templates
Class Templates
Nontype Parameters and Default Types for Class Templates
Wrap-Up
23
Stream Input/Output
23.1
23.2
Introduction
Streams
23.2.1 Classic Streams vs. Standard Streams
23.2.2 iostream Library Headers
23.2.3 Stream Input/Output Classes and Objects
Stream Output
23.3.1 Output of char * Variables
23.3.2 Character Output Using Member Function put
Stream Input
23.4.1 get and getline Member Functions
23.4.2 istream Member Functions peek, putback and ignore
23.4.3 Type-Safe I/O
Unformatted I/O Using read, write and gcount
Introduction to Stream Manipulators
23.6.1 Integral Stream Base: dec, oct, hex and setbase
23.6.2 Floating-Point Precision (precision, setprecision)
23.6.3 Field Width (width, setw)
23.6.4 User-Defined Output Stream Manipulators
Stream Format States and Stream Manipulators
23.7.1 Trailing Zeros and Decimal Points (showpoint)
23.7.2 Justification (left, right and internal)
23.7.3 Padding (fill, setfill)
23.7.4 Integral Stream Base (dec, oct, hex, showbase)
23.7.5 Floating-Point Numbers; Scientific and Fixed Notation
(scientific, fixed)
23.7.6 Uppercase/Lowercase Control (uppercase)
23.7.7 Specifying Boolean Format (boolalpha)
23.7.8 Setting and Resetting the Format State via Member
Function flags
23.3
23.4
23.5
23.6
23.7
810
813
817
817
823
824
824
827
828
834
835
839
840
841
841
842
842
845
845
845
846
846
849
849
849
850
851
851
853
854
856
856
857
859
860
861
862
862
863
Contents
23.8 Stream Error States
23.9 Tying an Output Stream to an Input Stream
23.10 Wrap-Up
xvii
864
866
867
24
Exception Handling: A Deeper Look
24.1
24.2
24.3
24.4
24.5
24.6
24.7
24.8
24.9
24.10
24.11
24.12
Introduction
Example: Handling an Attempt to Divide by Zero
When to Use Exception Handling
Rethrowing an Exception
Processing Unexpected Exceptions
Stack Unwinding
Constructors, Destructors and Exception Handling
Exceptions and Inheritance
Processing new Failures
Class unique_ptr and Dynamic Memory Allocation
Standard Library Exception Hierarchy
Wrap-Up
A
Operator Precedence Charts
902
B
ASCII Character Set
906
C
Number Systems
907
C.1
C.2
C.3
C.4
C.5
C.6
Introduction
Abbreviating Binary Numbers as Octal and Hexadecimal Numbers
Converting Octal and Hexadecimal Numbers to Binary Numbers
Converting from Binary, Octal or Hexadecimal to Decimal
Converting from Decimal to Binary, Octal or Hexadecimal
Negative Binary Numbers: Two’s Complement Notation
D
Game Programming: Solving Sudoku
D.1
D.2
D.3
D.4
D.5
D.6
Introduction
Deitel Sudoku Resource Center
Solution Strategies
Programming Sudoku Puzzle Solvers
Generating New Sudoku Puzzles
Conclusion
876
877
877
883
884
885
886
888
888
889
892
894
896
908
911
912
912
913
915
920
920
921
921
925
926
928
Appendices on the Web
929
Index
930
xviii
Contents
Appendices E through H are PDF documents posted online at the book’s Companion
Website (located at www.pearsonhighered.com/deitel).
E
Sorting: A Deeper Look
F
Introduction to the New C Standard
G
Using the Visual Studio Debugger
H
Using the GNU Debugger
Preface
Welcome to the C programming language—and to C++, too! This book presents leadingedge computing technologies for college students, instructors and software development
professionals.
At the heart of the book is the Deitel signature “live-code approach.” We present concepts in the context of complete working programs, rather than in code snippets. Each
code example is followed by one or more sample executions. Read the online Before You
Begin section (www.deitel.com/books/chtp7/chtp7_BYB.pdf) to learn how to set up
your computer to run the hundreds of code examples. All the source code is available at
www.deitel.com/books/chtp7/ and www.pearsonhighered.com/deitel. Use the source
code we provide to run every program as you study it.
We believe that this book and its support materials will give you an informative, challenging and entertaining introduction to C. As you read the book, if you have questions,
send an e-mail to deitel@deitel.com—we’ll respond promptly. For book updates, visit
www.deitel.com/books/chtp7/, join our communities on Facebook (www.deitel.com/
deitelfan), Twitter (@deitel) and Google+ (gplus.to/deitel), and subscribe to the
Deitel ® Buzz Online newsletter (www.deitel.com/newsletter/subscribe.html).
New and Updated Features
Here are some key features of C How to Program, 7/e:
• Coverage of the New C standard. The previous edition of the book conformed to
“standard C” and included a detailed appendix on the C99 standard. The New
C Standard was approved just before C How to Program, 7/e went to publication.
The new standard incorporates both C99 and the more recent C1X—now referred to as C11 or simply “the C standard” since its approval in 2011. Support
for the new standard varies by compiler. The vast majority of our readership uses
either the GNU gcc compiler—which supports several of the key features in the
new standard—or the Microsoft Visual C++ compiler. Microsoft supports only
a limited subset of the features that were added to C in C99 and C11—primarily
the features that are also required by the C++ standard. To accommodate all of
our readers, we placed the discussion of the new standard in optional, easy-to-useor-omit sections and in Appendix F, Introduction to the New C Standard. We’ve
also replaced various deprecated capabilities with newer preferred versions as a result of the new C standard.
• New Chapter 1. The new Chapter 1 engages students with intriguing facts and figures to get them excited about studying computers and computer programming.
The chapter includes a table of some of the research made possible by computers
and the Internet, current technology trends and hardware discussion, the data hierarchy, a new section on social networking, a table of business and technology pub-
xx
Preface
•
•
•
•
•
•
•
•
•
•
•
•
lications and websites that will help you stay up to date with the latest technology
news and trends, and updated exercises. We’ve included test-drives that show how
to run a command-line C program on Microsoft Windows, Linux and Mac OS X.
Secure C Programming Sections. We’ve added notes about secure C programming
to many of the C programming chapters. We’ve also posted a Secure C Programming Resource Center at www.deitel.com/SecureC/. For more details, see the section “A Note About Secure C Programming” in this Preface.
Focus on Performance Issues. C (and C++) are favored by designers of performance-intensive applications such as operating systems, real-time systems, embedded systems and communications systems, so we focus intensively on
performance issues.
“Making a Difference” Exercise Sets. We encourage you to use computers and the
Internet to research and solve problems that really matter. These exercises are
meant to increase awareness of important issues the world is facing. We hope
you’ll approach them with your own values, politics and beliefs.
All Code Tested on Windows and Linux. We’ve tested every example and exercise
program using Visual C++ and GNU gcc in Windows and Linux, respectively.
Updated Coverage of C++ and Object-Oriented Programming. We updated
Chapters 15–24 on object-oriented programming in C++ with material from our
textbook C++ How to Program, 8/e.
Sorting: A Deeper Look. Sorting places data in order, based on one or more sort
keys. We begin our presentation of sorting with a simple algorithm in Chapter 6—
in Appendix E, we present a deeper look. We consider several algorithms and compare them with regard to their memory consumption and processor demands. For
this purpose, we introduce Big O notation, which indicates how hard an algorithm
may have to work to solve a problem. Through examples and exercises, Appendix E
discusses the selection sort, insertion sort, recursive merge sort, recursive selection
sort, bucket sort and recursive Quicksort. Sorting is an interesting problem because
different sorting techniques achieve the same final result but they can vary hugely
in their consumption of memory, CPU time and other system resources.
Titled Programming Exercises. All the programming exercises are titled to help
instructors tune assignments for their classes.
Debugger Appendices. We’ve updated the Visual C++® and GNU gdb debugging
appendices.
Order of Evaluation. We added cautions about order of evaluation issues.
Additional Exercises. We added more function pointer exercises. We also added
a Fibonacci exercise project that improves the Fibonacci recursion example (tail
recursion).
C++-Style // Comments. We use the newer, more concise C++-style // comments in preference to C’s older style /*...*/ comments.
C Standard Library. Section 1.7 references P.J. Plauger’s Dinkumware website
(www.dinkumware.com/manuals/default.aspx) where students can find thorough searchable documentation for the C Standard Library functions.
A Note About Secure C Programming
xxi
A Note About Secure C Programming
Throughout this book, we focus on C programming fundamentals. When we write each
How to Program book, we search the corresponding language’s standards document for the
features that we feel novices need to learn in a first programming course, and features that
existing programmers need to know to begin working in that language. We must also cover
programming fundamentals and computer-science fundamentals for novice programmers—our core audience.
Industrial-strength coding techniques in any programming language are beyond the
scope of an introductory textbook. For that reason, our Secure C Programming sections
present some key issues and techniques, and provide links and references so you can continue learning.
Experience has shown that it’s difficult to build industrial-strength systems that stand
up to attacks from viruses, worms, etc. Today, via the Internet, such attacks can be instantaneous and global in scope. Software vulnerabilities often come from simple programming issues. Building security into software from the start of the development cycle can
greatly reduce costs and vulnerabilities.
The CERT® Coordination Center (www.cert.org) was created to analyze and
respond promptly to attacks. CERT—the Computer Emergency Response Team—publishes and promotes secure coding standards to help C programmers and others implement
industrial-strength systems that avoid the programming practices that open systems to
attacks. The CERT standards evolve as new security issues arise.
We’ve upgraded our code (as appropriate for an introductory book) to conform to various CERT recommendations. If you’ll be building C systems in industry, consider reading
The CERT C Secure Coding Standard (Robert Seacord, Addison-Wesley Professional, 2009)
and Secure Coding in C and C++ (Robert Seacord, Addison-Wesley Professional, 2006). The
CERT guidelines are available free online at www.securecoding.cert.org. Mr. Seacord, a
technical reviewer for the C portion of this book, provided specific recommendations on
each of our new Secure C Programming sections. Mr. Seacord is the Secure Coding Manager
at CERT at Carnegie Mellon University’s Software Engineering Institute (SEI) and an
adjunct professor in the Carnegie Mellon University School of Computer Science.
The Secure C Programming sections at the ends of Chapters 2–13 discuss many important topics, including testing for arithmetic overflows, using unsigned integer types, new
more secure functions in the C standard’s Annex K, the importance of checking the status
information returned by standard-library functions, range checking, secure random-number
generation, array bounds checking, techniques for preventing buffer overflows, input validation, avoiding undefined behaviors, choosing functions that return status information vs.
using similar functions that do not, ensuring that pointers are always NULL or contain valid
addresses, using C functions vs. using preprocessor macros, and more.
Web-Based Materials
This book is supported by substantial online materials. The book’s Companion Website
(www.pearsonhighered.com/deitel) contains source code for all the code examples and
the following appendices in searchable PDF format:
•
Appendix E, Sorting: A Deeper Look
•
Appendix F, Introduction to the New C Standard
xxii
Preface
•
Appendix G, Using the Visual Studio Debugger
•
Appendix H, Using the GNU Debugger
Dependency Charts
Figures 1 and 2 show the dependencies among the chapters to help instructors plan their
syllabi. C How to Program, 7/e is appropriate for CS1 and CS2 courses, and intermediatelevel C and C++ programming courses. The C++ part of the book assumes that you’ve
studied the C part.
C Chapter
Dependency
Chart
Introduction
1 Introduction to Computers,
the Internet and the Web
[Note: Arrows pointing into a
chapter indicate that chapter’s
dependencies.]
Intro to Programming
2 Intro to C Programming
Control Statements,
Functions and Arrays
3 Structured Program
Development in C
4 C Program Control
5 C Functions
6 C Arrays
Streams and Files
Pointers and Strings
9 C Formatted Input/Output
7 C Pointers
11 C File Processing
8 C Characters and Strings
Aggregate Types
10 C Structures, Unions, Bit
Manipulations and Enumerations
Data Structures
Other Topics and the New C Standard
5.14–5.16 Recursion
12 C Data Structures
13 C Preprocessor
14 Other C Topics
Fig. 1 | C chapter dependency chart.
F Intro to the New C Standard
E Sorting: A Deeper Look
Teaching Approach
C++ Chapter
Dependency
Chart
xxiii
Object-Based
Programming
15 C++ as a Better C;
Intro to Object Technology
[Note: Arrows pointing into a
chapter indicate that chapter’s
dependencies.]
16 Intro to Classes and Objects
17 Classes: A Deeper
Look, Part 1
18 Classes: A Deeper
Look, Part 2
19 Operator Overloading
Object-Oriented
Programming
20 OOP: Inheritance
21 OOP:
Polymorphism
22 Templates
23 Stream
Input/Output
24 Exception
Handling
Fig. 2 | C++ chapter dependency chart.
Teaching Approach
C How to Program, 7/e, contains a rich collection of examples. We focus on good software
engineering and stressing program clarity.
Syntax Shading. For readability, we syntax shade the code, similar to the way most IDEs
and code editors syntax color code. Our syntax-shading conventions are:
comments appear like this
keywords appear like this
constants and literal values appear like this
all other code appears in black
Code Highlighting. We place gray rectangles around the key code.
Using Fonts for Emphasis. We place the key terms and the index’s page reference for each
defining occurrence in bold blue text for easy reference. We emphasize on-screen components in the bold Helvetica font (e.g., the File menu) and C program text in the Lucida
font (for example, int x = 5;).
Objectives. The opening quotes are followed by a list of chapter objectives.
Illustrations/Figures. Abundant charts, tables, line drawings, UML diagrams, programs
and program output are included.
Programming Tips. We include programming tips to help you focus on important aspects
of program development. These tips and practices represent the best we’ve gleaned from a
combined seven decades of programming and teaching experience.
xxiv
Preface
Good Programming Practices
The Good Programming Practices call attention to techniques that will help you produce programs that are clearer, more understandable and more maintainable.
Common Programming Errors
Pointing out these Common Programming Errors reduces the likelihood that you’ll
make them.
Error-Prevention Tips
These tips contain suggestions for exposing and removing bugs from your programs; many
describe aspects of C that prevent bugs from getting into programs in the first place.
Performance Tips
These tips highlight opportunities for making your programs run faster or minimizing the
amount of memory that they occupy.
Portability Tips
The Portability Tips help you write code that will run on a variety of platforms.
Software Engineering Observations
The Software Engineering Observations highlight architectural and design issues that
affect the construction of software systems, especially large-scale systems.
Summary Bullets. We present a section-by-section, bullet-list summary of the chapter.
Terminology. We include an alphabetized list of the important terms defined in each chapter with the page number of each term’s defining occurrence for easy reference.
Self-Review Exercises and Answers. Extensive self-review exercises and answers are included for self-study.
Exercises. Each chapter concludes with a substantial set of exercises including:
• simple recall of important terminology and concepts
• identifying the errors in code samples
• writing individual program statements
• writing small portions of C functions and C++ member functions and classes
• writing complete programs
• implementing major projects
Index. We’ve included an extensive index, which is especially useful when you use the
book as a reference. Defining occurrences of key terms are highlighted with a bold blue
page number.
Software Used in C How to Program, 7/e
We wrote C How to Program, 7/e using Microsoft’s free Visual C++ Express Edition (which
can compile both C and C++ programs and can be downloaded from www.microsoft.com/
C++ IDE Resource Kit
express/downloads/)
xxv
and the free GNU C and C++ compilers (gcc.gnu.org/install/
binaries.html), which are already installed on most Linux systems and can be installed on
Mac OS X and Windows systems. Apple includes GNU C and C++ in their Xcode
development tools, which Mac OS X users can download from developer.apple.com/
technologies/tools/xcode.html.
For other free C and C++ compilers, visit:
www.thefreecountry.com/compilers/cpp.shtml
www.compilers.net/Dir/Compilers/CCpp.htm
www.freebyte.com/programming/cpp/#cppcompilers
en.wikipedia.org/wiki/List_of_compilers#C.2B.2B_compilers
C++ IDE Resource Kit
Your instructor may have ordered through your college bookstore a Value Pack edition of
C How to Program, 7/e that comes bundled with the C++ IDE Resource Kit—most C++
compilers also support C. This kit contains CD or DVD versions of:
•
Microsoft® Visual Studio 2010 Express Edition (www.microsoft.com/express/)
•
Dev C++ (www.bloodshed.net/download.html)
•
NetBeans (netbeans.org/downloads/index.html)
•
Eclipse (eclipse.org/downloads/)
•
CodeLite (codelite.org/LiteEditor/Download)
You can also download these software packages from the websites specified above. The C++
IDE Resource Kit also includes access to a Companion Website containing step-by-step
written instructions and VideoNotes to help you get started with each development environment. If your book did not come with the C++ IDE Resource Kit, you can purchase access
to the Resource Kit’s Companion Website from www.pearsonhighered.com/cppidekit/.
CourseSmart Web Books
Today’s students and instructors have increasing demands on their time and money. Pearson has responded to that need by offering digital texts and course materials online
through CourseSmart. CourseSmart allows faculty to review course materials online, saving time and costs. It offers students a high-quality digital version of the text for less than
the cost of a print copy. Students receive the same content offered in the print textbook
enhanced by search, note-taking and printing tools. For more information, visit
www.coursesmart.com.
Instructor Resources
The following supplements are available to qualified instructors only through Pearson
Education’s Instructor Resource Center (www.pearsonhighered.com/irc):
•
PowerPoint® slides containing all the code and figures in the text, plus bulleted
items that summarize key points.
•
Test Item File of multiple-choice questions (approximately two per book section)
xxvi
•
Preface
Solutions Manual with solutions to most of the end-of-chapter exercises. Please
check the Instructor Resource Center to determine which exercises have solutions.
Please do not write to us requesting access to the Pearson Instructor’s Resource Center.
Access is restricted to college instructors teaching from the book. Instructors may obtain
access only through their Pearson representatives. If you’re not a registered faculty member, contact your Pearson representative or visit www.pearsonhighered.com/educator/
replocator/.
Solutions are not provided for “project” exercises. Check out our Programming Projects Resource Center for lots of additional exercise and project possibilities
(www.deitel.com/ProgrammingProjects/).
Acknowledgments
We’d like to thank Abbey Deitel and Barbara Deitel for long hours devoted to this project.
We’re fortunate to have worked with the dedicated team of publishing professionals at
Pearson. We appreciate the guidance, savvy and energy of Michael Hirsch, Editor-inChief of Computer Science. Carole Snyder and Bob Engelhardt did a marvelous job managing the review and production processes, respectively.
C How to Program, 7/e Reviewers
We wish to acknowledge the efforts of our reviewers. Under tight deadlines, they scrutinized the text and the programs and provided countless suggestions for improving the presentation: Dr. John F. Doyle (Indiana University Southeast), Hemanth H.M. (Software
Engineer at SonicWALL), Vytautus Leonavicius (Microsoft), Robert Seacord (Secure
Coding Manager at SEI/CERT, author of The CERT C Secure Coding Standard and technical expert for the international standardization working group for the programming language C) and José Antonio González Seco (Parliament of Andalusia).
Other Recent Editions Reviewers
William Albrecht (University of South Florida), Ian Barland (Radford University), Ed
James Beckham (Altera), John Benito (Blue Pilot Consulting, Inc. and Convener of ISO
WG14—the Working Group responsible for the C Programming Language Standard),
Alireza Fazelpour (Palm Beach Community College), Mahesh Hariharan (Microsoft),
Kevin Mark Jones (Hewlett Packard), Lawrence Jones, (UGS Corp.), Don Kostuch (Independent Consultant), Xiaolong Li (Indiana State University), William Mike Miller
(Edison Design Group, Inc.), Tom Rethard (The University of Texas at Arlington), Benjamin Seyfarth (University of Southern Mississippi), Gary Sibbitts (St. Louis Community
College at Meramec), William Smith (Tulsa Community College) and Douglas Walls (Senior Staff Engineer, C compiler, Sun Microsystems).
Well, there you have it! C is a powerful programming language that will help you
write high-performance programs quickly and effectively. C scales nicely into the realm of
enterprise systems development to help organizations build their business-critical and mission-critical information systems. As you read the book, we would sincerely appreciate
your comments, criticisms, corrections and suggestions for improving the text. Please
address all correspondence to:
deitel@deitel.com
About the Authors
xxvii
We’ll respond promptly, and post corrections and clarifications on:
www.deitel.com/books/chtp7/
We hope you enjoy working with C How to Program, Seventh Edition as much as we enjoyed writing it!
Paul Deitel
Harvey Deitel
January 2012
About the Authors
Paul Deitel, CEO and Chief Technical Officer of Deitel & Associates, Inc., is a graduate
of MIT, where he studied Information Technology. Through Deitel & Associates, Inc.,
he has delivered hundreds of programming courses to industry clients, including Cisco,
IBM, Siemens, Sun Microsystems, Dell, Lucent Technologies, Fidelity, NASA at the Kennedy Space Center, the National Severe Storm Laboratory, White Sands Missile Range,
Rogue Wave Software, Boeing, SunGard Higher Education, Stratus, Cambridge Technology Partners, One Wave, Hyperion Software, Adra Systems, Entergy, CableData Systems,
Nortel Networks, Puma, iRobot, Invensys and many more. He and his co-author, Dr.
Harvey M. Deitel, are the world’s best-selling programming-language textbook/professional book/video authors.
Dr. Harvey Deitel, Chairman and Chief Strategy Officer of Deitel & Associates, Inc.,
has 50 years of experience in the computer field. Dr. Deitel earned B.S. and M.S. degrees
from MIT and a Ph.D. from Boston University. He has extensive college teaching experience, including earning tenure and serving as the Chairman of the Computer Science
Department at Boston College before founding Deitel & Associates, Inc., in 1991 with
his son, Paul Deitel. The Deitels’ publications have earned international recognition, with
translations published in Chinese, Korean, Japanese, German, Russian, Spanish, French,
Polish, Italian, Portuguese, Greek, Urdu and Turkish. Dr. Deitel has delivered hundreds
of professional programming seminars to major corporations, academic institutions, government organizations and the military.
Corporate Training from Deitel & Associates, Inc.
Deitel & Associates, Inc., founded by Paul Deitel and Harvey Deitel, is an internationally
recognized authoring, corporate training and software development organization specializing in computer programming languages, object technology, Android and iPhone app
development and Internet and web software technology. The company offers instructorled training courses delivered at client sites worldwide on major programming languages
and platforms, including C, C++, Visual C++®, Java™, Visual C#®, Visual Basic®,
XML®, Python®, object technology, Internet and web programming, Android app development, Objective-C and iPhone app development and a growing list of additional programming and software development courses. The company’s clients include many of the
world’s largest companies, government agencies, branches of the military, and academic
institutions.
Through its 36-year publishing partnership with Prentice Hall/Pearson, Deitel &
Associates, Inc., publishes leading-edge programming college textbooks, professional
xxviii
Preface
books and LiveLessons video courses. Deitel & Associates, Inc. and the authors can be
reached at:
deitel@deitel.com
To learn more about Deitel’s Dive Into® Series Corporate Training curriculum, visit:
www.deitel.com/training/
To request a proposal for worldwide on-site, instructor-led training at your company or
organization, e-mail deitel@deitel.com.
Individuals wishing to purchase Deitel books and LiveLessons video training can do so
through www.deitel.com. Bulk orders by corporations, the government, the military and
academic institutions should be placed directly with Pearson. For more information, visit
www.pearsoned.com/professional/index.htm.
1
Introduction to Computers,
the Internet and the Web
The chief merit of language is
clearness.
—Galen
Our life is frittered away by
detail. … Simplify, simplify.
—Henry David Thoreau
He had a wonderful talent for
packing thought close, and
rendering it portable.
—Thomas B. Macaulay
Man is still the most
extraordinary computer of all.
—John F. Kennedy
Objectives
In this chapter, you’ll learn:
■
Basic computer concepts.
■
The different types of
programming languages.
■
The history of the C
programming language.
■
The purpose of the C
Standard Library.
■
The elements of a typical C
program development
environment.
■
To test-drive a C application
in Windows, Linux and Mac
OS X.
■
Some basics of the Internet
and the World Wide Web.
2
Chapter 1 Introduction to Computers, the Internet and the Web
1.1 Introduction
1.2 Computers and the Internet in
Industry and Research
1.3 Hardware and Software
1.3.1 Moore’s Law
1.3.2 Computer Organization
1.4
1.5
1.6
1.7
1.8
1.9
1.10
Data Hierarchy
Programming Languages
The C Programming Language
C Standard Library
C++ and Other C-Based Languages
Object Technology
Typical C Program Development
Environment
1.10.1 Phase 1: Creating a Program
1.10.2 Phases 2 and 3: Preprocessing and
Compiling a C Program
1.10.3 Phase 4: Linking
1.10.4 Phase 5: Loading
1.10.5 Phase 6: Execution
1.10.6 Problems That May Occur at
Execution Time
1.10.7 Standard Input, Standard Output and
Standard Error Streams
1.11 Test-Driving a C Application in
Windows, Linux and Mac OS X
1.11.1 Running a C Application from the
Windows Command Prompt
1.11.2 Running a C Application Using GNU
C with Linux
1.11.3 Running a C Application Using GNU
C with Mac OS X
1.12 Operating Systems
1.12.1 Windows—A Proprietary Operating
System
1.12.2 Linux—An Open-Source Operating
System
1.12.3 Apple’s Mac OS X; Apple’s iOS for
iPhone®, iPad® and iPod Touch®
Devices
1.12.4 Google’s Android
1.13 The Internet and World Wide Web
1.14 Some Key Software Development
Terminology
1.15 Keeping Up-to-Date with
Information Technologies
1.16 Web Resources
Terminology | Self-Review Exercises | Answers to Self-Review Exercises | Exercises
Making a Difference
1.1 Introduction
Welcome to C and C++! C is a concise yet powerful computer programming language
that’s appropriate for technically oriented people with little or no programming experience
and for experienced programmers to use in building substantial software systems. C How
to Program, Seventh Edition, is an effective learning tool for each of these audiences.
The core of the book emphasizes effective software engineering through the proven
methodologies of structured programming in C and object-oriented programming in C++. The
book presents hundreds of complete working programs and shows the outputs produced when
those programs are run on a computer. We call this the “live-code approach.” All of these
example programs may be downloaded from our website www.deitel.com/books/chtp7/.
Most people are familiar with the exciting tasks that computers perform. Using this
textbook, you’ll learn how to command computers to perform those tasks. It’s software
(i.e., the instructions you write to command computers to perform actions and make decisions) that controls computers (often referred to as hardware).
1.2 Computers and the Internet in Industry and Research
These are exciting times in the computer field. Many of the most influential and successful
businesses of the last two decades are technology companies, including Apple, IBM, Hew-
1.2 Computers and the Internet in Industry and Research
3
lett Packard, Dell, Intel, Motorola, Cisco, Microsoft, Google, Amazon, Facebook, Twitter, Groupon, Foursquare, Yahoo!, eBay and many more. These companies are major
employers of people who study computer science, computer engineering, information systems or related disciplines. At the time of this writing, Apple was the most valuable company in the world. Figure 1.1 provides a few examples of the ways in which computers are
used in research and industry.
Name
Description
Electronic health
records
These might include a patient's medical history, prescriptions, immunizations, lab results, allergies, insurance information and more. Making
this information available to health care providers across a secure network improves patient care, reduces the probability of error and
increases overall efficiency of the health care system.
Human Genome
Project
The Human Genome Project was founded to identify and analyze the
20,000+ genes in human DNA. The project used computer programs
to analyze complex genetic data, determine the sequences of the billions of chemical base pairs that make up human DNA and store the
information in databases which have been made available over the
Internet to researchers in many fields.
AMBER™ Alert
The AMBER (America’s Missing: Broadcast Emergency Response)
Alert System is used to find abducted children. Law enforcement
notifies TV and radio broadcasters and state transportation officials,
who then broadcast alerts on TV, radio, computerized highway signs,
the Internet and wireless devices. AMBER Alert recently partnered
with Facebook, whose users can “Like” AMBER Alert pages by location
to receive alerts in their news feeds.
World
Community Grid
People worldwide can donate their unused computer processing power
by installing a free secure software program that allows the World
Community Grid (www.worldcommunitygrid.org) to harness unused
capacity. This computing power, accessed over the Internet, is used in
place of expensive supercomputers to conduct scientific research
projects that are making a difference—providing clean water to thirdworld countries, fighting cancer, growing more nutritious rice for
regions fighting hunger and more.
Medical imaging
X-ray computed tomography (CT) scans, also called CAT (computerized axial tomography) scans, take X-rays of the body from hundreds of
different angles. Computers are used to adjust the intensity of the Xrays, optimizing the scan for each type of tissue, then to combine all of
the information to create a 3D image. MRI scanners use a technique
called magnetic resonance imaging, also to produce internal images
non-invasively.
One Laptop Per
Child (OLPC)
One Laptop Per Child (one.laptop.org) is providing low-power, inexpensive, Internet-enabled laptops to children in third-world countries—enabling learning and reducing the digital divide.
Fig. 1.1 | A few uses for computers. (Part 1 of 3.)
4
Chapter 1 Introduction to Computers, the Internet and the Web
Name
Description
Cloud
computing
Cloud computing allows you to use software, hardware and information stored in the “cloud”—i.e., accessed on remote computers via the
Internet and available on demand—rather than having it stored on
your personal computer. These services allow you to increase or
decrease resources to meet your needs at any given time, so they can be
more cost effective than purchasing expensive hardware to ensure that
you have enough storage and processing power to meet your needs at
their peak levels. Business applications often are expensive, and require
significant hardware to run them and knowledgeable support staff to
ensure that they’re running properly and securely. Using cloud computing services shifts the burden of managing these applications from the
business to the service provider, saving businesses money.
GPS
Global Positioning System (GPS) devices use a network of satellites to
retrieve location-based information. Multiple satellites send timestamped signals to the GPS device, which calculates the distance to
each satellite based on the time the signal left the satellite and the time
the signal arrived. This information is used to determine the exact location of the device. GPS devices can provide step-by-step directions and
help you locate nearby businesses (restaurants, gas stations, etc.) and
points of interest. GPS is used in numerous location-based Internet services such as check-in apps to help you find your friends (e.g., Foursquare and Facebook), exercise apps such as RunKeeper that track the
time, distance and average speed of your outdoor jog, dating apps that
help you find a match nearby and apps that dynamically update changing traffic conditions.
Robots
Robots can be used for day-to-day tasks (e.g., iRobot’s Roomba vacuuming robot), entertainment (e.g., robotic pets), military combat, deep
sea and space exploration (e.g., NASA’s Mars rover) and more.
RoboEarth (www.roboearth.org) is “a World Wide Web for robots.” It
allows robots to learn from each other by sharing information and thus
improving their abilities to perform tasks, navigate, recognize objects
and more.
E-mail, Instant
Messaging,
Video Chat
and FTP
Internet-based servers support all of your online messaging. E-mail
messages go through a mail server that also stores the messages. Instant
Messaging (IM) and Video Chat apps, such as AIM, Skype, Yahoo!
Messenger and others allow you to communicate with others in real
time by sending your messages and live video through servers. FTP (file
transfer protocol) allows you to exchange files between multiple computers (e.g., a client computer such as your desktop and a file server)
over the Internet.
Internet TV
Internet TV set-top boxes (such as Apple TV, Google TV and TiVo)
allow you to access an enormous amount of content on demand, such
as games, news, movies, television shows and more, and they help
ensure that the content is streamed to your TV smoothly.
Fig. 1.1 | A few uses for computers. (Part 2 of 3.)
1.3 Hardware and Software
Name
Description
Game
programming
Analysts expect global video game revenues to reach $91 billion by
2015 (www.vg247.com/2009/06/23/global-industry-analystspredicts-gaming-market-to-reach-91-billion-by-2015/). The most
sophisticated games can cost as much as $100 million to develop.
Activision’s Call of Duty: Black Ops—one of the best-selling games of all
time—earned $360 million in just one day (www.forbes.com/sites/
5
insertcoin/2011/03/11/call-of-duty-black-ops-now-the-bestselling-video-game-of-all-time/)! Online social gaming, which
enables users worldwide to compete with one another over the Internet,
is growing rapidly. Zynga—creator of popular online games such as
Farmville and Mafia Wars—was founded in 2007 and already has over
200 million monthly users. To accommodate the growth in traffic,
Zynga is adding nearly 1,000 servers each week (techcrunch.com/
2010/09/22/zynga-moves-1-petabyte-of-data-daily-adds-1000servers-a-week/)!
Fig. 1.1 | A few uses for computers. (Part 3 of 3.)
1.3 Hardware and Software
In use today are more than a billion general-purpose computers, and billions more embedded computers are used in cell phones, smartphones, tablet computers, home appliances,
automobiles and more. Computers can perform computations and make logical decisions
phenomenally faster than human beings can. Many of today’s personal computers can perform billions of calculations in one second—more than a human can perform in a lifetime.
Supercomputers are already performing thousands of trillions (quadrillions) of instructions
per second! In 2011, Fujitsu announced that its “K” supercomputer can perform over 10
quadrillion calculations per second (10 petaflops)! To put that in perspective, the K supercomputer can perform in one second more than 1,000,000 calculations for every person on the
planet! And—these “upper limits” are growing quickly!
Computers process data under the control of sequences of instructions called computer programs. These programs guide the computer through ordered actions specified by
people called computer programmers. The programs that run on a computer are referred
to as software. In this book, you’ll learn key programming methodologies that are
enhancing programmer productivity, thereby reducing software development costs—
structured programming (in C) and object-oriented programming in C++.
A computer consists of various devices referred to as hardware (e.g., the keyboard,
screen, mouse, hard disks, memory, DVD drives and processing units). Computing costs
are dropping dramatically, owing to rapid developments in hardware and software technologies. Computers that might have filled large rooms and cost millions of dollars decades
ago are now inscribed on silicon chips smaller than a fingernail, costing perhaps a few dollars each. Ironically, silicon is one of the most abundant materials—it’s an ingredient in
common sand. Silicon-chip technology has made computing so economical that computers have become a commodity.
6
Chapter 1 Introduction to Computers, the Internet and the Web
1.3.1 Moore’s Law
Every year, you probably expect to pay at least a little more for most products and services.
The opposite has been the case in the computer and communications fields, especially with
regard to the costs of hardware supporting these technologies. For many decades, hardware
costs have fallen rapidly. Every year or two, the capacities of computers have approximately
doubled inexpensively. This remarkable trend often is called Moore’s Law, named for the
person who identified it, Gordon Moore, co-founder of Intel—the leading manufacturer of
the processors in today’s computers and embedded systems. Moore’s Law and related observations apply especially to the amount of memory that computers have for programs, the
amount of secondary storage (such as disk storage) they have to hold programs and data over
longer periods of time, and their processor speeds—the speeds at which computers execute
their programs (i.e., do their work). Similar growth hpas occurred in the communications
field, in which costs have plummeted as enormous demand for communications bandwidth
(i.e., information-carrying capacity) has attracted intense competition. We know of no other
fields in which technology improves so quickly and costs fall so rapidly. Such phenomenal
improvement is truly fostering the Information Revolution.
1.3.2 Computer Organization
Regardless of differences in physical appearance, computers can be envisioned as divided
into various logical units or sections (Fig. 1.2).
Logical unit
Description
Input unit
This “receiving” section obtains information (data and computer
programs) from input devices and places it at the disposal of the
other units for processing. Most information is entered into computers through keyboards, touch screens and mouse devices. Other
forms of input include receiving voice commands, scanning images
and barcodes, reading from secondary storage devices (such as hard
drives, DVD drives, Blu-ray Disc™ drives and USB flash drives—
also called “thumb drives” or “memory sticks”), receiving video from
a webcam and having your computer receive information from the
Internet (such as when you download videos from YouTube™ or ebooks from Amazon). Newer forms of input include position data
from a GPS device, and motion and orientation information from an
accelerometer in a smartphone or game controller (such as Microsoft® Kinect™, Wii™ Remote and PlayStation® Move).
This “shipping” section takes information that the computer has processed and places it on various output devices to make it available for
use outside the computer. Most information that’s output from computers today is displayed on screens, printed on paper, played as
audio or video on PCs and media players (such as Apple’s popular
iPods) and giant screens in sports stadiums, transmitted over the
Internet or used to control other devices, such as robots and “intelligent” appliances.
Output unit
Fig. 1.2 | Logical units of a computer. (Part 1 of 2.)
1.4 Data Hierarchy
Logical unit
Description
Memory unit
This rapid-access, relatively low-capacity “warehouse” section retains
information that has been entered through the input unit, making it
immediately available for processing when needed. The memory unit
also retains processed information until it can be placed on output
devices by the output unit. Information in the memory unit is volatile—it’s typically lost when the computer’s power is turned off. The
memory unit is often called either memory or primary memory.
Typical main memories on desktop and notebook computers contain
between 1 and 8 GB (GB stands for gigabytes; a gigabyte is approximately one billion bytes).
This “manufacturing” section performs calculations, such as addition,
subtraction, multiplication and division. It also contains the decision
mechanisms that allow the computer, for example, to compare two
items from the memory unit to determine whether they’re equal. In
today’s systems, the ALU is usually implemented as part of the next
logical unit, the CPU.
This “administrative” section coordinates and supervises the operation of the other sections. The CPU tells the input unit when information should be read into the memory unit, tells the ALU when
information from the memory unit should be used in calculations
and tells the output unit when to send information from the memory
unit to certain output devices. Many of today’s computers have multiple CPUs and, hence, can perform many operations simultaneously. A multi-core processor implements multiple processors on a
single integrated-circuit chip—a dual-core processor has two CPUs
and a quad-core processor has four CPUs. Today’s desktop computers
have processors that can execute billions of instructions per second.
This is the long-term, high-capacity “warehousing” section. Programs
or data not actively being used by the other units normally are placed
on secondary storage devices (e.g., your hard drive) until they’re again
needed, possibly hours, days, months or even years later. Information
on secondary storage devices is persistent—it’s preserved even when
the computer’s power is turned off. Secondary storage information
takes much longer to access than information in primary memory,
but the cost per unit of secondary storage is much less than that of
primary memory. Examples of secondary storage devices include CD
drives, DVD drives and flash drives, some of which can hold up to
512 GB. Typical hard drives on desktop and notebook computers
can hold up to 2 TB (TB stands for terabytes; a terabyte is approximately one trillion bytes).
Arithmetic
and logic unit
(ALU)
Central
processing
unit (CPU)
Secondary
storage unit
7
Fig. 1.2 | Logical units of a computer. (Part 2 of 2.)
1.4 Data Hierarchy
Data items processed by computers form a data hierarchy that becomes larger and more
complex in structure as we progress from bits to characters to fields, and so on. Figure 1.3
8
Chapter 1 Introduction to Computers, the Internet and the Web
illustrates a portion of the data hierarchy. Figure 1.4 summarizes the data hierarchy’s levels.
Judy
Black
Tom
Blue
Judy
Green
Iris
Orange
Randy
Red
Green
J u d y
File
Record
Field
00000000 01001010
1
Sally
16-bit Unicode character J
Bit
Fig. 1.3 | Data hierarchy.
Level
Description
Bits
The smallest data item in a computer can assume the value 0 or the value 1.
Such a data item is called a bit (short for “binary digit”—a digit that can
assume one of two values). It’s remarkable that the impressive functions performed by computers involve only the simplest manipulations of 0s and 1s—
examining a bit’s value, setting a bit’s value and reversing a bit’s value (from 1 to 0
or from 0 to 1).
It’s tedious for people to work with data in the low-level form of bits. Instead,
they prefer to work with decimal digits (0–9), letters (A–Z and a–z), and special
symbols (e.g., $, @, %, &, *, (, ), –, +, ", :, ? and / ). Digits, letters and special
symbols are known as characters. The computer’s character set is the set of all
the characters used to write programs and represent data items. Computers process only 1s and 0s, so every character is represented as a pattern of 1s and 0s.
The Unicode character set contains characters for many of the world’s languages. C supports several character sets, including 16-bit Unicode® characters
Characters
Fig. 1.4 | Levels of the data hierarchy. (Part 1 of 2.)
1.5 Programming Languages
Level
Description
Characters
(cont.)
that are composed of two bytes, each composed of eight bits. See Appendix B
for more information on the ASCII (American Standard Code for Information Interchange) character set—the popular subset of Unicode that represents
uppercase and lowercase letters, digits and some common special characters.
Fields
Just as characters are composed of bits, fields are composed of characters or
bytes. A field is a group of characters or bytes that conveys meaning. For example, a field consisting of uppercase and lowercase letters could be used to represent a person’s name, and a field consisting of decimal digits could represent a
person’s age.
Records
Several related fields can be used to compose a record. In a payroll system, for
example, the record for an employee might consist of the following fields (possible types for these fields are shown in parentheses):
• Employee identification number (a whole number)
• Name (a string of characters)
• Address (a string of characters)
• Hourly pay rate (a number with a decimal point)
• Year-to-date earnings (a number with a decimal point)
• Amount of taxes withheld (a number with a decimal point)
Thus, a record is a group of related fields. In the preceding example, all the
fields belong to the same employee. A company might have many employees
and a payroll record for each one.
Files
A file is a group of related records. [Note: More generally, a file contains arbitrary data in arbitrary formats. In some operating systems, a file is viewed simply as a sequence of bytes—any organization of the bytes in a file, such as
organizing the data into records, is a view created by the application programmer.] It’s not unusual for an organization to have many files, some containing
billions, or even trillions, of characters of information.
Database
A database is an electronic collection of data that’s organized for easy access and
manipulation. The most popular database model is the relational database in
which data is stored in simple tables. A table includes records and fields. For
example, a table of students might include first name, last name, major, year,
student ID number and grade point average. The data for each student is a
record, and the individual pieces of information in each record are the fields.
You can search, sort and manipulate the data based on its relationship to multiple tables or databases. For example, a university might use data from the student database in combination with databases of courses, on-campus housing,
meal plans, etc.
9
Fig. 1.4 | Levels of the data hierarchy. (Part 2 of 2.)
1.5 Programming Languages
Programmers write instructions in various programming languages, some directly understandable by computers and others requiring intermediate translation steps.
10
Chapter 1
Introduction to Computers, the Internet and the Web
Machine Languages
Any computer can directly understand only its own machine language, defined by its
hardware architecture. Machine languages generally consist of numbers (ultimately reduced to 1s and 0s). Such languages are cumbersome for humans.
Assembly Languages
Programming in machine language was simply too slow and tedious for most programmers. Instead, they began using Englishlike abbreviations to represent elementary operations. These abbreviations formed the basis of assembly languages. Translator programs
called assemblers were developed to convert assembly-language programs to machine language. Although assembly-language code is clearer to humans, it’s incomprehensible to
computers until translated to machine language.
High-Level Languages
To speed the programming process even further, high-level languages were developed in
which single statements could be written to accomplish substantial tasks. High-level languages allow you to write instructions that look almost like everyday English and contain
commonly used mathematical expressions. Translator programs called compilers convert
high-level language programs into machine language.
The process of compiling a large high-level language program into machine language
can take a considerable amount of computer time. Interpreter programs were developed
to execute high-level language programs directly (without the need for compilation),
although more slowly than compiled programs. Scripting languages such as JavaScript
and PHP are processed by interpreters.
Performance Tip 1.1
Interpreters have an advantage over compilers in Internet scripting. An interpreted program can begin executing as soon as it’s downloaded to the client’s machine, without needing to be compiled before it can execute. On the downside, interpreted scripts generally run
slower than compiled code.
1.6 The C Programming Language
C evolved from two previous languages, BCPL and B. BCPL was developed in 1967 by
Martin Richards as a language for writing operating systems and compilers. Ken Thompson modeled many features in his B language after their counterparts in BCPL, and in
1970 he used B to create early versions of the UNIX operating system at Bell Laboratories.
The C language was evolved from B by Dennis Ritchie at Bell Laboratories and was
originally implemented in 1972. C initially became widely known as the development language of the UNIX operating system. Many of today’s leading operating systems are
written in C and/or C++. C is mostly hardware independent—with careful design, it’s possible to write C programs that are portable to most computers.
Built for Performance
C is widely used to develop systems that demand performance, such as operating systems,
embedded systems, real-time systems and communications systems (Figure 1.5).
1.6 The C Programming Language
Application
Description
Operating systems
C’s portability and performance make it desirable for
implementing operating systems, such as Linux and
portions of Microsoft’s Windows and Google’s
Android. Apple’s OS X is built in Objective-C, which
was derived from C. We discuss some key popular
desktop/notebook operating systems and mobile
operating systems in Section 1.12.
The vast majority of the microprocessors produced
each year are embedded in devices other than generalpurpose computers. These embedded systems include
navigation systems, smart home appliances, home
security systems, smartphones, robots, intelligent traffic intersections and more. C is one of the most popular programming languages for developing embedded
systems, which typically need to run as fast as possible
and conserve memory. For example, a car’s anti-lock
brakes must respond immediately to slow or stop the
car without skidding; game controllers used for video
games should respond instantaneously to prevent any
lag between the controller and the action in the game,
and to ensure smooth animations.
Real-time systems are often used for “mission-critical” applications that require nearly instantaneous
response times. For example, an air-traffic-control
system must constantly monitor the positions and
velocities of the planes and report that information to
air-traffic controllers without delay so that they can
alert the planes to change course if there’s a possibility of a collision.
Communications systems need to route massive
amounts of data to their destinations quickly to
ensure that things such as audio and video are delivered smoothly and without delay.
Embedded systems
Real-time systems
Communications systems
11
Fig. 1.5 | Some popular performance-oriented C applications.
By the late 1970s, C had evolved into what’s now referred to as “traditional C.” The
publication in 1978 of Kernighan and Ritchie’s book, The C Programming Language, drew
wide attention to the language. This became one of the most successful computer science
books of all time.
Standardization
The rapid expansion of C over various types of computers (sometimes called hardware
platforms) led to many variations that were similar but often incompatible. This was a serious problem for programmers who needed to develop code that would run on several
12
Chapter 1
Introduction to Computers, the Internet and the Web
platforms. It became clear that a standard version of C was needed. In 1983, the X3J11
technical committee was created under the American National Standards Committee on
Computers and Information Processing (X3) to “provide an unambiguous and machineindependent definition of the language.” In 1989, the standard was approved as ANSI
X3.159-1989 in the United States through the American National Standards Institute
(ANSI), then worldwide through the International Standards Organization (ISO). We
call this simply Standard C. This standard was updated in 1999—its standards document
is referred to as INCITS/ISO/IEC 9899-1999 and often referred to simply as C99. Copies
may be ordered from the American National Standards Institute (www.ansi.org) at webstore.ansi.org/ansidocstore.
The New C Standard
We also introduce the new C standard (referred to as C11), which was approved as this
book went to publication. The new standard refines and expands the capabilities of C. Not
all popular C compilers support the new features. Of those that do, most implement only
a subset of the new features. We’ve integrated into the text (and appendices) in easy-toinclude-or-omit sections many of the new features implemented in leading compilers.
Portability Tip 1.1
Because C is a hardware-independent, widely available language, applications written in
C often can run with little or no modification on a range of different computer systems.
1.7 C Standard Library
As you’ll learn in Chapter 5, C programs consist of pieces called functions. You can program all the functions that you need to form a C program, but most C programmers take
advantage of the rich collection of existing functions called the C Standard Library. Thus,
there are really two parts to learning how to program in C—learning the C language itself
and learning how to use the functions in the C Standard Library. Throughout the book,
we discuss many of these functions. P. J. Plauger’s book The Standard C Library is must
reading for programmers who need a deep understanding of the library functions, how to
implement them and how to use them to write portable code. We use and explain many
C library functions throughout this text. Visit the following website for the C Standard
Library documentation:
www.dinkumware.com/manuals/#Standard%20C%20Library
C How to Program, 7/e encourages a building-block approach to creating programs.
Avoid “reinventing the wheel.” Instead, use existing pieces—this is called software reuse.
When programming in C you’ll typically use the following building blocks:
•
C Standard Library functions
•
Functions you create yourself
•
Functions other people (whom you trust) have created and made available to you
The advantage of creating your own functions is that you’ll know exactly how they
work. You’ll be able to examine the C code. The disadvantage is the time-consuming effort
that goes into designing, developing and debugging new functions.
1.8 C++ and Other C-Based Languages
13
Performance Tip 1.2
Using Standard C library functions instead of writing your own comparable versions can
improve program performance, because these functions are carefully written to perform efficiently.
Portability Tip 1.2
Using Standard C library functions instead of writing your own comparable versions can
improve program portability, because these functions are used in virtually all Standard C
implementations.
1.8 C++ and Other C-Based Languages
C++ was developed by Bjarne Stroustrup at Bell Laboratories. It has its roots in C, providing a number of features that “spruce up” the C language. More important, it provides capabilities for object-oriented programming. Objects are essentially reusable software
components that model items in the real world. Using a modular, object-oriented design
and implementation approach can make software development groups more productive.
Chapters 15–24 present a condensed treatment of C++ selected from our book C++
How to Program, 8/e. As you study C++, check out our online C++ Resource Center at
www.deitel.com/cplusplus/. Figure 1.6 introduces several other popular C-based programming languages.
Programming
language
Objective-C
Visual C#
Java
Description
Objective-C is an object-oriented language based on C. It was developed in
the early 1980s and later acquired by NeXT, which in turn was acquired by
Apple. It has become the key programming language for the Mac OS X
operating system and all iOS-based devices (such as iPods, iPhones and
iPads).
Microsoft’s three primary object-oriented programming languages are
Visual Basic, Visual C++ (based on C++) and C# (based on C++ and Java,
and developed for integrating the Internet and the web into computer
applications).
Sun Microsystems in 1991 funded an internal corporate research project
which resulted in the C++-based object-oriented programming language
called Java. A key goal of Java is to enable the writing of programs that will
run on a broad variety of computer systems and computer-controlled
devices. This is sometimes called “write once, run anywhere.” Java is used to
develop large-scale enterprise applications, to enhance the functionality of
web servers (the computers that provide the content we see in our web
browsers), to provide applications for consumer devices (smartphones, television set-top boxes and more) and for many other purposes.
Fig. 1.6 | Popular C-based programming languages. (Part 1 of 2.)
14
Chapter 1
Programming
language
PHP
JavaScript
Introduction to Computers, the Internet and the Web
Description
PHP—an object-oriented, open-source (see Section 1.12) scripting language
based on C and supported by a community of users and developers—is used
by many websites including Wikipedia and Facebook. PHP is platform independent—implementations exist for all major UNIX, Linux, Mac and Windows operating systems. PHP also supports many databases, including
MySQL. Other languages similar in concept to PHP are Perl and Python.
JavaScript—developed by Netscape—is the most widely used scripting language. It’s primarily used to add programmability to web pages—for example, animations and interactivity with the user. It’s provided with all major
web browsers.
Fig. 1.6 | Popular C-based programming languages. (Part 2 of 2.)
1.9 Object Technology
Building software quickly, correctly and economically remains an elusive goal at a time
when demands for new and more powerful software are soaring. Objects, or more precisely
the classes objects come from, are essentially reusable software components. There are date
objects, time objects, audio objects, video objects, automobile objects, people objects, etc.
Almost any noun can be reasonably represented as a software object in terms of attributes
(e.g., name, color and size) and behaviors (e.g., calculating, moving and communicating).
Software developers are discovering that using a modular, object-oriented design-andimplementation approach can make software-development groups much more productive
than was possible with earlier techniques—object-oriented programs are often easier to
understand, correct and modify.
The Automobile as an Object
Let’s begin with a simple analogy. Suppose you want to drive a car and make it go faster by
pressing its accelerator pedal. What must happen before you can do this? Well, before you
can drive a car, someone has to design it. A car typically begins as engineering drawings,
similar to the blueprints that describe the design of a house. These drawings include the
design for an accelerator pedal. The pedal hides from the driver the complex mechanisms
that actually make the car go faster, just as the brake pedal hides the mechanisms that slow
the car, and the steering wheel hides the mechanisms that turn the car. This enables people
with little or no knowledge of how engines, braking and steering mechanisms work to
drive a car easily.
Before you can drive a car, it must be built from the engineering drawings that
describe it. A completed car has an actual accelerator pedal to make the car go faster, but
even that’s not enough—the car won’t accelerate on its own (hopefully!), so the driver
must press the pedal to accelerate the car.
Methods and Classes
Let’s use our car example to introduce some key object-oriented programming concepts.
Performing a task in a program requires a method. The method houses the program state-
1.9 Object Technology
15
ments that actually perform its tasks. It hides these statements from its user, just as a car’s
accelerator pedal hides from the driver the mechanisms of making the car go faster. In object-oriented programming languages, we create a program unit called a class to house the
set of methods that perform the class’s tasks. For example, a class that represents a bank
account might contain one method to deposit money to an account, another to withdraw
money from an account and a third to inquire what the account’s current balance is. A class
is similar in concept to a car’s engineering drawings, which house the design of an accelerator pedal, steering wheel, and so on.
Instantiation
Just as someone has to build a car from its engineering drawings before you can actually
drive a car, you must build an object from a class before a program can perform the tasks
that the class’s methods define. The process of doing this is called instantiation. An object
is then referred to as an instance of its class.
Reuse
Just as a car’s engineering drawings can be reused many times to build many cars, you can
reuse a class many times to build many objects. Reuse of existing classes when building new
classes and programs saves time and effort. Reuse also helps you build more reliable and
effective systems, because existing classes and components often have gone through extensive testing, debugging and performance tuning. Just as the notion of interchangeable parts
was crucial to the Industrial Revolution, reusable classes are crucial to the software revolution that has been spurred by object technology.
Messages and Method Calls
When you drive a car, pressing its gas pedal sends a message to the car to perform a task—
that is, to go faster. Similarly, you send messages to an object. Each message is implemented
as a method call that tells a method of the object to perform its task. For example, a program might call a particular bank-account object’s deposit method to increase the account’s
balance.
Attributes and Instance Variables
A car, besides having capabilities to accomplish tasks, also has attributes, such as its color,
its number of doors, the amount of gas in its tank, its current speed and its record of total
miles driven (i.e., its odometer reading). Like its capabilities, the car’s attributes are represented as part of its design in its engineering diagrams (which, for example, include an
odometer and a fuel gauge). As you drive an actual car, these attributes are carried along
with the car. Every car maintains its own attributes. For example, each car knows how
much gas is in its own gas tank, but not how much is in the tanks of other cars.
An object, similarly, has attributes that it carries along as it’s used in a program. These
attributes are specified as part of the object’s class. For example, a bank-account object has
a balance attribute that represents the amount of money in the account. Each bankaccount object knows the balance in the account it represents, but not the balances of the
other accounts in the bank. Attributes are specified by the class’s instance variables.
Encapsulation
Classes encapsulate (i.e., wrap) attributes and methods into objects—an object’s attributes
and methods are intimately related. Objects may communicate with one another, but nor-
16
Chapter 1
Introduction to Computers, the Internet and the Web
mally they’re not allowed to know how other objects are implemented—implementation
details are hidden within the objects themselves. This information hiding is crucial to good
software engineering.
Inheritance
A new class of objects can be created quickly and conveniently by inheritance—the new
class absorbs the characteristics of an existing class, possibly customizing them and adding
unique characteristics of its own. In our car analogy, an object of class “convertible” certainly is an object of the more general class “automobile,” but more specifically, the roof can
be raised or lowered.
1.10 Typical C Program Development Environment
C systems generally consist of several parts: a program development environment, the language and the C Standard Library. The following discussion explains the typical C development environment shown in Fig. 1.7.
C programs typically go through six phases to be executed (Fig. 1.7). These are: edit,
preprocess, compile, link, load and execute. Although C How to Program, Seventh Edition
is a generic C textbook (written independently of the details of any particular operating
system), we concentrate in this section on a typical Linux-based C system. [Note: The programs in this book will run with little or no modification on most current C systems,
including Microsoft Windows-based systems.] If you’re not using a Linux system, refer to
the documentation for your system or ask your instructor how to accomplish these tasks
in your environment. Check out our C Resource Center at www.deitel.com/C to locate
“getting started” tutorials for popular C compilers and development environments.
1.10.1 Phase 1: Creating a Program
Phase 1 consists of editing a file. This is accomplished with an editor program. Two editors widely used on Linux systems are vi and emacs. Software packages for the C/C++ integrated program development environments such as Eclipse and Microsoft Visual Studio
have editors that are integrated into the programming environment. You type a C program
with the editor, make corrections if necessary, then store the program on a secondary storage device such as a hard disk. C program file names should end with the .c extension.
1.10.2 Phases 2 and 3: Preprocessing and Compiling a C Program
In Phase 2, the you give the command to compile the program. The compiler translates the
C program into machine language-code (also referred to as object code). In a C system, a
preprocessor program executes automatically before the compiler’s translation phase begins.
The C preprocessor obeys special commands called preprocessor directives, which indicate
that certain manipulations are to be performed on the program before compilation. These
manipulations usually consist of including other files in the file to be compiled and performing various text replacements. The most common preprocessor directives are discussed in the
early chapters; a detailed discussion of preprocessor features appears in Chapter 13.
In Phase 3, the compiler translates the C program into machine-language code. A
syntax error occurs when the compiler cannot recognize a statement because it violates the
1.10 Typical C Program Development Environment
Editor
Disk
Preprocessor
Disk
Compiler
Disk
Linker
Disk
17
Phase 1:
Programmer creates program
in the editor and stores it on
disk.
Phase 2:
Preprocessor program
processes the code.
Phase 3:
Compiler creates
object code and stores
it on disk.
Phase 4:
Linker links the object
code with the libraries,
creates an executable file and
stores it on disk.
Primary
Memory
Loader
Phase 5:
Loader puts program
in memory.
...
Disk
Primary
Memory
CPU
...
Phase 6:
CPU takes each
instruction and
executes it, possibly
storing new data
values as the program
executes.
Fig. 1.7 | Typical C development environment.
rules of the language. The compiler issues an error message to help you locate and fix the
incorrect statement. The C Standard does not specify the wording for error messages
issued by the compiler, so the error messages you see on your system may differ from those
on other systems. Syntax errors are also called compile errors, or compile-time errors.
18
Chapter 1
Introduction to Computers, the Internet and the Web
1.10.3 Phase 4: Linking
The next phase is called linking. C programs typically contain references to functions defined elsewhere, such as in the standard libraries or in the private libraries of groups of programmers working on a particular project. The object code produced by the C compiler
typically contains “holes” due to these missing parts. A linker links the object code with
the code for the missing functions to produce an executable image (with no missing pieces). On a typical Linux system, the command to compile and link a program is called gcc
(the GNU C compiler). To compile and link a program named welcome.c, type
gcc welcome.c
at the Linux prompt and press the Enter key (or Return key). [Note: Linux commands are
case sensitive; make sure that each c is lowercase and that the letters in the filename are in
the appropriate case.] If the program compiles and links correctly, a file called a.out is produced. This is the executable image of our welcome.c program.
1.10.4 Phase 5: Loading
The next phase is called loading. Before a program can be executed, the program must first
be placed in memory. This is done by the loader, which takes the executable image from
disk and transfers it to memory. Additional components from shared libraries that support
the program are also loaded.
1.10.5 Phase 6: Execution
Finally, the computer, under the control of its CPU, executes the program one instruction
at a time. To load and execute the program on a Linux system, type ./a.out at the Linux
prompt and press Enter.
1.10.6 Problems That May Occur at Execution Time
Programs do not always work on the first try. Each of the preceding phases can fail because
of various errors that we’ll discuss. For example, an executing program might attempt to
divide by zero (an illegal operation on computers just as in arithmetic). This would cause
the computer to display an error message. You would then return to the edit phase, make
the necessary corrections and proceed through the remaining phases again to determine
that the corrections work properly.
Common Programming Error 1.1
Errors such as division-by-zero occur as a program runs, so they are called runtime errors
or execution-time errors. Divide-by-zero is generally a fatal error, i.e., one that causes the
program to terminate immediately without successfully performing its job. Nonfatal errors
allow programs to run to completion, often producing incorrect results.
1.10.7 Standard Input, Standard Output and Standard Error Streams
Most C programs input and/or output data. Certain C functions take their input from
stdin (the standard input stream), which is normally the keyboard, but stdin can be connected to another stream. Data is often output to stdout (the standard output stream),
which is normally the computer screen, but stdout can be connected to another stream.
1.11 Test-Driving a C Application in Windows, Linux and Mac OS X
19
When we say that a program prints a result, we normally mean that the result is displayed
on a screen. Data may be output to devices such as disks and printers. There’s also a standard error stream referred to as stderr. The stderr stream (normally connected to the
screen) is used for displaying error messages. It’s common to route regular output data,
i.e., stdout, to a device other than the screen while keeping stderr assigned to the screen
so that the user can be immediately informed of errors.
1.11 Test-Driving a C Application in Windows, Linux
and Mac OS X
In this section, you’ll run and interact with your first C application. You’ll begin by running
a guess-the-number game, which randomly picks a number from 1 to 1000 and prompts you
to guess it. If your guess is correct, the game ends. If your guess is not correct, the application
indicates whether your guess is higher or lower than the correct number. There’s no limit on
the number of guesses you can make but you should be able to guess any of the numbers in
this range correctly in 10 or fewer tries. There’s some nice computer science behind this
game—in Section 6.8, Searching Arrays, you’ll explore the binary search technique.
For this test-drive only, we’ve modified this application from the exercise you’ll be
asked to create in Chapter 5. Normally this application randomly selects the correct
answers. The modified application uses the same sequence of correct answers every time
you execute the program (though this may vary by compiler), so you can use the same
guesses we use in this section and see the same results.
We’ll demonstrate running a C application using the Windows Command Prompt, a
shell on Linux and a Terminal window in Mac OS X. The application runs similarly on all
three platforms. After you perform the test drive for your platform, you can try the randomized version of the game, which we’ve provided with each test drive’s version of the
example in a subfolder named randomized_version.
Many development environments are available in which you can compile, build and
run C applications, such as GNU C, Dev C++, Microsoft Visual C++, CodeLite, NetBeans, Eclipse, Xcode, etc. Consult your instructor for information on your specific development environment. Most C++ development environments can compile both C and
C++ programs.
In the following steps, you’ll run the application and enter various numbers to guess the
correct number. The elements and functionality that you see in this application are typical
of those you’ll learn to program in this book. We use fonts to distinguish between features
you see on the screen (e.g., the Command Prompt) and elements that are not directly related
to the screen. We emphasize screen features like titles and menus (e.g., the File menu) in a
semibold sans-serif Helvetica font, and to emphasize filenames, text displayed by an application and values you should enter into an application (e.g., GuessNumber or 500) we use a
sans-serif Lucida font. As you’ve noticed, the defining occurrence of each key term is set
in bold blue type. For the Windows version of the test drive in this section, we’ve modified
the background color of the Command Prompt window to make the Command Prompt windows more readable. To modify the Command Prompt colors on your system, open a Command Prompt by selecting Start > All Programs > Accessories > Command Prompt, then right
click the title bar and select Properties. In the "Command Prompt" Properties dialog box that
appears, click the Colors tab, and select your preferred text and background colors.
20
Chapter 1
Introduction to Computers, the Internet and the Web
1.11.1 Running a C Application from the Windows Command Prompt
1. Checking your setup. It’s important to read the Before You Begin section at
www.deitel.com/books/chtp7/ to make sure that you’ve copied the book’s examples to your hard drive correctly.
2. Locating the completed application. Open a Command Prompt window. To
change to the directory for the completed GuessNumber application, type
cd C:\examples\ch01\GuessNumber\Windows, then press Enter (Fig. 1.8). The
command cd is used to change directories.
Fig. 1.8 | Opening a Command Prompt window and changing the directory.
3. Running the GuessNumber application. Now that you are in the directory that
contains the GuessNumber application, type the command GuessNumber
(Fig. 1.9) and press Enter. [Note: GuessNumber.exe is the actual name of the application; however, Windows assumes the .exe extension by default.]
Fig. 1.9 | Running the GuessNumber application.
4. Entering your first guess. The application displays "Please type your first
guess.", then displays a question mark (?) as a prompt on the next line
(Fig. 1.9). At the prompt, enter 500 (Fig. 1.10).
Fig. 1.10 | Entering your first guess.
5. Entering another guess. The application displays "Too high. Try again.", meaning that the value you entered is greater than the number the application chose as
1.11 Test-Driving a C Application in Windows, Linux and Mac OS X
21
the correct guess. So, you should enter a lower number for your next guess. At the
prompt, enter 250 (Fig. 1.11). The application again displays "Too high. Try
again.", because the value you entered is still greater than the number that the
application chose.
Fig. 1.11 | Entering a second guess and receiving feedback.
6. Entering additional guesses. Continue to play the game by entering values until
you guess the correct number. The application will display "Excellent! You
guessed the number!" (Fig. 1.12).
7. Playing the game again or exiting the application. After you guess correctly, the application asks if you’d like to play another game (Fig. 1.12). At the prompt, entering 1 causes the application to choose a new number and displays the message
“Please type your first guess.” followed by a question-mark prompt
(Fig. 1.13), so you can make your first guess in the new game. Entering 2 ends the
application and returns you to the application’s directory at the Command Prompt
Fig. 1.12 | Entering additional guesses and guessing the correct number.
22
Chapter 1
Introduction to Computers, the Internet and the Web
(Fig. 1.14). Each time you execute this application from the beginning (i.e., Step 3),
it will choose the same numbers for you to guess.
8. Close the Command Prompt window.
Fig. 1.13 | Playing the game again.
Fig. 1.14 | Exiting the game.
1.11.2 Running a C Application Using GNU C with Linux
For this test drive, we assume that you know how to copy the examples into your home
directory. Please see your instructor if you have any questions regarding copying the files
to your Linux system. Also, for the figures in this section, we use a bold font to point out
the user input required by each step. The prompt in the shell on our system uses the tilde
(~) character to represent the home directory, and each prompt ends with the dollar-sign
($) character. The prompt will vary among Linux systems.
1. Checking your setup. It’s important to read the Before You Begin section at
www.deitel.com/books/chtp7/ to make sure that you’ve copied the book’s examples to your hard drive correctly.
2. Locating the completed application. From a Linux shell, change to the completed
GuessNumber application directory (Fig. 1.15) by typing
cd examples/ch01/GuessNumber/GNU
then pressing Enter. The command cd is used to change directories.
~$ cd examples/ch01/GuessNumber/GNU
~/examples/ch01/GuessNumber/GNU$
Fig. 1.15 | Changing to the GuessNumber application’s directory.
1.11 Test-Driving a C Application in Windows, Linux and Mac OS X
23
3. Compiling the GuessNumber application. To run an application on the GNU
C++ compiler, you must first compile it by typing
gcc GuessNumber.c -o GuessNumber
as in Fig. 1.16. This command compiles the application and produces an executable file called GuessNumber.
~/examples/ch01/GuessNumber/GNU$ gcc GuessNumber.c -o GuessNumber
~/examples/ch01/GuessNumber/GNU$
Fig. 1.16 | Compiling the GuessNumber application using the gcc command.
4. Running the GuessNumber application. To run the executable file GuessNumber,
type ./GuessNumber at the next prompt, then press Enter (Fig. 1.17).
~/examples/ch01/GuessNumber/GNU$ ./GuessNumber
I have a number between 1 and 1000.
Can you guess my number?
Please type your first guess.
?
Fig. 1.17 | Running the GuessNumber application.
5. Entering your first guess. The application displays "Please type your first
guess.", then displays a question mark (?) as a prompt on the next line
(Fig. 1.17). At the prompt, enter 500 (Fig. 1.18).
~/examples/ch01/GuessNumber/GNU$ ./GuessNumber
I have a number between 1 and 1000.
Can you guess my number?
Please type your first guess.
? 500
Too high. Try again.
?
Fig. 1.18 | Entering an initial guess.
6. Entering another guess. The application displays "Too high. Try again.", meaning that the value you entered is greater than the number the application chose as
the correct guess (Fig. 1.18). At the next prompt, enter 250 (Fig. 1.19). This time
the application displays "Too low. Try again.", because the value you entered is
less than the correct guess.
7. Entering additional guesses. Continue to play the game (Fig. 1.20) by entering
values until you guess the correct number. When you guess correctly, the application displays "Excellent! You guessed the number!"
24
Chapter 1
Introduction to Computers, the Internet and the Web
~/examples/ch01/GuessNumber/GNU$ ./GuessNumber
I have a number between 1 and 1000.
Can you guess my number?
Please type your first guess.
? 500
Too high. Try again.
? 250
Too low. Try again.
?
Fig. 1.19 | Entering a second guess and receiving feedback.
Too low. Try again.
? 375
Too low. Try again.
? 437
Too high. Try again.
? 406
Too high. Try again.
? 391
Too high. Try again.
? 383
Too low. Try again.
? 387
Too high. Try again.
? 385
Too high. Try again.
? 384
Excellent! You guessed the number!
Would you like to play again?
Please type ( 1=yes, 2=no )?
Fig. 1.20 | Entering additional guesses and guessing the correct number.
8. Playing the game again or exiting the application. After you guess the correct
number, the application asks if you’d like to play another game. At the prompt,
entering 1 causes the application to choose a new number and displays the message "Please type your first guess." followed by a question-mark prompt
(Fig. 1.21) so that you can make your first guess in the new game. Entering 2 ends
the application and returns you to the application’s directory in the shell
(Fig. 1.22). Each time you execute this application from the beginning (i.e., Step
4), it will choose the same numbers for you to guess.
Excellent! You guessed the number!
Would you like to play again?
Please type ( 1=yes, 2=no )? 1
I have a number between 1 and 1000.
Can you guess my number?
Please type your first guess.
?
Fig. 1.21 | Playing the game again.
1.11 Test-Driving a C Application in Windows, Linux and Mac OS X
25
Excellent! You guessed the number!
Would you like to play again?
Please type ( 1=yes, 2=no )? 2
~/examples/ch01/GuessNumber/GNU$
Fig. 1.22 | Exiting the game.
1.11.3 Running a C Application Using GNU C with Mac OS X
For the figures in this section, we use a bold font to point out the user input required by
each step. You’ll use Mac OS X’s Terminal window to perform this test dive. To open a
Terminal window, click the Spotlight Search icon in the upper-right corner of your screen,
then type Terminal to locate the Terminal application. Under Applications in the Spotlight
Search results, select Terminal to open a Terminal window. The prompt in a Terminal window has the form hostName:~ userFolder$ to represent your user directory. For the figures
in this section we remove the hostName: part and used the generic name userFolder to represent your user account’s folder.
1. Checking your setup. It’s important to read the Before You Begin section at
www.deitel.com/books/chtp7/ to make sure that you’ve copied the book’s examples to your hard drive correctly. We assume that the examples are located in
your user account’s Documents/examples folder.
2. Locating the completed application. In the Terminal window, change to the completed GuessNumber application directory (Fig. 1.23) by typing
cd Documents/examples/ch01/GuessNumber/GNU
then pressing Enter. The command cd is used to change directories.
hostName:~ userFolder$ cd Documents/examples/ch01/GuessNumber/GNU
hostName:GNU$
Fig. 1.23 | Changing to the GuessNumber application’s directory.
3. Compiling the GuessNumber application. To run an application on the GNU C
compiler, you must first compile it by typing
gcc GuessNumber.c -o GuessNumber
as in Fig. 1.24. This command compiles the application and produces an executable file called GuessNumber.
hostName:GNU~ userFolder$ gcc GuessNumber.c -o GuessNumber
hostName:GNU~ userFolder$
Fig. 1.24 | Compiling the GuessNumber application using the gcc command.
4. Running the GuessNumber application. To run the executable file GuessNumber,
type ./GuessNumber at the next prompt, then press Enter (Fig. 1.25).
26
Chapter 1
Introduction to Computers, the Internet and the Web
hostName:GNU~ userFolder$ ./GuessNumber
I have a number between 1 and 1000.
Can you guess my number?
Please type your first guess.
?
Fig. 1.25 | Running the GuessNumber application.
5. Entering your first guess. The application displays "Please type your first
guess.", then displays a question mark (?) as a prompt on the next line
(Fig. 1.25). At the prompt, enter 500 (Fig. 1.26).
hostName:GNU~ userFolder$ ./GuessNumber
I have a number between 1 and 1000.
Can you guess my number?
Please type your first guess.
? 500
Too low. Try again.
?
Fig. 1.26 | Entering an initial guess.
6. Entering another guess. The application displays "Too low. Try again."
(Fig. 1.26), meaning that the value you entered is greater than the number the application chose as the correct guess. At the next prompt, enter 750 (Fig. 1.27).
Again the application displays "Too low. Try again.", because the value you entered is less than the correct guess.
hostName:GNU~ userFolder$ ./GuessNumber
I have a number between 1 and 1000.
Can you guess my number?
Please type your first guess.
? 500
Too low. Try again.
? 750
Too low. Try again.
?
Fig. 1.27 | Entering a second guess and receiving feedback.
7. Entering additional guesses. Continue to play the game (Fig. 1.28) by entering
values until you guess the correct number. When you guess correctly, the application displays "Excellent! You guessed the number!"
8. Playing the game again or exiting the application. After you guess the correct
number, the application asks if you’d like to play another game. At the prompt,
1.12 Operating Systems
27
Too low. Try again.
? 825
Too high. Try again.
? 788
Too low. Try again.
? 806
Too low. Try again.
? 815
Too high. Try again.
? 811
Too high. Try again.
? 808
Excellent! You guessed the number!
Would you like to play again?
Please type ( 1=yes, 2=no )?
Fig. 1.28 | Entering additional guesses and guessing the correct number.
entering 1 causes the application to choose a new number and displays the message "Please type your first guess." followed by a question-mark prompt
(Fig. 1.29) so you can make your first guess in the new game. Entering 2 ends the
application and returns you to the application’s folder in the Terminal window
(Fig. 1.30). Each time you execute this application from the beginning (i.e., Step
3), it will choose the same numbers for you to guess.
Excellent! You guessed the number!
Would you like to play again?
Please type ( 1=yes, 2=no )? 1
I have a number between 1 and 1000.
Can you guess my number?
Please type your first guess.
?
Fig. 1.29 | Playing the game again.
Excellent! You guessed the number!
Would you like to play again?
Please type ( 1=yes, 2=no )? 2
hostName:GNU~ userFolder$
Fig. 1.30 | Exiting the game.
1.12 Operating Systems
Operating systems are software systems that make using computers more convenient for
users, application developers and system administrators. They provide services that allow
each application to execute safely, efficiently and concurrently (i.e., in parallel) with other
28
Chapter 1
Introduction to Computers, the Internet and the Web
applications. The software that contains the core components of the operating system is
called the kernel. Popular desktop operating systems include Linux, Windows and Mac
OS X. Popular mobile operating systems used in smartphones and tablets include Google’s
Android, Apple’s iOS (for iPhone, iPad and iPod Touch devices), BlackBerry OS and
Windows Phone 7. You can develop applications in C for all four of the following key operating systems, including several of the latest mobile operating systems.
1.12.1 Windows—A Proprietary Operating System
In the mid-1980s, Microsoft developed the Windows operating system, consisting of a
graphical user interface built on top of DOS—an enormously popular personal-computer
operating system that users interacted with by typing commands. Windows borrowed from
many concepts (such as icons, menus and windows) developed by Xerox PARC and popularized by early Apple Macintosh operating systems. Windows 7 is Microsoft’s latest operating system—its features include enhancements to the user interface, faster startup
times, further refinement of security features, touch-screen and multitouch support, and
more. Windows is a proprietary operating system—it’s controlled by Microsoft exclusively.
Windows is by far the world’s most widely used operating system.
1.12.2 Linux—An Open-Source Operating System
The Linux operating system is perhaps the greatest success of the open-source movement.
Open-source software departs from the proprietary software development style that dominated software’s early years. With open-source development, individuals and companies
contribute their efforts in developing, maintaining and evolving software in exchange for
the right to use that software for their own purposes, typically at no charge. Open-source
code is often scrutinized by a much larger audience than proprietary software, so errors often get removed faster. Open source also encourages more innovation. Enterprise systems
companies, such as IBM, Oracle and many others, have made significant investments in
Linux open-source development.
Some key organizations in the open-source community are the Eclipse Foundation
(the Eclipse Integrated Development Environment helps programmers conveniently
develop software), the Mozilla Foundation (creators of the Firefox web browser), the
Apache Software Foundation (creators of the Apache web server used to develop webbased applications) and SourceForge (which provides the tools for managing open-source
projects—it has over 322,000 of them under development). Rapid improvements to computing and communications, decreasing costs and open-source software have made it
much easier and more economical to create a software-based business now than just a
decade ago. A great example is Facebook, which was launched from a college dorm room
and built with open-source software.
The Linux kernel is the core of the most popular open-source, freely distributed, fullfeatured operating system. It’s developed by a loosely organized team of volunteers and is
popular in servers, personal computers and embedded systems. Unlike that of proprietary
operating systems like Microsoft’s Windows and Apple’s Mac OS X, Linux source code
(the program code) is available to the public for examination and modification and is free
to download and install. As a result, Linux users benefit from a community of developers
1.12 Operating Systems
29
actively debugging and improving the kernel, an absence of licensing fees and restrictions,
and the ability to completely customize the operating system to meet specific needs.
A variety of issues—such as Microsoft’s market power, the small number of userfriendly Linux applications and the diversity of Linux distributions, such as Red Hat
Linux, Ubuntu Linux and many others—have prevented widespread Linux use on
desktop computers. Linux has become extremely popular on servers and in embedded systems, such as Google’s Android-based smartphones.
1.12.3 Apple’s Mac OS X; Apple’s iOS for iPhone®, iPad® and iPod
Touch® Devices
Apple, founded in 1976 by Steve Jobs and Steve Wozniak, quickly became a leader in personal computing. In 1979, Jobs and several Apple employees visited Xerox PARC (Palo
Alto Research Center) to learn about Xerox’s desktop computer that featured a graphical
user interface (GUI). That GUI served as the inspiration for the Apple Macintosh,
launched with much fanfare in a memorable Super Bowl ad in 1984.
The Objective-C programming language, created by Brad Cox and Tom Love at
Stepstone in the early 1980s, added capabilities for object-oriented programming (OOP)
to the C programming language. Steve Jobs left Apple in 1985 and founded NeXT Inc.
In 1988, NeXT licensed Objective-C from StepStone and developed an Objective-C compiler and libraries which were used as the platform for the NeXTSTEP operating system’s
user interface and Interface Builder—used to construct graphical user interfaces.
Jobs returned to Apple in 1996 when Apple bought NeXT. Apple’s Mac OS X operating system is a descendant of NeXTSTEP. Apple’s proprietary operating system, iOS, is
derived from Apple’s Mac OS X and is used in the iPhone, iPad and iPod Touch devices.
1.12.4 Google’s Android
Android—the fastest growing mobile and smartphone operating system—is based on the
Linux kernel and Java. Experienced Java programmers can quickly dive into Android development. One benefit of developing Android apps is the openness of the platform. The
operating system is open source and free.
The Android operating system was developed by Android, Inc., which was acquired
by Google in 2005. In 2007, the Open Handset Alliance™—a consortium of 34 companies initially and 84 by 2011—was formed to continue developing Android. As of June
2011, more than 500,000 Android smartphones were being activated each day!1 Android
smartphones are now outselling iPhones in the United States.2 The Android operating
system is used in numerous smartphones (such as the Motorola Droid, HTC EVO™ 4G,
Samsung Vibrant™ and many more), e-reader devices (such as the Barnes and Noble
Nook™), tablet computers (such as the Dell Streak and the Samsung Galaxy Tab), instore touch-screen kiosks, cars, robots, multimedia players and more.
1.
2.
news.cnet.com/8301-13506_3-20074956-17/google-500000-android-devices-activatedeach-day/.
www.pcworld.com/article/196035/android_outsells_the_iphone_no_big_surprise.html.
30
Chapter 1
Introduction to Computers, the Internet and the Web
1.13 The Internet and World Wide Web
The Internet—a global network of computers—was made possible by the convergence of
computing and communications technologies. In the late 1960s, ARPA (the Advanced Research Projects Agency) rolled out blueprints for networking the main computer systems
of about a dozen ARPA-funded universities and research institutions. Academic research
was about to take a giant leap forward. ARPA proceeded to implement the ARPANET,
which eventually evolved into today’s Internet. It rapidly became clear that communicating quickly and easily via electronic mail was the key early benefit of the ARPANET. This
is true even today on the Internet, which facilitates communications of all kinds among
the world’s Internet users.
Packet Switching
A primary goal for ARPANET was to allow multiple users to send and receive information
simultaneously over the same communications paths (e.g., phone lines). The network operated with a technique called packet switching, in which digital data was sent in small
bundles called packets. The packets contained address, error-control and sequencing information. The address information allowed packets to be routed to their destinations. The
sequencing information helped in reassembling the packets—which, because of complex
routing mechanisms, could actually arrive out of order—into their original order for presentation to the recipient. Packets from different senders were intermixed on the same lines
to efficiently use the available bandwidth. This packet-switching technique greatly reduced transmission costs, as compared with the cost of dedicated communications lines.
The network was designed to operate without centralized control. If a portion of the
network failed, the remaining working portions would still route packets from senders to
receivers over alternative paths for reliability.
TCP/IP
The protocol (i.e., set of rules) for communicating over the ARPANET became known as
TCP—the Transmission Control Protocol. TCP ensured that messages were properly
routed from sender to receiver and that they arrived intact.
As the Internet evolved, organizations worldwide were implementing their own networks. One challenge was to get these different networks to communicate. ARPA accomplished this with the development of IP—the Internet Protocol, truly creating a network
of networks, the current architecture of the Internet. The combined set of protocols is now
commonly called TCP/IP.
World Wide Web, HTML, HTTP
The World Wide Web allows you to execute web-based applications and to locate and
view multimedia-based documents on almost any subject over the Internet. The web is a
relatively recent creation. In 1989, Tim Berners-Lee of CERN (the European Organization for Nuclear Research) began to develop a technology for sharing information via hyperlinked text documents. Berners-Lee called his invention the HyperText Markup
Language (HTML). He also wrote communication protocols to form the backbone of his
new information system, which he called the World Wide Web. In particular, he wrote
the Hypertext Transfer Protocol (HTTP)—a communications protocol used to send information over the web. The URL (Uniform Resource Locator) specifies the address (i.e.,
1.14 Some Key Software Development Terminology
31
location) of the web page displayed in the browser window. Each web page on the Internet
is associated with a unique URL. Hypertext Transfer Protocol Secure (HTTPS) is the
standard for transferring encrypted data on the web.
Mosaic, Netscape, Emergence of Web 2.0
Web use exploded with the availability in 1993 of the Mosaic browser, which featured a
user-friendly graphical interface. Marc Andreessen, whose team at the National Center for
Supercomputing Applications developed Mosaic, went on to found Netscape, the company that many people credit with igniting the explosive Internet economy of the late 1990s.
In 2003 there was a noticeable shift in how people and businesses were using the web
and developing web-based applications. The term Web 2.0 was coined by Dale Dougherty
of O’Reilly Media3 in 2003 to describe this trend. Generally, Web 2.0 companies use the
web as a platform to create collaborative, community-based sites (e.g., social networking
sites, blogs, wikis).
Companies with Web 2.0 characteristics are Google (web search), YouTube (video
sharing), Facebook (social networking), Twitter (microblogging), Groupon (social commerce), Foursquare (mobile check-in), Salesforce (business software offered as online services “in the cloud”), Craigslist (mostly free classified listings), Flickr (photo sharing),
Skype (Internet telephony and video calling and conferencing) and Wikipedia (a free
online encyclopedia).
Web 2.0 involves the users—not only do they create content, but they help organize
it, share it, remix it, critique it, update it, etc. Web 2.0 is a conversation, with everyone
having the opportunity to speak and share views. Companies that understand Web 2.0
realize that their products and services are conversations as well.
Architecture of Participation
Web 2.0 embraces an architecture of participation—a design that encourages user interaction and community contributions. You, the user, are the most important aspect of Web
2.0—so important, in fact, that in 2006, TIME magazine’s “Person of the Year” was
“You.”4 The article recognized the social phenomenon of Web 2.0—the shift away from
a powerful few to an empowered many. Popular blogs now compete with traditional media
powerhouses, and many Web 2.0 companies are built almost entirely on user-generated
content. For websites like Facebook®, Twitter™, YouTube, eBay® and Wikipedia® users
create the content, while the companies provide the platforms on which to enter, manipulate and share the information.
1.14 Some Key Software Development Terminology
Figure 1.31 lists a number of buzzwords that you’ll hear in the software development community. We’ve created Resource Centers on most of these topics, with more on the way.
3.
T. O’Reilly, “What is Web 2.0: Design Patterns and Business Models for the Next Generation
of Software.” September 2005 <http://www.oreillynet.com/pub/a/oreilly/tim/news/2005/
4.
L. Grossman, “TIME’s Person of the Year: You.” TIME, December 2006
09/30/what-is-web-20.html?page=1>.
www.time.com/time/magazine/article/0,9171,1569514,00.html>.
<http://
32
Chapter 1
Introduction to Computers, the Internet and the Web
Technology
Description
Ajax
Ajax is one of the premier Web 2.0 software technologies. Ajax helps Internet-based applications perform like desktop applications—a difficult task,
given that such applications suffer transmission delays as data is shuttled
back and forth between your computer and servers on the Internet.
Agile software development is a set of methodologies that try to get software implemented faster and using fewer resources than previous methodologies. Check out the Agile Alliance (www.agilealliance.org) and the Agile
Manifesto (www.agilemanifesto.org).
Refactoring involves reworking programs to make them clearer and easier
to maintain while preserving their correctness and functionality. It’s widely
employed with agile development methodologies. Many IDEs include refactoring tools to do major portions of the reworking automatically.
Design patterns are proven architectures for constructing flexible and maintainable object-oriented software. The field of design patterns tries to enumerate those recurring patterns, encouraging software designers to reuse
them to develop better-quality software using less time, money and effort.
MySQL is an open-source database management system. PHP is the most
popular open-source server-side Internet “scripting” language for developing Internet-based applications. LAMP is an acronym for the set of opensource technologies that many developers use to build web applications—it
stands for Linux, Apache, MySQL and PHP (or Perl or Python—two other
languages used for similar purposes).
Software has generally been viewed as a product; most software still is
offered this way. If you want to run an application, you buy a software package from a software vendor—often a CD, DVD or web download. You then
install that software on your computer and run it as needed. As new versions
of the software appear, you upgrade your software, often requiring significant time and at considerable expense. This process can become cumbersome for organizations with tens of thousands of systems that must be
maintained on a diverse array of computer equipment. With Software as a
Service (SaaS), the software runs on servers elsewhere on the Internet.
When that server is updated, all clients worldwide see the new capabilities—no local installation is needed. You access the service through a
browser. Browsers are quite portable, so you can run the same applications
on a wide variety of computers from anywhere in the world. Salesforce.com,
Google, and Microsoft’s Office Live and Windows Live all offer SaaS. SaaS
is a capability of cloud computing.
Platform as a Service (PaaS), another capability of cloud computing, provides
a computing platform for developing and running applications as a service
over the web, rather than installing the tools on your computer. PaaS providers include Google App Engine, Amazon EC2, Bungee Labs and more.
Software Development Kits (SDKs) include the tools and documentation
developers use to program applications.
Agile software
development
Refactoring
Design
patterns
LAMP
Software as a
Service (SaaS)
Platform as a
Service (PaaS)
Software
Development
Kit (SDK)
Fig. 1.31 | Software technologies.
1.15 Keeping Up-to-Date with Information Technologies
33
Figure 1.32 describes software product-release categories.
Version
Description
Alpha
An alpha version is the earliest release of a software product that’s still under
active development. Alpha versions are often buggy, incomplete and unstable
and are released to a relatively small number of developers for testing new features, getting early feedback, etc.
Beta versions are released to a larger number of developers later in the development process after most major bugs have been fixed and new features are
nearly complete. Beta software is more stable, but still subject to change.
Release candidates are generally feature complete and (supposedly) bug free and
ready for use by the community, which provides a diverse testing environment—the software is used on different systems, with varying constraints and
for a variety of purposes. Any bugs that appear are corrected, and eventually
the final product is released to the general public. Software companies often
distribute incremental updates over the Internet.
Software that’s developed using this approach generally does not have version
numbers (for example, Google search or Gmail). The software, which is
hosted in the cloud (not installed on your computer), is constantly evolving
so that users always have the latest version.
Beta
Release
candidates
Continuous
beta
Fig. 1.32 | Software product-release terminology.
1.15 Keeping Up-to-Date with Information Technologies
Figure 1.33 lists key technical and business publications that will help you stay up-to-date
with the latest news and trends and technology. You can also find a growing list of
Internet- and web-related Resource Centers at www.deitel.com/resourcecenters.html.
Publication
URL
ACM TechNews
ACM Transactions on
Accessible Computing
ACM Transactions on Internet
Technology
Bloomberg BusinessWeek
CNET
Communications of the ACM
Computer World
Engadget
eWeek
technews.acm.org/
www.gccis.rit.edu/taccess/index.html
toit.acm.org/
www.businessweek.com
news.cnet.com
cacm.acm.org/
www.computerworld.com
www.engadget.com
www.eweek.com
Fig. 1.33 | Technical and business publications. (Part 1 of 2.)
34
Chapter 1
Introduction to Computers, the Internet and the Web
Publication
URL
Fast Company
Fortune
IEEE Computer
IEEE Internet Computing
InfoWorld
Mashable
PCWorld
SD Times
Slashdot
Smarter Technology
Technology Review
Techcrunch
Wired
www.fastcompany.com/
money.cnn.com/magazines/fortune/
www.computer.org/portal/web/computer
www.computer.org/portal/web/internet/home
www.infoworld.com
mashable.com
www.pcworld.com
www.sdtimes.com
slashdot.org/
www.smartertechnology.com
technologyreview.com
techcrunch.com
www.wired.com
Fig. 1.33 | Technical and business publications. (Part 2 of 2.)
1.16 Web Resources
This section provides links to our C and related Resource Centers that will be useful to
you as you learn C. These Resource Centers include various C resources, including blogs,
articles, whitepapers, compilers, development tools, downloads, FAQs, tutorials, webcasts,
wikis and links to resources for C game programming with the Allegro libraries. For updates on Deitel publications, Resource Centers, training courses, partner offers and more,
follow us on Facebook® at www.facebook.com/deitelfan/ and on Twitter® @deitel.
Deitel & Associates Websites
www.deitel.com/books/chtp7/
The Deitel & Associates C How to Program, 7/e site. Here you’ll find links to the book’s examples
and other resources.
www.deitel.com/C/
www.deitel.com/visualcplusplus/
www.deitel.com/codesearchengines/
www.deitel.com/programmingprojects/
Check these Resource Centers for compilers, code downloads, tutorials, documentation, books, ebooks, articles, blogs, RSS feeds and more that will help you develop C applications.
www.deitel.com
Check this site for updates, corrections and additional resources for all Deitel publications.
www.deitel.com/newsletter/subscribe.html
Subscribe here for the Deitel® Buzz Online e-mail newsletter to follow the Deitel & Associates publishing program, including updates and errata to C How to Program, 7/e.
Terminology
35
Terminology
actions (computers perform) 2
agile software development 32
Ajax 32
American National Standards Institute
(ANSI) 12
Android 29
architecture of participation 31
arithmetic and logic unit (ALU) 7
ASCII 9
assembler 10
assembly language 10
bit 8
bytes 8
C preprocessor 16
C Standard Library 12
central processing unit (CPU) 7
character 8
character set 8
class 15
cloud computing 4
compile 16
compile error 17
compile phase 16
compile-time error 17
compiler 10
components (software) 13
computer program 5
data hierarchy 7
database 9
decisions (made by computers) 2
design pattern 32
edit phase 16
editor program 16
embedded systems 11
encapsulate 15
executable image 18
execute 18
execute phase 16
field 9
file 9
function 12
gcc compilation command 18
hardware 2
hardware platform 11
high-level language 10
information hiding 16
inheritance 16
input device 6
input unit 6
instance 15
instance variable 15
International Standards Organization (ISO) 12
interpreter 10
iOS 29
kernel 28
LAMP 32
link phase 16
linker 18
linking 18
Linux 28
load phase 16
loader 18
loading 18
logical unit 6
machine language 10
memory 7
memory unit 7
method 14
method call 15
Moore’s Law 6
multi-core processor 7
object 13
object code 16
object-oriented programming (OOP) 13
open source 28
operating system 27
output device 6
output unit 6
Platform as a Service (PaaS) 32
portable program 10
preprocess phase 16
preprocessor 16
preprocessor directive 16
primary memory 7
programmer 5
record 9
refactoring 32
scripting language 10
secondary storage unit 7
software 2
Software as a Service (SaaS) 32
Software Development Kit (SDK) 32
standard error stream (stderr) 19
standard input stream (stdin) 18
standard output stream (stdout) 18
syntax error 16
Unicode 8
Web 2.0 31
Windows Operating System 28
36
Chapter 1
Introduction to Computers, the Internet and the Web
Self-Review Exercises
1.1
Fill in the blanks in each of the following:
a) Computers process data under the control of sequences of instructions called computer
.
is a type of computer language that uses Englishlike abbreviations for mab)
chine-language instructions.
c)
languages are most convenient to the programmer for writing programs
quickly and easily.
.
d) The only language a computer can directly understand is that computer’s
e) The programs that translate high-level language programs into machine language are
.
called
f) With
development, individuals and companies contribute their efforts in developing, maintaining and evolving software in exchange for the right to use that software for their own purposes, typically at no charge.
g) C is widely known as the development language of the
operating system.
1.2
Fill in the blanks in each of the following sentences about the C environment.
program.
a) C programs are normally typed into a computer using a(n)
b) In a C system, a(n)
program automatically executes before the translation
phase begins.
and
.
c) The two most common kinds of preprocessor directives are
d) The
program combines the output of the compiler with various library functions to produce an executable image.
program transfers the executable image from disk to memory.
e) The
1.3
Fill in the blanks in each of the following statements (based on Section 1.9):
a) Objects have the property of
—although objects may know how to communicate with one another across well-defined interfaces, they normally are not allowed to
know how other objects are implemented.
b) In object-oriented programming languages, we create
to house the set of
methods that perform tasks.
, new classes of objects are derived by absorbing characteristics of existing
c) With
classes, then adding unique characteristics of their own.
d) The size, shape, color and weight of an object are considered
of the object’s class.
Answers to Self-Review Exercises
1.1
a) programs. b) Assembly language. c) High-level. d) machine language. e) compilers.
f) open-source. g) UNIX.
1.2
a) editor. b) preprocessor. c) including other files in the file to be compiled, performing various text replacements. d) linker. e) loader.
1.3
a) information hiding. b) classes. c) inheritance. d) attributes.
Exercises
1.4
Categorize each of the following items as either hardware or software:
a) CPU
b) C++ compiler
c) ALU
d) C++ preprocessor
e) input unit
f) an editor program
Making a Difference
37
Fill in the blanks in each of the following statements:
.
a) The process of instructing the computer to solve a problem is called
b) What type of computer language uses Englishlike abbreviations for machine-language
instructions?
.
c) The level of computer language at which it’s most convenient to write programs quickly
.
and easily is
d) The only language that a computer directly understands is called that computer's
.
e) Web 2.0 embraces a(n)
—a design that encourages user interaction and community contributions.
1.6
Fill in the blanks in each of the following statements:
a)
is now used to develop large-scale enterprise applications, to enhance the
functionality of web servers, to provide applications for consumer devices and for many
other purposes.
b)
initially became widely known as the development language of the UNIX operating system.
programming language was developed by Bjarne Stroustrup in the early
c) The
1980s at Bell Laboratories.
1.7
Discuss the meaning of each of the following names:
a) stdin
b) stdout
c) stderr
1.8
Why is so much attention today focused on object-oriented programming?
1.9
(Internet in Industry and Research) Figure 1.1 provides examples of how computers and
the Internet are being used in industry and research. Find three additional examples and describe
how each is using the Internet and the web.
1.10 (Cloud Computing) Describe three benefits of the cloud computing model.
1.11 (Internet Negatives) Besides their numerous benefits, the Internet and the web have several
downsides, such as privacy issues, identity theft, spam and malware. Research some of the negative aspects of the Internet. List five problems and describe what could possibly be done to help solve each.
1.12 (Watch as an Object) You are probably wearing on your wrist one of the most common
types of objects—a watch. Discuss how each of the following terms and concepts applies to the notion of a watch: object, attributes, behaviors, class, inheritance (consider, for example, an alarm
clock), messages, encapsulation and information hiding.
1.5
Making a Difference
1.13 (Test-Drive: Carbon Footprint Calculator) Some scientists believe that carbon emissions,
especially from the burning of fossil fuels, contribute significantly to global warming and that this
can be combatted if individuals take steps to limit their use of carbon-based fuels. Organizations and
individuals are increasingly concerned about their “carbon footprints.” Websites such as TerraPass
www.terrapass.com/carbon-footprint-calculator/
and Carbon Footprint
www.carbonfootprint.com/calculator.aspx
provide carbon footprint calculators. Test-drive these calculators to estimate your carbon footprint.
Exercises in later chapters will ask you to program your own carbon footprint calculator. To prepare for this, use the web to research the formulas for calculating carbon footprints.
38
Chapter 1
Introduction to Computers, the Internet and the Web
1.14 (Test-Drive: Body Mass Index Calculator) By recent estimates, two-thirds of the people in
the United States are overweight and about half of those are obese. This causes significant increases
in illnesses such as diabetes and heart disease. To determine whether a person is overweight or obese,
you can use a measure called the body mass index (BMI). The United States Department of Health
and Human Services provides a BMI calculator at www.nhlbisupport.com/bmi/. Use it to calculate
your own BMI. An exercise in Chapter 2 will ask you to program your own BMI calculator. To prepare for this, use the web to research the formulas for calculating BMI.
1.15 (Gender Neutrality) Many people want to eliminate sexism in all forms of communication.
You’ve been asked to create a program that can process a paragraph of text and replace gender-specific words with gender-neutral ones. Assuming that you’ve been given a list of gender-specific
words and their gender-neutral replacements (e.g., replace “wife” with “spouse,” “man” with “person,” “daughter” with “child” and so on), explain the procedure you’d use to read through a paragraph of text and manually perform these replacements. How might your procedure generate a
strange term like “woperchild?” In Chapter 4, you’ll learn that a more formal term for “procedure”
is “algorithm,” and that an algorithm specifies the steps to be performed and the order in which to
perform them.
1.16 (Privacy) Some online e-mail services save all e-mail correspondence for some period of
time. Suppose a disgruntled employee were to post all of the e-mail correspondences for millions of
people, including yours, on the Internet. Discuss the issues.
1.17 (Programmer Responsibility and Liability) As a programmer in industry, you may develop
software that could affect people’s health or even their lives. Suppose a software bug in one of your
programs causes a cancer patient to receive an excessive dose during radiation therapy and that the
person is severely injured or dies. Discuss the issues.
1.18 (2010 “Flash Crash”) An example of the consequences of our excessive dependence on
computers was the so-called “flash crash” which occurred on May 6, 2010, when the U.S. stock market fell precipitously in a matter of minutes, wiping out trillions of dollars of investments, and then
recovered within minutes. Research online the causes of this crash and discuss the issues it raises.
1.19 (Making a Difference Projects) The following is a list of just a few worldwide organizations
that are working to make a difference. Visit these sites and our Making a Difference Resource Center
at www.deitel.com/makingadifference. Prepare a top-10 list of programming projects that you
think could indeed “make a difference.”
•
www.imaginecup.com/
The Microsoft Image Cup is a global competition in which students use technology to try to solve
some of the world’s most difficult problems, such as environmental sustainability, ending hunger, emergency response, literacy and combating HIV/AIDS. Visit www.imaginecup.com/about
for more information about the competition and to learn about the projects developed by previous winners. You can also find several project ideas submitted by worldwide charitable organizations at www.imaginecup.com/students/imagine-cup-solve-this. For additional ideas for
programming projects that can make a difference, search the web for “making a difference” and
visit the following websites:
•
www.un.org/millenniumgoals
The United Nations Millennium Project seeks solutions to major worldwide issues such as environmental sustainability, gender equality, child and maternal health, universal education and
more.
•
www.ibm.com/smarterplanet/
The IBM® Smarter Planet website
discusses how IBM is using technology to solve issues related
to business, cloud computing, education, sustainability and more.
Making a Difference
•
39
www.gatesfoundation.org/Pages/home.aspx
The Bill and Melinda Gates Foundation provides grants to organizations that work to alleviate
hunger, poverty and disease in developing countries. In the United States, the foundation focuses
on improving public education, particularly for people with few resources.
•
www.nethope.org/
NetHope is a collaboration of humanitarian organizations worldwide working to solve technology problems such as connectivity, emergency response and more.
•
www.rainforestfoundation.org/home
The Rainforest Foundation works to preserve rainforests and to protect the rights of the indigenous people who call the rainforests home. The site includes a list of things you can do to help.
•
www.undp.org/
The United Nations Development Programme (UNDP) seeks solutions to global challenges
such as crisis prevention and recovery, energy and the environment and democratic governance.
•
www.unido.org
The United Nations Industrial Development Organization (UNIDO) seeks to reduce poverty,
give developing countries the opportunity to participate in global trade, and promote energy efficiency and sustainability.
•
www.usaid.gov/
USAID promotes global democracy, health, economic growth, conflict prevention, humanitarian aid and more.
•
www.toyota.com/ideas-for-good/
Toyota’s Ideas for Good website describes several Toyota technologies that are making a difference—including their Advanced Parking Guidance System, Hybrid Synergy Drive®, Solar Powered Ventilation System, T.H.U.M.S. (Total Human Model for Safety) and Touch Tracer
Display. You can participate in the Ideas for Good challenge by submitting a short essay or video
describing how these technologies can be used for other good purposes.
2
What’s in a name?
That which we call a rose
By any other name would
smell as sweet.
—William Shakespeare
“Take some more tea,” the
March Hare said to Alice, very
earnestly. “I’ve had nothing yet,”
Alice replied in an offended
tone: “so I can’t take more.” “You
mean you can’t take less,” said
the Hatter: “it’s very easy to take
more than nothing.”
—Lewis Carroll
High thoughts must have high
language.
—Aristophanes
Objectives
In this chapter, you’ll:
■
■
■
■
■
■
■
Write simple computer
programs in C.
Use simple input and output
statements.
Use the fundamental data
types.
Learn computer memory
concepts.
Use arithmetic operators.
Learn the precedence of
arithmetic operators.
Write simple decisionmaking statements.
Introduction to C
Programming
2.1 Introduction
2.1 Introduction
2.2 A Simple C Program: Printing a Line
of Text
2.3 Another Simple C Program: Adding
Two Integers
41
2.4 Memory Concepts
2.5 Arithmetic in C
2.6 Decision Making: Equality and
Relational Operators
2.7 Secure C Programming
Summary | Terminology | Self-Review Exercises | Answers to Self-Review Exercises | Exercises |
Making a Difference
2.1 Introduction
The C language facilitates a structured and disciplined approach to computer-program
design. In this chapter we introduce C programming and present several examples that
illustrate many important features of C. Each example is analyzed one statement at a time.
In Chapters 3 and 4 we present an introduction to structured programming in C. We then
use the structured approach throughout the remainder of the C portion of the text.
2.2 A Simple C Program: Printing a Line of Text
C uses some notations that may appear strange to people who have not programmed computers. We begin by considering a simple C program. Our first example prints a line of
text. The program and its screen output are shown in Fig. 2.1.
1
2
3
4
5
6
7
8
9
// Fig. 2.1: fig02_01.c
// A first program in C.
#include <stdio.h>
// function main begins program execution
int main( void )
{
printf( "Welcome to C!\n" );
} // end function main
Welcome to C!
Fig. 2.1 | A first program in C.
Comments
Even though this program is simple, it illustrates several important features of the C language. Lines 1 and 2
// Fig. 2.1: fig02_01.c
// A first program in C
begin with //, indicating that these two lines are comments. You insert comments to document programs and improve program readability. Comments do not cause the computer
to perform any action when the program is run. Comments are ignored by the C compiler
and do not cause any machine-language object code to be generated. The preceding com-
42
Chapter 2
Introduction to C Programming
ment simply describes the figure number, file name and purpose of the program. Comments also help other people read and understand your program.
You can also use /*…*/ multi-line comments in which everything from /* on the
first line to */ at the end of the last line is a comment. We prefer // comments because
they’re shorter and they eliminate common programming errors that occur with /*…*/
comments, especially when the closing */ is omitted.
#include
Preprocessor Directive
Line 3
#include <stdio.h>
is a directive to the C preprocessor. Lines beginning with # are processed by the preprocessor before compilation. Line 3 tells the preprocessor to include the contents of the standard input/output header (<stdio.h>) in the program. This header contains information
used by the compiler when compiling calls to standard input/output library functions such
as printf (line 8). We explain the contents of headers in more detail in Chapter 5.
Blank Lines and White Space
Line 4 is simply a blank line. You use blank lines, space characters and tab characters (i.e.,
“tabs”) to make programs easier to read. Together, these characters are known as white
space. White-space characters are normally ignored by the compiler.
The main Function
Line 6
int main( void )
is a part of every C program. The parentheses after main indicate that main is a program
building block called a function. C programs contain one or more functions, one of which
must be main. Every program in C begins executing at the function main. Functions can
return information. The keyword int to the left of main indicates that main “returns” an
integer (whole-number) value. We’ll explain what it means for a function to “return a value” when we demonstrate how to create your own functions in Chapter 5. For now, simply include the keyword int to the left of main in each of your programs. Functions also
can receive information when they’re called upon to execute. The void in parentheses here
means that main does not receive any information. In Chapter 14, we’ll show an example
of main receiving information.
Good Programming Practice 2.1
Every function should be preceded by a comment describing the purpose of the function.
A left brace, {, begins the body of every function (line 7). A corresponding right brace
ends each function (line 9). This pair of braces and the portion of the program between
the braces is called a block. The block is an important program unit in C.
An Output Statement
Line 8
printf( "Welcome to C!\n" );
2.2 A Simple C Program: Printing a Line of Text
43
instructs the computer to perform an action, namely to print on the screen the string of
characters marked by the quotation marks. A string is sometimes called a character string,
a message or a literal. The entire line, including the printf function (the “f” stands for
“formatted”), its argument within the parentheses and the semicolon (;), is called a statement. Every statement must end with a semicolon (also known as the statement terminator). When the preceding printf statement is executed, it prints the message Welcome to
C! on the screen. The characters normally print exactly as they appear between the double
quotes in the printf statement.
Escape Sequences
Notice that the characters \n were not printed on the screen. The backslash (\) is called an
escape character. It indicates that printf is supposed to do something out of the ordinary.
When encountering a backslash in a string, the compiler looks ahead at the next character
and combines it with the backslash to form an escape sequence. The escape sequence \n
means newline. When a newline appears in the string output by a printf, the newline
causes the cursor to position to the beginning of the next line on the screen. Some common escape sequences are listed in Fig. 2.2.
Escape sequence
Description
\n
Newline. Position the cursor at the beginning of the next line.
Horizontal tab. Move the cursor to the next tab stop.
Alert. Produces a sound or visible alert without changing the current
cursor position.
Backslash. Insert a backslash character in a string.
Double quote. Insert a double-quote character in a string.
\t
\a
\\
\"
Fig. 2.2 | Some common escape sequences .
Because the backslash has special meaning in a string, i.e., the compiler recognizes it
as an escape character, we use a double backslash (\\) to place a single backslash in a string.
Printing a double quote also presents a problem because double quotes mark the boundaries of a string—such quotes are not printed. By using the escape sequence \" in a string
to be output by printf, we indicate that printf should display a double quote. The right
brace, }, (line 9) indicates that the end of main has been reached.
Good Programming Practice 2.2
Add a comment to the line containing the right brace, }, that closes every function, including main.
We said that printf causes the computer to perform an action. As any program
executes, it performs a variety of actions and makes decisions. Section 2.6 discusses decision making. Chapter 3 discusses this action/decision model of programming in depth.
The Linker and Executables
Standard library functions like printf and scanf are not part of the C programming language. For example, the compiler cannot find a spelling error in printf or scanf. When
44
Chapter 2
Introduction to C Programming
the compiler compiles a printf statement, it merely provides space in the object program
for a “call” to the library function. But the compiler does not know where the library functions are—the linker does. When the linker runs, it locates the library functions and inserts
the proper calls to these library functions in the object program. Now the object program
is complete and ready to be executed. For this reason, the linked program is called an executable. If the function name is misspelled, the linker will spot the error, because it will
not be able to match the name in the C program with the name of any known function in
the libraries.
Common Programming Error 2.1
Mistyping the name of the output function printf as print in a program.
Good Programming Practice 2.3
Indent the entire body of each function one level of indentation (we recommend three
spaces) within the braces that define the body of the function. This indentation emphasizes
the functional structure of programs and helps make programs easier to read.
Good Programming Practice 2.4
Set a convention for the size of indent you prefer and then uniformly apply that convention. The tab key may be used to create indents, but tab stops may vary.
Using Multiple printfs
The printf function can print Welcome to C! several different ways. For example, the program of Fig. 2.3 produces the same output as the program of Fig. 2.1. This works because
each printf resumes printing where the previous printf stopped printing. The first
printf (line 8) prints Welcome followed by a space, and the second printf (line 9) begins
printing on the same line immediately following the space.
1
2
3
4
5
6
7
8
9
10
// Fig. 2.3: fig02_03.c
// Printing on one line with two printf statements.
#include <stdio.h>
// function main begins program execution
int main( void )
{
printf( "Welcome " );
printf( "to C!\n" );
} // end function main
Welcome to C!
Fig. 2.3 | Printing on one line with two printf statements.
One printf can print several lines by using additional newline characters as in
Fig. 2.4. Each time the \n (newline) escape sequence is encountered, output continues at
the beginning of the next line.
2.3 Another Simple C Program: Adding Two Integers
1
2
3
4
5
6
7
8
9
45
// Fig. 2.4: fig02_04.c
// Printing multiple lines with a single printf.
#include <stdio.h>
// function main begins program execution
int main( void )
{
printf( "Welcome\nto\nC!\n" );
} // end function main
Welcome
to
C!
Fig. 2.4 | Printing multiple lines with a single printf.
2.3 Another Simple C Program: Adding Two Integers
Our next program uses the Standard Library function scanf to obtain two integers typed
by a user at the keyboard, computes the sum of these values and prints the result using
printf. The program and sample output are shown in Fig. 2.5. [In the input/output dialog of Fig. 2.5, we emphasize the numbers entered by the user in bold.]
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
// Fig. 2.5: fig02_05.c
// Addition program.
#include <stdio.h>
// function main begins program execution
int main( void )
{
int integer1; // first number to be entered by user
int integer2; // second number to be entered by user
int sum; // variable in which sum will be stored
printf( "Enter first integer\n" ); // prompt
scanf( "%d", &integer1 ); // read an integer
printf( "Enter second integer\n" ); // prompt
scanf( "%d", &integer2 ); // read an integer
sum = integer1 + integer2; // assign total to sum
printf( "Sum is %d\n", sum ); // print sum
} // end function main
Enter first integer
45
Enter second integer
72
Sum is 117
Fig. 2.5 | Addition program.
46
Chapter 2
Introduction to C Programming
The comment in lines 1–2 states the purpose of the program. As we stated earlier,
every program begins execution with main. The left brace { (line 7) marks the beginning
of the body of main, and the corresponding right brace } (line 21) marks the end of main.
Variables and Variable Definitions
Lines 8–10
int integer1; // first number to be entered by user
int integer2; // second number to be entered by user
int sum; // variable in which sum will be stored
are definitions. The names integer1, integer2 and sum are the names of variables—locations in memory where values can be stored for use by a program. These definitions
specify that variables integer1, integer2 and sum are of type int, which means that
they’ll hold integer values, i.e., whole numbers such as 7, –11, 0, 31914 and the like.
All variables must be defined with a name and a data type before they can be used in a
program. For readers using the Microsoft Visual C++ compiler, note that we’re placing our
variable definitions immediately after the left brace that begins the body of main. The C standard allows you to place each variable definition anywhere in main before that variable’s first
use in the code. Some compilers, such as GNU gcc, have implemented this capability. We’ll
address this issue in more depth in later chapters.
The preceding definitions could have been combined into a single definition statement as follows:
int integer1, integer2, sum;
but that would have made it difficult to describe the variables with corresponding comments as we did in lines 8–10.
Identifiers and Case Sensitivity
A variable name in C is any valid identifier. An identifier is a series of characters consisting
of letters, digits and underscores ( _ ) that does not begin with a digit. C is case sensitive—
uppercase and lowercase letters are different in C, so a1 and A1 are different identifiers.
Common Programming Error 2.2
Using a capital letter where a lowercase letter should be used (for example, typing Main
instead of main).
Error-Prevention Tip 2.1
Avoid starting identifiers with the underscore character ( _ ) to prevent conflicts with compiler-generated identifiers and standard library identifiers.
Good Programming Practice 2.5
Choosing meaningful variable names helps make a program self-documenting—that is,
fewer comments are needed.
Good Programming Practice 2.6
The first letter of an identifier used as a simple variable name should be a lowercase letter.
Later in the text we’ll assign special significance to identifiers that begin with a capital
letter and to identifiers that use all capital letters.
2.3 Another Simple C Program: Adding Two Integers
47
Good Programming Practice 2.7
Multiple-word variable names can help make a program more readable. Separate the words
with underscores as in total_commissions, or, if you run the words together, begin each
word after the first with a capital letter as in totalCommissions. The latter style is preferred.
Syntax Errors
We discussed what syntax errors are in Chapter 1. Recall that the Microsoft Visual C++
compiler requires variable definitions to be placed after the left brace of a function and before any executable statements. Therefore, in the program in Fig. 2.5, inserting the definition of integer1 after the first printf would cause a syntax error in Visual C++.
Common Programming Error 2.3
Placing variable definitions among executable statements causes syntax errors in the Microsoft Visual C++ Compiler.
Prompting Messages
Line 12
printf( "Enter first integer\n" ); // prompt
displays the literal "Enter first integer" and positions the cursor to the beginning of the
next line. This message is called a prompt because it tells the user to take a specific action.
The scanf Function and Formatted Inputs
The next statement
scanf( "%d", &integer1 ); // read an integer
uses scanf (the “f” stands for “formatted”) to obtain a value from the user. The function
reads from the standard input, which is usually the keyboard. This scanf has two arguments,
"%d" and &integer1. The first, the format control string, indicates the type of data that
should be entered by the user. The %d conversion specifier indicates that the data should be
an integer (the letter d stands for “decimal integer”). The % in this context is treated by scanf
(and printf as we’ll see) as a special character that begins a conversion specifier. The second
argument of scanf begins with an ampersand (&)—called the address operator—followed
by the variable name. The &, when combined with the variable name, tells scanf the location
(or address) in memory at which the variable integer1 is stored. The computer then stores
the value that the user enters for integer1 at that location. The use of ampersand (&) is often
confusing to novice programmers or to people who have programmed in other languages
that do not require this notation. For now, just remember to precede each variable in every
call to scanf with an ampersand. Some exceptions to this rule are discussed in Chapters 6
and 7. The use of the ampersand will become clear after we study pointers in Chapter 7.
Good Programming Practice 2.8
Place a space after each comma (,) to make programs more readable.
When the computer executes the preceding scanf, it waits for the user to enter a value
for variable integer1. The user responds by typing an integer, then pressing the Enter key
to send the number to the computer. The computer then assigns this number, or value, to
48
Chapter 2
Introduction to C Programming
the variable integer1. Any subsequent references to integer1 in this program will use this
same value. Functions printf and scanf facilitate interaction between the user and the computer. Because this interaction resembles a dialogue, it’s often called interactive computing.
Line 15
printf( "Enter second integer\n" ); // prompt
displays the message Enter second integer on the screen, then positions the cursor to the
beginning of the next line. This printf also prompts the user to take action.
Line 16
scanf( "%d", &integer2 ); // read an integer
obtains a value for variable integer2 from the user.
Assignment Statement
The assignment statement in line 18
sum = integer1 + integer2; // assign total to sum
calculates the total of variables integer1 and integer2 and assigns the result to variable
using the assignment operator =. The statement is read as, “sum gets the value of
integer1 + integer2.” Most calculations are performed in assignments. The = operator
and the + operator are called binary operators because each has two operands. The + operator’s two operands are integer1 and integer2. The = operator’s two operands are sum
and the value of the expression integer1 + integer2.
sum
Good Programming Practice 2.9
Place spaces on either side of a binary operator. This makes the operator stand out and
makes the program more readable.
Common Programming Error 2.4
A calculation in an assignment statement must be on the right side of the = operator. It’s
a compilation error to place a calculation on the left side of an assignment operator.
Printing with a Format Control String
Line 20
printf( "Sum is %d\n", sum ); // print sum
calls function printf to print the literal Sum is followed by the numerical value of variable
sum on the screen. This printf has two arguments, "Sum is %d\n" and sum. The first argument is the format control string. It contains some literal characters to be displayed, and
it contains the conversion specifier %d indicating that an integer will be printed. The second argument specifies the value to be printed. Notice that the conversion specifier for an
integer is the same in both printf and scanf—this is the case for most C data types.
Calculations in printf Statements
Calculations can also be performed inside printf statements. We could have combined
the previous two statements into the statement
printf( "Sum is %d\n", integer1 + integer2 );
The right brace, }, at line 21 indicates that the end of function main has been reached.
2.4 Memory Concepts
49
Common Programming Error 2.5
Forgetting to precede a variable in a scanf statement with an ampersand when that variable should, in fact, be preceded by an ampersand results in an execution-time error. On
many systems, this causes a “segmentation fault” or “access violation.” Such an error occurs
when a user’s program attempts to access a part of the computer’s memory to which it does
not have access privileges. The precise cause of this error will be explained in Chapter 7.
Common Programming Error 2.6
Preceding a variable included in a printf statement with an ampersand when, in fact,
that variable should not be preceded by an ampersand.
2.4 Memory Concepts
Variable names such as integer1, integer2 and sum actually correspond to locations in
the computer’s memory. Every variable has a name, a type and a value.
In the addition program of Fig. 2.5, when the statement (line 13)
scanf( "%d", &integer1 ); // read an integer
is executed, the value entered by the user is placed into a memory location to which the
name integer1 has been assigned. Suppose the user enters the number 45 as the value for
integer1. The computer will place 45 into location integer1, as shown in Fig. 2.6.
integer1
45
Fig. 2.6 | Memory location showing the name and value of a variable.
Whenever a value is placed in a memory location, the value replaces the previous value
in that location; thus, this process is said to be destructive.
Returning to our addition program again, when the statement (line 16)
scanf( "%d", &integer2 ); // read an integer
executes, suppose the user enters the value 72. This value is placed into location integer2,
and memory appears as in Fig. 2.7. These locations are not necessarily adjacent in memory.
Once the program has obtained values for integer1 and integer2, it adds these
values and places the total into variable sum. The statement (line 18)
sum = integer1 + integer2; // assign total to sum
integer1
45
integer2
72
Fig. 2.7 | Memory locations after both variables are input.
50
Chapter 2
Introduction to C Programming
that performs the addition also replaces whatever value was stored in sum. This occurs when
the calculated total of integer1 and integer2 is placed into location sum (destroying the
value already in sum). After sum is calculated, memory appears as in Fig. 2.8. The values of
integer1 and integer2 appear exactly as they did before they were used in the calculation.
They were used, but not destroyed, as the computer performed the calculation. Thus, when
a value is read from a memory location, the process is said to be nondestructive.
integer1
45
integer2
72
sum
117
Fig. 2.8 | Memory locations after a calculation.
2.5 Arithmetic in C
Most C programs perform calculations using the C arithmetic operators (Fig. 2.9). Note
the use of various special symbols not used in algebra. The asterisk (*) indicates multiplication and the percent sign (%) denotes the remainder operator, which is introduced below.
In algebra, to multiply a times b, we simply place these single-letter variable names side by
side, as in ab. In C, however, if we were to do this, ab would be interpreted as a single,
two-letter name (or identifier). Therefore, C (and many other programming languages) require that multiplication be explicitly denoted by using the * operator, as in a * b. The
arithmetic operators are all binary operators. For example, the expression 3 + 7 contains
the binary operator + and the operands 3 and 7.
C operation
Arithmetic operator
Algebraic expression
C expression
Addition
Subtraction
Multiplication
+
f+7
p–c
bm
f + 7
–
*
Division
/
x
x / y or -- or x ÷ y
y
Remainder
%
r mod s
p - c
b * m
x / y
r % s
Fig. 2.9 | Arithmetic operators.
Integer Division and the Remainder Operator
Integer division yields an integer result. For example, the expression 7 / 4 evaluates to 1
and the expression 17 / 5 evaluates to 3. C provides the remainder operator, %, which
yields the remainder after integer division. The remainder operator is an integer operator
that can be used only with integer operands. The expression x % y yields the remainder after x is divided by y. Thus, 7 % 4 yields 3 and 17 % 5 yields 2. We’ll discuss many interesting
applications of the remainder operator.
2.5 Arithmetic in C
51
Common Programming Error 2.7
An attempt to divide by zero is normally undefined on computer systems and generally results in a fatal error, i.e., an error that causes the program to terminate immediately without having successfully performed its job. Nonfatal errors allow programs to run to
completion, often producing incorrect results.
Arithmetic Expressions in Straight-Line Form
Arithmetic expressions in C must be written in straight-line form to facilitate entering
programs into the computer. Thus, expressions such as “a divided by b” must be written
as a/b so that all operators and operands appear in a straight line. The algebraic notation
--a
b
is generally not acceptable to compilers, although some special-purpose software packages
do support more natural notation for complex mathematical expressions.
Parentheses for Grouping Subexpressions
Parentheses are used in C expressions in the same manner as in algebraic expressions. For
example, to multiply a times the quantity b + c we write a * ( b + c ).
Rules of Operator Precedence
C applies the operators in arithmetic expressions in a precise sequence determined by the
following rules of operator precedence, which are generally the same as those in algebra:
1. Operators in expressions contained within pairs of parentheses are evaluated first.
Parentheses are said to be at the “highest level of precedence.” In cases of nested,
or embedded, parentheses, such as
( ( a + b ) + c )
the operators in the innermost pair of parentheses are applied first.
2. Multiplication, division and remainder operations are applied next. If an expression contains several multiplication, division and remainder operations, evaluation proceeds from left to right. Multiplication, division and remainder are said
to be on the same level of precedence.
3. Addition and subtraction operations are evaluated next. If an expression contains
several addition and subtraction operations, evaluation proceeds from left to right.
Addition and subtraction also have the same level of precedence, which is lower
than the precedence of the multiplication, division and remainder operations.
4. The assignment operator (=) is evaluated last.
The rules of operator precedence specify the order C uses to evaluate expressions.1
When we say evaluation proceeds from left to right, we’re referring to the associativity of
the operators. We’ll see that some operators associate from right to left. Figure 2.10 summarizes these rules of operator precedence for the operators we’ve seen so far.
1.
We use simple examples to explain the order of evaluation of expressions. Subtle issues occur in more
complex expressions that you’ll encounter later in the book. We’ll discuss these issues as they arise.
52
Chapter 2
Introduction to C Programming
Operator(s)
Operation(s)
Order of evaluation (precedence)
( )
Parentheses
Evaluated first. If the parentheses are nested,
the expression in the innermost pair is evaluated first. If there are several pairs of parentheses “on the same level” (i.e., not nested),
they’re evaluated left to right.
*
/
%
Multiplication
Division
Remainder
Evaluated second. If there are several, they’re
evaluated left to right.
+
-
Addition
Subtraction
Evaluated third. If there are several, they’re
evaluated left to right.
=
Assignment
Evaluated last.
Fig. 2.10 | Precedence of arithmetic operators.
Sample Algebraic and C Expressions
Now let’s consider several expressions in light of the rules of operator precedence. Each
example lists an algebraic expression and its C equivalent. The following expression calculates the arithmetic mean (average) of five terms.
+b+c+d+e
Algebra: m = a------------------------------------5
C:
m = ( a + b + c + d + e ) / 5;
The parentheses are required to group the additions because division has higher precedence than addition. The entire quantity ( a + b + c + d + e ) should be divided by 5. If
the parentheses are erroneously omitted, we obtain a + b + c + d + e / 5, which evaluates
incorrectly as
a + b + c + d + --e5
The following expression is the equation of a straight line:
Algebra: y = mx + b
C:
y = m * x + b;
No parentheses are required. The multiplication is evaluated first because multiplication
has a higher precedence than addition.
The following expression contains remainder (%), multiplication, division, addition,
subtraction and assignment operations:
Algebra:
C:
z = pr %q + w/x – y
z
=
6
p
*
1
r
%
2
q
+
4
w
/
3
x
- y;
5
The circled numbers indicate the order in which C evaluates the operators. The multiplication, remainder and division are evaluated first in left-to-right order (i.e., they associate
2.5 Arithmetic in C
53
from left to right) because they have higher precedence than addition and subtraction. The
addition and subtraction are evaluated next. They’re also evaluated left to right. Finally,
the result is assigned to the variable z.
Not all expressions with several pairs of parentheses contain nested parentheses. For
example, the following expression does not contain nested parentheses—instead, the
parentheses are said to be “on the same level.”
a * ( b + c ) + c * ( d + e )
Evaluation of a Second-Degree Polynomial
To develop a better understanding of the rules of operator precedence, let’s see how C evaluates a second-degree polynomial.
y
=
a
6
*
1
x
*
x
2
+
4
b
*
x
3
+ c;
5
The circled numbers under the statement indicate the order in which C performs the operations. There’s no arithmetic operator for exponentiation in C, so we’ve represented x2 as
x * x. The C Standard Library includes the pow (“power”) function to perform exponentiation. Because of some subtle issues related to the data types required by pow, we defer
a detailed explanation of pow until Chapter 4.
Suppose variables a, b, c and x in the preceding second-degree polynomial are initialized as follows: a = 2, b = 3, c = 7 and x = 5. Figure 2.11 illustrates the order in which the
operators are applied.
Step 1.
y = 2 * 5 * 5 + 3 * 5 + 7;
(Leftmost multiplication)
2 * 5 is 10
Step 2.
y = 10 * 5 + 3 * 5 + 7;
(Leftmost multiplication)
10 * 5 is 50
Step 3.
y = 50 + 3 * 5 + 7;
(Multiplication before addition)
3 * 5 is 15
Step 4.
y = 50 + 15 + 7;
(Leftmost addition)
50 + 15 is 65
Step 5.
y = 65 + 7;
(Last addition)
65 + 7 is 72
Step 6.
y = 72
(Last operation—place 72 in y)
Fig. 2.11 | Order in which a second-degree polynomial is evaluated.
54
Chapter 2
Introduction to C Programming
As in algebra, it’s acceptable to place unnecessary parentheses in an expression to make
the expression clearer. These are called redundant parentheses. For example, the preceding statement could be parenthesized as follows:
y = ( a * x * x ) + ( b * x ) + c;
2.6 Decision Making: Equality and Relational Operators
Executable statements either perform actions (such as calculations or input or output of data)
or make decisions (we’ll soon see several examples of these). We might make a decision in a
program, for example, to determine whether a person’s grade on an exam is greater than or
equal to 60 and whether the program should print the message “Congratulations! You
passed.” This section introduces a simple version of C’s if statement that allows a program
to make a decision based on the truth or falsity of a statement of fact called a condition. If
the condition is true (i.e., the condition is met), the statement in the body of the if statement is executed. If the condition is false (i.e., the condition isn’t met), the body statement
isn’t executed. Whether the body statement is executed or not, after the if statement completes, execution proceeds with the next statement after the if statement.
Conditions in if statements are formed by using the equality operators and relational
operators summarized in Fig. 2.12. The relational operators all have the same level of precedence and they associate left to right. The equality operators have a lower level of precedence
than the relational operators and they also associate left to right. [Note: In C, a condition may
actually be any expression that generates a zero (false) or nonzero (true) value.]
Common Programming Error 2.8
A syntax error occurs if the two symbols in any of the operators ==, !=, >= and <= are separated by spaces.
Common Programming Error 2.9
Confusing the equality operator == with the assignment operator. To avoid this confusion,
the equality operator should be read “double equals” and the assignment operator should
be read “gets” or “is assigned the value of.” As you’ll see, confusing these operators may not
cause an easy-to-recognize compilation error, but may cause extremely subtle logic errors.
Algebraic equality or
relational operator
C equality or
relational operator
Example of
C condition
Meaning of C condition
==
x == y
x
!=
x != y
x
>
x > y
x
<
x < y
x
>=
x >= y
<=
x <= y
Equality operators
=
≠
is equal to y
is not equal to y
Relational operators
>
<
≥
≤
Fig. 2.12 | Equality and relational operators.
is greater than y
is less than y
x is greater than or equal to y
x is less than or equal to y
2.6 Decision Making: Equality and Relational Operators
55
Figure 2.13 uses six if statements to compare two numbers entered by the user. If the
condition in any of these if statements is true, the printf statement associated with that
if executes. The program and three sample execution outputs are shown in the figure.
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
// Fig. 2.13: fig02_13.c
// Using if statements, relational
// operators, and equality operators.
#include <stdio.h>
// function main begins program execution
int main( void )
{
int num1; // first number to be read from user
int num2; // second number to be read from user
printf( "Enter two integers, and I will tell you\n" );
printf( "the relationships they satisfy: " );
scanf( "%d%d", &num1, &num2 ); // read two integers
if ( num1 == num2 ) {
printf( "%d is equal to %d\n", num1, num2 );
} // end if
if ( num1 != num2 ) {
printf( "%d is not equal to %d\n", num1, num2 );
} // end if
if ( num1 < num2 ) {
printf( "%d is less than %d\n", num1, num2 );
} // end if
if ( num1 > num2 ) {
printf( "%d is greater than %d\n", num1, num2 );
} // end if
if ( num1 <= num2 ) {
printf( "%d is less than or equal to %d\n", num1, num2 );
} // end if
if ( num1 >= num2 ) {
printf( "%d is greater than or equal to %d\n", num1, num2 );
} // end if
} // end function main
Enter two integers, and I will tell you
the relationships they satisfy: 3 7
3 is not equal to 7
3 is less than 7
3 is less than or equal to 7
Fig. 2.13 | Using if statements, relational operators, and equality operators. (Part 1 of 2.)
56
Chapter 2
Introduction to C Programming
Enter two integers, and I will tell you
the relationships they satisfy: 22 12
22 is not equal to 12
22 is greater than 12
22 is greater than or equal to 12
Enter two integers, and I will tell you
the relationships they satisfy: 7 7
7 is equal to 7
7 is less than or equal to 7
7 is greater than or equal to 7
Fig. 2.13 | Using if statements, relational operators, and equality operators. (Part 2 of 2.)
The program uses scanf (line 15) to input two numbers. Each conversion specifier
has a corresponding argument in which a value will be stored. The first %d converts a value
to be stored in the variable num1, and the second %d converts a value to be stored in the
variable num2.
Good Programming Practice 2.10
Although it’s allowed, there should be no more than one statement per line in a program.
Common Programming Error 2.10
Placing commas (when none are needed) between conversion specifiers in the format control string of a scanf statement.
Comparing Numbers
The if statement in lines 17–19
if ( num1 == num2 ) {
printf( "%d is equal to %d\n", num1, num2 );
}
compares the values of variables num1 and num2 to test for equality. If the values are equal,
the statement in line 18 displays a line of text indicating that the numbers are equal. If the
conditions are true in one or more of the if statements starting in lines 21, 25, 29, 33
and 37, the corresponding body statement displays an appropriate line of text. Indenting
the body of each if statement and placing blank lines above and below each if statement
enhances program readability.
Common Programming Error 2.11
Placing a semicolon immediately to the right of the right parenthesis after the condition
in an if statement.
A left brace, {, begins the body of each if statement (e.g., line 17). A corresponding
right brace, }, ends each if statement’s body (e.g., line 19). Any number of statements can
be placed in the body of an if statement.2
2.6 Decision Making: Equality and Relational Operators
57
Good Programming Practice 2.11
A lengthy statement may be spread over several lines. If a statement must be split across
lines, choose breaking points that make sense (such as after a comma in a comma-separated
list). If a statement is split across two or more lines, indent all subsequent lines. It’s not
correct to split identifiers.
Figure 2.14 lists from highest to lowest the precedence of the operators introduced in
this chapter. Operators are shown top to bottom in decreasing order of precedence. The
equals sign is also an operator. All these operators, with the exception of the assignment
operator =, associate from left to right. The assignment operator (=) associates from right
to left.
Good Programming Practice 2.12
Refer to the operator precedence chart when writing expressions containing many operators. Confirm that the operators in the expression are applied in the proper order. If you’re
uncertain about the order of evaluation in a complex expression, use parentheses to group
expressions or break the statement into several simpler statements. Be sure to observe that
some of C’s operators such as the assignment operator (=) associate from right to left rather
than from left to right.
Operators
Associativity
()
left to right
left to right
left to right
left to right
left to right
right to left
*
/
+
-
<
<=
==
!=
=
%
>
>=
Fig. 2.14 | Precedence and associativity of the operators discussed so far.
Some of the words we’ve used in the C programs in this chapter—in particular int
and if—are keywords or reserved words of the language. Figure 2.15 contains the C keywords. These words have special meaning to the C compiler, so you must be careful not
to use these as identifiers such as variable names.
In this chapter, we’ve introduced many important features of the C programming language, including displaying data on the screen, inputting data from the user, performing
calculations and making decisions. In the next chapter, we build upon these techniques as
we introduce structured programming. You’ll become more familiar with indentation
techniques. We’ll study how to specify the order in which statements are executed—this is
called flow of control.
2.
Using braces to delimit the body of an if statement is optional when the body contains only one
statement. Many programmers consider it good practice to always use these braces. In Chapter 3,
we’ll explain the issues.
58
Chapter 2
Introduction to C Programming
Keywords
auto
double
int
struct
break
else
long
switch
case
enum
register
typedef
char
extern
return
union
const
float
short
unsigned
continue
for
signed
void
default
goto
sizeof
volatile
do
if
static
while
Keywords added in C99 standard
_Bool
_Complex
_Imaginary
inline
restrict
Keywords added in C11 draft standard
_Alignas
_Alignof
_Atomic
_Generic
_Noreturn
_Static_assert
_Thread_local
Fig. 2.15 | C’s keywords.
2.7 Secure C Programming
We mentioned The CERT C Secure Coding Standard in the Preface and indicated that we
would follow certain guidelines that will help you avoid programming practices that open
systems to attacks.
Avoid Single-Argument printfs3
One such guideline is to avoid using printf with a single string argument. If you need to display a string that terminates with a newline, use the puts function, which displays its string
argument followed by a newline character. For example, in Fig. 2.1, line 8
printf( "Welcome to C!\n" );
should be written as:
puts( "Welcome to C!" );
We did not include \n in the preceding string because puts adds it automatically.
If you need to display a string without a terminating newline character, use printf
with two arguments—a "%s" format control string and the string to display. The %s conversion specifier is for displaying a string. For example, in Fig. 2.3, line 8
printf( "Welcome " );
should be written as:
printf( "%s", "Welcome " );
3.
For more information, see CERT C Secure Coding rule FIO30-C (www.securecoding.cert.org/
confluence/display/seccode/FIO30-C.+Exclude+user+input+from+format+strings). In Chapter 6’s Secure C Programming section, we’ll explain the notion of user input as referred to by this CERT
guideline.
Summary
59
Although the printfs in this chapter as written are actually not insecure, these changes
are responsible coding practices that will eliminate certain security vulnerabilities as we get
deeper into C—we’ll explain the rationale later in the book. From this point forward, we use
these practices in the chapter examples and you should use them in your exercise solutions.
and printf, scanf_s and printf_s
We introduced scanf and printf in this chapter. We’ll be saying more about these in subsequent Secure C Coding Guidelines sections. We’ll also discuss scanf_s and printf_s,
which were introduced in C11.
scanf
Summary
Section 2.1 Introduction
• The C language facilitates a structured and disciplined approach to computer-program design.
Section 2.2 A Simple C Program: Printing a Line of Text
• Comments begin with //. Comments document programs and improve program readability. C
also supports older-style multi-line comments that begin with /* and end with */.
• Comments do not cause the computer to perform any action when the program is run. They’re
ignored by the C compiler and do not cause any machine-language object code to be generated.
• Lines beginning with # are processed by the preprocessor before the program is compiled. The
#include directive tells the preprocessor to include the contents of another file.
• The <stdio.h> header contains information used by the compiler when compiling calls to standard input/output library functions such as printf.
• The function main is a part of every C program. The parentheses after main indicate that main is
a program building block called a function. C programs contain one or more functions, one of
which must be main. Every program in C begins executing at the function main.
• Functions can return information. The keyword int to the left of main indicates that main “returns” an integer (whole-number) value.
• Functions can receive information when they’re called upon to execute. The void in parentheses
after main indicates that main does not receive any information.
• A left brace, {, begins the body of every function. A corresponding right brace, }, ends each function. This pair of braces and the portion of the program between the braces is called a block.
• The printf function instructs the computer to display information on the screen.
• A string is sometimes called a character string, a message or a literal.
• Every statement must end with a semicolon (also known as the statement terminator).
• In \n, the backslash (\) is called an escape character. When encountering a backslash in a string,
the compiler looks ahead at the next character and combines it with the backslash to form an
escape sequence. The escape sequence \n means newline.
• When a newline appears in the string output by a printf, the newline causes the cursor to position to the beginning of the next line on the screen.
• The double backslash (\\) escape sequence can be used to place a single backslash in a string.
• The escape sequence \" represents a literal double-quote character.
Section 2.3 Another Simple C Program: Adding Two Integers
• A variable is a location in memory where a value can be stored for use by a program.
60
Chapter 2
Introduction to C Programming
• Variables of type int hold integer values, i.e., whole numbers such as 7, –11, 0, 31914.
• All variables must be defined with a name and a data type before they can be used in a program.
• A variable name in C is any valid identifier. An identifier is a series of characters consisting of
letters, digits and underscores ( _ ) that does not begin with a digit.
• C is case sensitive—uppercase and lowercase letters are different in C.
• Micosoft Visual C++ requires variable definitions in C programs to be placed after the left brace
of a function and before any executable statements. GNU gcc and some other compilers do not
have this restriction.
• A syntax error is caused when the compiler cannot recognize a statement. The compiler normally
issues an error message to help you locate and fix the incorrect statement. Syntax errors are violations of the language. Syntax errors are also called compile errors, or compile-time errors.
• Standard Library function scanf can be used to obtain input from the standard input, which is
usually the keyboard.
• The scanf format control string indicates the type(s) of data that should be input.
• The %d conversion specifier indicates that the data should be an integer (the letter d stands for
“decimal integer”). The % in this context is treated by scanf (and printf) as a special character
that begins a conversion specifier.
• The arguments that follow scanf’s format control string begin with an ampersand (&)—called
the address operator in C—followed by a variable name. The ampersand, when combined with
a variable name, tells scanf the location in memory at which the variable is located. The computer then stores the value for the variable at that location.
• Most calculations are performed in assignment statements.
• The = operator and the + operator are binary operators—each has two operands.
• In a printf that specifies a format control string as its first argument the conversion specifiers
indicate placeholders for data to output.
Section 2.4 Memory Concepts
• Variable names correspond to locations in the computer’s memory. Every variable has a name, a
type and a value.
• Whenever a value is placed in a memory location, the value replaces the previous value in that
location; thus, placing a new value into a memory location is said to be destructive.
• When a value is read from a memory location, the process is said to be nondestructive.
Section 2.5 Arithmetic in C
• In algebra, if we want to multiply a times b, we can simply place these single-letter variable names
side by side as in ab. In C, however, if we were to do this, ab would be interpreted as a single,
two-letter name (or identifier). Therefore, C (like other programming languages, in general) requires that multiplication be explicitly denoted by using the * operator, as in a * b.
• Arithmetic expressions in C must be written in straight-line form to facilitate entering programs
into the computer. Thus, expressions such as “a divided by b” must be written as a/b, so that all
operators and operands appear in a straight line.
• Parentheses are used to group terms in C expressions in much the same manner as in algebraic
expressions.
• C evaluates arithmetic expressions in a precise sequence determined by the following rules of operator precedence, which are generally the same as those followed in algebra.
Terminology
61
• Multiplication, division and remainder operations are applied first. If an expression contains several multiplication, division and remainder operations, evaluation proceeds from left to right.
Multiplication, division and remainder are said to be on the same level of precedence.
• Addition and subtraction operations are evaluated next. If an expression contains several addition
and subtraction operations, evaluation proceeds from left to right. Addition and subtraction also
have the same level of precedence, which is lower than the precedence of the multiplication, division and remainder operators.
• The rules of operator precedence specify the order C uses to evaluate expressions. The associativity of the operators specifies whether they evaluate from left to right or from right to left.
Section 2.6 Decision Making: Equality and Relational Operators
• Executable C statements either perform actions or make decisions.
• C’s if statement allows a program to make a decision based on the truth or falsity of a statement
of fact called a condition. If the condition is met (i.e., the condition is true) the statement in the
body of the if statement executes. If the condition isn’t met (i.e., the condition is false) the body
statement does not execute. Whether the body statement is executed or not, after the if statement completes, execution proceeds with the next statement after the if statement.
• Conditions in if statements are formed by using the equality operators and relational operators.
• The relational operators all have the same level of precedence and associate left to right. The
equality operators have a lower level of precedence than the relational operators and they also associate left to right.
• To avoid confusing assignment (=) and equality (==), the assignment operator should be read
“gets” and the equality operator should be read “double equals.”
• In C programs, white-space characters such as tabs, newlines and spaces are normally ignored.
So, statements may be split over several lines. It’s not correct to split identifiers.
• Keywords (or reserved words) have special meaning to the C compiler, so you cannot use them
as identifiers such as variable names.
Section 2.7 Secure C Programming
• One practice to help avoid leaving systems open to attacks is to avoid using printf with a single
string argument.
• To display a string followed by a newline character, use the puts function, which displays it’s
string argument followed by a newline character.
• To display a string without a trailing newline character, you can use printf the format string argument "%s" followed by a second argument representing the string to display. The conversion
specification %s is for displaying a string.
Terminology
addition operator 50
division operator 50
* multiplication operator 50
% remainder operator 50
- subtraction operator 50
%d conversion specifier 47
%s conversion specifier 58
action 43
action/decision model 43
address operator (&) 47
+
/
argument 43
arithmetic operators 50
assignment statement 48
associativity 51
body 42
C preprocessor 42
case sensitive 46
character string 43
comment (//) 41
comment (/*...*/) 42
62
Chapter 2
Introduction to C Programming
condition 54
decision 43
definition 46
destructive 49
document a program 41
embedded parentheses 51
Enter key 47
equality operator 54
escape character 43
escape sequence 43
executable 44
false 54
flow of control 57
format control string 47
function 42
identifier 46
if statement 54
int type 46
integer 46
integer division 50
interactive computing 48
keyword 57
literal 43
message 43
nested parentheses 51
newline (\n) 43
nondestructive 50
operand 48
percent sign (%) 50
printf function 43
prompt 47
puts function 58
redundant parentheses 54
relational operator 54
right brace (}) 42
rules of operator precedence 51
scanf function 47
single-line comment (//) 41
statement 43
statement terminator (;) 43
<stdio.h> (standard input/output) header 42
straight-line form 51
string 43
true 54
type 49
value 49
variable 46
white space 42
Self-Review Exercises
2.1
2.2
Fill in the blanks in each of the following.
.
a) Every C program begins execution at the function
b) Every function’s body begins with
and ends with
.
c) Every statement ends with a(n)
.
standard library function displays information on the screen.
d) The
e) The escape sequence \n represents the
character, which causes the cursor
to position to the beginning of the next line on the screen.
Standard Library function is used to obtain data from the keyboard.
f) The
g) The conversion specifier
is used in a scanf format control string to indicate
that an integer will be input and in a printf format control string to indicate that an
integer will be output.
h) Whenever a new value is placed in a memory location, that value overrides the previous
value in that location. This process is said to be
.
i) When a value is read from a memory location, the value in that location is preserved;
.
this process is said to be
j) The
statement is used to make decisions.
State whether each of the following is true or false. If false, explain why.
a) Function printf always begins printing at the beginning of a new line.
b) Comments cause the computer to display the text after // on the screen when the program is executed.
c) The escape sequence \n when used in a printf format control string causes the cursor
to position to the beginning of the next line on the screen.
d) All variables must be defined before they’re used.
e) All variables must be given a type when they’re defined.
Answers to Self-Review Exercises
63
f) C considers the variables number and NuMbEr to be identical.
g) Definitions can appear anywhere in the body of a function.
h) All arguments following the format control string in a printf function must be preceded by an ampersand (&).
i) The remainder operator (%) can be used only with integer operands.
j) The arithmetic operators *, /, %, + and - all have the same level of precedence.
k) A program that prints three lines of output must contain three printf statements.
2.3
Write a single C statement to accomplish each of the following:
a) Define the variables c, thisVariable, q76354 and number to be of type int.
b) Prompt the user to enter an integer. End your prompting message with a colon (:) followed by a space and leave the cursor positioned after the space.
c) Read an integer from the keyboard and store the value entered in integer variable a.
d) If number is not equal to 7, print "The variable number is not equal to 7."
e) Print the message "This is a C program." on one line.
f) Print the message "This is a C program." on two lines so that the first line ends with C.
g) Print the message "This is a C program." with each word on a separate line.
h) Print the message "This is a C program." with the words separated by tabs.
2.4
Write a statement (or comment) to accomplish each of the following:
a) State that a program will calculate the product of three integers.
b) Define the variables x, y, z and result to be of type int.
c) Prompt the user to enter three integers.
d) Read three integers from the keyboard and store them in the variables x, y and z.
e) Compute the product of the three integers contained in variables x, y and z, and assign
the result to the variable result.
f) Print "The product is" followed by the value of the integer variable result.
2.5
Using the statements you wrote in Exercise 2.4, write a complete program that calculates
the product of three integers.
2.6
Identify and correct the errors in each of the following statements:
a) printf( "The value is %d\n", &number );
b) scanf( "%d%d", &number1, number2 );
c) if ( c < 7 );{
printf( "C is less than 7\n" );
}
d)
if ( c => 7 ) {
printf( "C is greater than or equal to 7\n" );
}
Answers to Self-Review Exercises
2.1
a) main. b) left brace ({), right brace (}). c) semicolon. d) printf. e) newline. f) scanf.
g) %d. h) destructive. i) nondestructive. j) if.
2.2
a) False. Function printf always begins printing where the cursor is positioned,
and this may be anywhere on a line of the screen.
b) False. Comments do not cause any action to be performed when the program is executed. They’re used to document programs and improve their readability.
c) True.
d) True.
e) True.
f) False. C is case sensitive, so these variables are unique.
64
Chapter 2
Introduction to C Programming
g) False. A variable’s definition must appear before its first use in the code. In Microsoft
Visual C++, variable definitions must appear immediately following the left brace that
begins the body of main. Later in the book we’ll discuss this in more depth as we encounter additional C features that can affect this issue.
h) False. Arguments in a printf function ordinarily should not be preceded by an ampersand. Arguments following the format control string in a scanf function ordinarily
should be preceded by an ampersand. We’ll discuss exceptions to these rules in
Chapter 6 and Chapter 7.
i) True.
j) False. The operators *, / and % are on the same level of precedence, and the operators +
and - are on a lower level of precedence.
k) False. A printf statement with multiple \n escape sequences can print several lines.
2.3
a)
b)
c)
d)
int c, thisVariable, q76354, number;
printf( "Enter an integer: " );
scanf( "%d", &a );
if ( number != 7 ) {
printf( "The variable number is not equal to 7.\n" );
}
2.4
2.5
1
2
3
4
5
6
7
8
9
10
11
12
2.6
e)
f)
g)
h)
printf( "This is a C program.\n" );
a)
b)
c)
d)
e)
f)
// Calculate the product of three integers
printf( "This is a C\nprogram.\n" );
printf( "This\nis\na\nC\nprogram.\n" );
printf( "This\tis\ta\tC\tprogram.\n" );
int x, y, z, result;
printf( "Enter three integers: " );
scanf( "%d%d%d", &x, &y, &z );
result = x * y * z;
printf( "The product is %d\n", result );
See below.
// Calculate the product of three integers
#include <stdio.h>
int main( void )
{
int x, y, z, result; // declare variables
printf( "Enter three integers: " ); // prompt
scanf( "%d%d%d", &x, &y, &z ); // read three integers
result = x * y * z; // multiply values
printf( "The product is %d\n", result ); // display result
} // end function main
a) Error: &number.
Correction: Eliminate the &. We discuss exceptions to this later.
b) Error: number2 does not have an ampersand.
Correction: number2 should be &number2. Later in the text we discuss exceptions to this.
c) Error: Semicolon after the right parenthesis of the condition in the if statement.
Correction: Remove the semicolon after the right parenthesis. [Note: The result of this
error is that the printf statement will be executed whether or not the condition in the
Exercises
65
if statement is true. The semicolon after the right parenthesis is considered an empty
statement—a statement that does nothing.]
d) Error: => is not an operator in C.
Correction: The relational operator => should be changed to >= (greater than or equal to).
Exercises
2.7
Identify and correct the errors in each of the following statements. (Note: There may be
more than one error per statement.)
a) scanf( "d", value );
b) printf( "The product of %d and %d is %d"\n, x, y );
c) firstNumber + secondNumber = sumOfNumbers
d) if ( number => largest )
largest == number;
e)
f)
g)
h)
*/ Program to determine the largest of three integers /*
Scanf( "%d", anInteger );
printf( "Remainder of %d divided by %d is\n", x, y, x % y );
if ( x = y );
printf( %d is equal to %d\n", x, y );
i) print( "The sum is %d\n," x + y );
j) Printf( "The value you entered is: %d\n, &value );
2.8
Fill in the blanks in each of the following:
a)
are used to document a program and improve its readability.
b) The function used to display information on the screen is
.
.
c) A C statement that makes a decision is
d) Calculations are normally performed by
statements.
e) The
function inputs values from the keyboard.
2.9
Write a single C statement or line that accomplishes each of the following:
a) Print the message “Enter two numbers.”
b) Assign the product of variables b and c to variable a.
c) State that a program performs a sample payroll calculation (i.e., use text that helps to
document a program).
d) Input three integer values from the keyboard and place them in integer variables a, b
and c.
2.10 State which of the following are true and which are false. If false, explain your answer.
a) C operators are evaluated from left to right.
b) The following are all valid variable names: _under_bar_, m928134, t5, j7, her_sales,
his_account_total, a, b, c, z, z2.
c) The statement printf("a = 5;"); is a typical example of an assignment statement.
d) A valid arithmetic expression containing no parentheses is evaluated from left to right.
e) The following are all invalid variable names: 3g, 87, 67h2, h22, 2h.
2.11 Fill in the blanks in each of the following:
a) What arithmetic operations are on the same level of precedence as multiplication?
.
b) When parentheses are nested, which set of parentheses is evaluated first in an arithmetic
.
expression?
c) A location in the computer’s memory that may contain different values at various times
throughout the execution of a program is called a
.
2.12 What, if anything, prints when each of the following statements is performed? If nothing
prints, then answer “Nothing.” Assume x = 2 and y = 3.
66
Chapter 2
Introduction to C Programming
a) printf( "%d", x );
b) printf( "%d", x + x );
c) printf( "x=" );
d) printf( "x=%d", x );
e) printf( "%d = %d", x + y, y + x );
f) z = x + y;
g) scanf( "%d%d", &x, &y );
h) // printf( "x + y = %d", x + y );
i) printf( "\n" );
2.13 Which, if any, of the following C statements contain variables whose values are replaced?
a) scanf( "%d%d%d%d%d", &b, &c, &d, &e, &f );
b) p = i + j + k + 7;
c) printf( "Values are replaced" );
d) printf( "a = 5" );
2.14 Given the equation y = ax3 + 7, which of the following, if any, are correct C statements for
this equation?
a) y = a * x * x * x + 7;
b) y = a * x * x * ( x + 7 );
c) y = ( a * x ) * x * ( x + 7 );
d) y = ( a * x ) * x * x + 7;
e) y = a * ( x * x * x ) + 7;
f) y = a * x * ( x * x + 7 );
2.15 State the order of evaluation of the operators in each of the following C statements and
show the value of x after each statement is performed.
a) x = 7 + 3 * 6 / 2 - 1;
b) x = 2 % 2 + 2 * 2 - 2 / 2;
c) x = ( 3 * 9 * ( 3 + ( 9 * 3 / ( 3 ) ) ) );
2.16 (Arithmetic) Write a program that asks the user to enter two numbers, obtains them from
the user and prints their sum, product, difference, quotient and remainder.
2.17 (Printing Values with printf) Write a program that prints the numbers 1 to 4 on the same
line. Write the program using the following methods.
a) Using one printf statement with no conversion specifiers.
b) Using one printf statement with four conversion specifiers.
c) Using four printf statements.
2.18 (Comparing Integers) Write a program that asks the user to enter two integers, obtains the
numbers from the user, then prints the larger number followed by the words “is larger.” If the
numbers are equal, print the message “These numbers are equal.” Use only the single-selection
form of the if statement you learned in this chapter.
2.19 (Arithmetic, Largest Value and Smallest Value) Write a program that inputs three different
integers from the keyboard, then prints the sum, the average, the product, the smallest and the largest of these numbers. Use only the single-selection form of the if statement you learned in this chapter. The screen dialogue should appear as follows:
Enter three different integers: 13 27 14
Sum is 54
Average is 18
Product is 4914
Smallest is 13
Largest is 27
Exercises
67
2.20 (Diameter, Circumference and Area of a Circle) Write a program that reads in the radius
of a circle and prints the circle’s diameter, circumference and area. Use the constant value 3.14159
for π. Perform each of these calculations inside the printf statement(s) and use the conversion specifier %f. [Note: In this chapter, we’ve discussed only integer constants and variables. In Chapter 3
we’ll discuss floating-point numbers, i.e., values that can have decimal points.]
2.21 (Shapes with Asterisks) Write a program that prints the following shapes with asterisks.
*********
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*********
2.22
***
*
*
*
*
*
*
*
*
*
*
*
*
*
*
***
*
***
*****
*
*
*
*
*
*
*
* *
*
*
*
*
*
*
*
*
*
*
* *
*
What does the following code print?
printf( "*\n**\n***\n****\n*****\n" );
2.23 (Largest and Smallest Integers) Write a program that reads in three integers and then determines and prints the largest and the smallest integers in the group. Use only the programming techniques you have learned in this chapter.
2.24 (Odd or Even) Write a program that reads an integer and determines and prints whether
it’s odd or even. [Hint: Use the remainder operator. An even number is a multiple of two. Any multiple of two leaves a remainder of zero when divided by 2.]
2.25 Print your initials in block letters down the page. Construct each block letter out of the letter it represents, as shown below.
PPPPPPPPP
P
P
P
P
P
P
P P
JJ
J
J
J
JJJJJJJ
DDDDDDDDD
D
D
D
D
D
D
DDDDD
2.26 (Multiples) Write a program that reads in two integers and determines and prints whether
the first is a multiple of the second. [Hint: Use the remainder operator.]
2.27 (Checkerboard Pattern of Asterisks) Display the following checkerboard pattern with eight
printf statements and then display the same pattern with as few printf statements as possible.
* * * * * * * *
* * * * * * * *
* * * * * * * *
* * * * * * * *
* * * * * * * *
* * * * * * * *
* * * * * * * *
* * * * * * * *
68
Chapter 2
Introduction to C Programming
2.28 Distinguish between the terms fatal error and nonfatal error. Why might you prefer to experience a fatal error rather than a nonfatal error?
2.29 (Integer Value of a Character) Here’s a peek ahead. In this chapter you learned about integers and the type int. C can also represent uppercase letters, lowercase letters and a considerable
variety of special symbols. C uses small integers internally to represent each different character. The
set of characters a computer uses together with the corresponding integer representations for those
characters is called that computer’s character set. You can print the integer equivalent of uppercase
A, for example, by executing the statement
printf( "%d", 'A' );
Write a C program that prints the integer equivalents of some uppercase letters, lowercase letters,
digits and special symbols. As a minimum, determine the integer equivalents of the following:
A B C a b c 0 1 2 $ * + / and the blank character.
2.30 (Separating Digits in an Integer) Write a program that inputs one five-digit number, separates the number into its individual digits and prints the digits separated from one another by three
spaces each. [Hint: Use combinations of integer division and the remainder operation.] For example, if the user types in 42139, the program should print
4
2
1
3
9
2.31 (Table of Squares and Cubes) Using only the techniques you learned in this chapter, write
a program that calculates the squares and cubes of the numbers from 0 to 10 and uses tabs to print
the following table of values:
number
0
1
2
3
4
5
6
7
8
9
10
square
0
1
4
9
16
25
36
49
64
81
100
cube
0
1
8
27
64
125
216
343
512
729
1000
Making a Difference
2.32 (Body Mass Index Calculator) We introduced the body mass index (BMI) calculator in
Exercise 1.14. The formulas for calculating BMI are
weightInPounds × 703
BMI = ----------------------------------------------------------------------------------heightInInches × heightInInches
or
weightInKi log rams
BMI = ------------------------------------------------------------------------------------heightInMeters × heightInMeters
Create a BMI calculator application that reads the user’s weight in pounds and height in inches
(or, if you prefer, the user’s weight in kilograms and height in meters), then calculates and displays
the user’s body mass index. Also, the application should display the following information from
the Department of Health and Human Services/National Institutes of Health so the user can evaluate his/her BMI:
Making a Difference
BMI VALUES
Underweight:
Normal:
Overweight:
Obese:
69
less than 18.5
between 18.5 and 24.9
between 25 and 29.9
30 or greater
[Note: In this chapter, you learned to use the int type to represent whole numbers. The BMI calculations when done with int values will both produce whole-number results. In Chapter 4 you’ll
learn to use the double type to represent numbers with decimal points. When the BMI calculations
are performed with doubles, they’ll both produce numbers with decimal points—these are called
“floating-point” numbers.]
2.33 (Car-Pool Savings Calculator) Research several car-pooling websites. Create an application
that calculates your daily driving cost, so that you can estimate how much money could be saved by
car pooling, which also has other advantages such as reducing carbon emissions and reducing traffic
congestion. The application should input the following information and display the user’s cost per
day of driving to work:
a) Total miles driven per day.
b) Cost per gallon of gasoline.
c) Average miles per gallon.
d) Parking fees per day.
e) Tolls per day.
3
Let’s all move one place on.
—Lewis Carroll
The wheel is come full circle.
—William Shakespeare
All the evolution we know of
proceeds from the vague to the
definite.
—Charles Sanders Peirce
Objectives
In this chapter, you’ll:
■
■
■
■
■
■
■
Use basic problem-solving
techniques.
Develop algorithms through
the process of top-down,
stepwise refinement.
Use the if selection
statement and the if…else
selection statement to select
actions.
Use the while repetition
statement to execute
statements in a program
repeatedly.
Use counter-controlled
repetition and sentinelcontrolled repetition.
Learn structured
programming.
Use increment, decrement
and assignment operators.
Structured Program
Development in C
3.1 Introduction
3.1 Introduction
3.2 Algorithms
3.3 Pseudocode
3.4 Control Structures
3.5 The if Selection Statement
3.6 The if…else Selection Statement
3.7 The while Repetition Statement
3.8 Formulating Algorithms Case Study
1: Counter-Controlled Repetition
71
3.9 Formulating Algorithms with TopDown, Stepwise Refinement Case
Study 2: Sentinel-Controlled
Repetition
3.10 Formulating Algorithms with TopDown, Stepwise Refinement Case
Study 3: Nested Control Statements
3.11 Assignment Operators
3.12 Increment and Decrement Operators
3.13 Secure C Programming
Summary |Terminology | Self-Review Exercises | Answers to Self-Review Exercises | Exercises |
Making a Difference
3.1 Introduction
Before writing a program to solve a particular problem, we must have a thorough understanding of the problem and a carefully planned solution approach. The next two chapters
discuss techniques that facilitate the development of structured computer programs. In
Section 4.12, we present a summary of the structured programming techniques developed
here and in Chapter 4.
3.2 Algorithms
The solution to any computing problem involves executing a series of actions in a specific
order. A procedure for solving a problem in terms of
1. the actions to be executed, and
2. the order in which these actions are to be executed
is called an algorithm. The following example demonstrates that correctly specifying the
order in which the actions are to be executed is important.
Consider the “rise-and-shine algorithm” followed by one junior executive for getting
out of bed and going to work: (1) Get out of bed, (2) take off pajamas, (3) take a shower,
(4) get dressed, (5) eat breakfast, (6) carpool to work. This routine gets the executive to
work well prepared to make critical decisions. Suppose that the same steps are performed
in a slightly different order: (1) Get out of bed, (2) take off pajamas, (3) get dressed, (4)
take a shower, (5) eat breakfast, (6) carpool to work. In this case, our junior executive
shows up for work soaking wet. Specifying the order in which statements are to be executed in a computer program is called program control. In this and the next chapter, we
investigate C’s program control capabilities.
3.3 Pseudocode
Pseudocode is an artificial and informal language that helps you develop algorithms. The
pseudocode we present here is particularly useful for developing algorithms that will be
72
Chapter 3
Structured Program Development in C
converted to structured C programs. Pseudocode is similar to everyday English; it’s convenient and user friendly although it’s not an actual computer programming language.
Pseudocode programs are not executed on computers. Rather, they merely help you
“think out” a program before attempting to write it in a programming language like C.
Pseudocode consists purely of characters, so you may conveniently type pseudocode
programs into a computer using an editor program. A carefully prepared pseudocode program may be easily converted to a corresponding C program. This is done in many cases
simply by replacing pseudocode statements with their C equivalents.
Pseudocode consists only of action statements—those that are executed when the program has been converted from pseudocode to C and is run in C. Definitions are not executable statements—they’re simply messages to the compiler. For example, the definition
int i;
tells the compiler the type of variable i and instructs the compiler to reserve space in memory for the variable. But this definition does not cause any action—such as input, output,
a calculation or a comparison—to occur when the program is executed. Some programmers choose to list each variable and briefly mention the purpose of each at the beginning
of a pseudocode program.
3.4 Control Structures
Normally, statements in a program are executed one after the other in the order in which
they’re written. This is called sequential execution. Various C statements we’ll soon discuss enable you to specify that the next statement to be executed may be other than the
next one in sequence. This is called transfer of control.
During the 1960s, it became clear that the indiscriminate use of transfers of control
was the root of a great deal of difficulty experienced by software development groups. The
finger of blame was pointed at the goto statement that allows you to specify a transfer of
control to one of many possible destinations in a program. The notion of so-called structured programming became almost synonymous with “goto elimination.”
The research of Bohm and Jacopini1 had demonstrated that programs could be
written without any goto statements. The challenge of the era was for programmers to shift
their styles to “goto-less programming.” It was not until well into the 1970s that the programming profession started taking structured programming seriously. The results were
impressive, as software development groups reported reduced development times, more
frequent on-time delivery of systems and more frequent within-budget completion of software projects. Programs produced with structured techniques were clearer, easier to debug
and modify and more likely to be bug free in the first place.
Bohm and Jacopini’s work demonstrated that all programs could be written in terms
of only three control structures, namely the sequence structure, the selection structure
and the repetition structure. The sequence structure is simple—unless directed otherwise,
the computer executes C statements one after the other in the order in which they’re
written. The flowchart segment of Fig. 3.1 illustrates C’s sequence structure.
1.
C. Bohm and G. Jacopini, “Flow Diagrams, Turing Machines, and Languages with Only Two Formation Rules,” Communications of the ACM, Vol. 9, No. 5, May 1966, pp. 336–371.
3.4 Control Structures
73
Flowcharts
A flowchart is a graphical representation of an algorithm or of a portion of an algorithm.
Flowcharts are drawn using certain special-purpose symbols such as rectangles, diamonds,
rounded rectaingles, and small circles; these symbols are connected by arrows called flowlines.
Like pseudocode, flowcharts are useful for developing and representing algorithms,
although pseudocode is preferred by most programmers. Flowcharts clearly show how
control structures operate; that’s what we use them for in this text.
Consider the flowchart for the sequence structure in Fig. 3.1. We use the rectangle
symbol, also called the action symbol, to indicate any type of action including a calculation or an input/output operation. The flowlines in the figure indicate the order in which
the actions are performed—first, grade is added to total, then 1 is added to counter. C
allows us to have as many actions as we want in a sequence structure. As we’ll soon see,
anywhere a single action may be placed, we may place several actions in sequence.
add grade to total
total = total + grade;
add 1 to counter
counter = counter + 1;
Fig. 3.1 | Flowcharting C’s sequence structure.
When drawing a flowchart that represents a complete algorithm, a rounded rectangle
symbol containing the word “Begin” is the first symbol used in the flowchart; a rounded
rectangle symbol containing the word “End” is the last symbol used. When drawing only
a portion of an algorithm as in Fig. 3.1, the rounded rectangle symbols are omitted in favor
of using small circle symbols, also called connector symbols.
Perhaps the most important flowcharting symbol is the diamond symbol, also called
the decision symbol, which indicates that a decision is to be made. We’ll discuss the
diamond symbol in the next section.
Selection Statements in C
C provides three types of selection structures in the form of statements. The if selection
statement (Section 3.5) either selects (performs) an action if a condition is true or skips the
action if the condition is false. The if…else selection statement (Section 3.6) performs
an action if a condition is true and performs a different action if the condition is false. The
switch selection statement (discussed in Chapter 4) performs one of many different actions, depending on the value of an expression. The if statement is called a single-selection statement because it selects or ignores a single action. The if…else statement is
called a double-selection statement because it selects between two different actions. The
switch statement is called a multiple-selection statement because it selects among many
different actions.
74
Chapter 3
Structured Program Development in C
Repetition Statements in C
C provides three types of repetition structures in the form of statements, namely while
(Section 3.7), do…while, and for (both discussed in Chapter 4).
That’s all there is. C has only seven control statements: sequence, three types of
selection and three types of repetition. Each C program is formed by combining as many
of each type of control statement as is appropriate for the algorithm the program implements. As with the sequence structure of Fig. 3.1, we’ll see that the flowchart representation of each control statement has two small circle symbols, one at the entry point to the
control statement and one at the exit point. These single-entry/single-exit control statements make it easy to build clear programs. The control-statement flowchart segments can
be attached to one another by connecting the exit point of one control statement to the
entry point of the next. This is much like the way in which a child stacks building blocks,
so we call this control-statement stacking. We’ll learn that there’s only one other way control statements may be connected—a method called control-statement nesting. Thus, any
C program we’ll ever need to build can be constructed from only seven different types of
control statements combined in only two ways. This is the essence of simplicity.
3.5 The if Selection Statement
Selection statements are used to choose among alternative courses of action. For example,
suppose the passing grade on an exam is 60. The pseudocode statement
If student’s grade is greater than or equal to 60
Print “Passed”
determines whether the condition “student’s grade is greater than or equal to 60” is true
or false. If the condition is true, then “Passed” is printed, and the next pseudocode statement in order is “performed” (remember that pseudocode isn’t a real programming language). If the condition is false, the printing is ignored, and the next pseudocode statement
in order is performed. The second line of this selection structure is indented. Such indentation is optional, but it’s highly recommended, as it helps emphasize the inherent structure of structured programs. The C compiler ignores white-space characters such as
blanks, tabs and newlines used for indentation and vertical spacing.
The preceding pseudocode If statement may be written in C as
if ( grade >= 60 ) {
printf( "Passed\n" );
} // end if
Notice that the C code corresponds closely to the pseudocode (of course you’ll also
need to declare the int variable grade). This is one of the properties of pseudocode that
makes it such a useful program development tool.
The flowchart of Fig. 3.2 illustrates the single-selection if statement. This flowchart
contains what is perhaps the most important flowcharting symbol—the diamond symbol,
also called the decision symbol, which indicates that a decision is to be made. The decision
symbol contains an expression, such as a condition, that can be either true or false. The
decision symbol has two flowlines emerging from it. One indicates the direction to take
when the expression in the symbol is true and the other the direction to take when the
expression is false. Decisions can be based on conditions containing relational or equality
3.6 The if…else Selection Statement
75
operators. In fact, a decision can be based on any expression—if the expression evaluates
to zero, it’s treated as false, and if it evaluates to nonzero, it’s treated as true.
grade >= 60
true
print “Passed”
false
Fig. 3.2 | Flowcharting the single-selection if statement.
The if statement, too, is a single-entry/single-exit statement. We’ll soon learn that the
flowcharts for the remaining control structures can also contain (besides small circle symbols and flowlines) only rectangle symbols to indicate the actions to be performed, and
diamond symbols to indicate decisions to be made. This is the action/decision model of programming we’ve been emphasizing.
We can envision seven bins, each containing only control-statement flowcharts of one
of the seven types. These flowchart segments are empty—nothing is written in the rectangles and nothing in the diamonds. Your task, then, is assembling a program from as many
of each type of control statement as the algorithm demands, combining them in only two
possible ways (stacking or nesting), and then filling in the actions and decisions in a manner
appropriate for the algorithm. We’ll discuss the variety of ways in which actions and decisions may be written.
3.6 The if…else Selection Statement
The if selection statement performs an indicated action only when the condition is true;
otherwise the action is skipped. The if…else selection statement allows you to specify
that different actions are to be performed when the condition is true and when it’s false.
For example, the pseudocode statement
If student’s grade is greater than or equal to 60
Print “Passed”
else
Print “Failed”
prints Passed if the student’s grade is greater than or equal to 60 and Failed if the student’s
grade is less than 60. In either case, after printing occurs, the next pseudocode statement
in sequence is “performed.” The body of the else is also indented.
Good Programming Practice 3.1
Indent both body statements of an if…else statement.
76
Chapter 3
Structured Program Development in C
Good Programming Practice 3.2
If there are several levels of indentation, each level should be indented the same additional
amount of space.
The preceding pseudocode If…else statement may be written in C as
if ( grade >= 60 ) {
puts( "Passed" );
} // end if
else {
puts( "Failed" );
} // end else
The flowchart of Fig. 3.3 illustrates the flow of control in the if…else statement.
Once again, besides small circles and arrows, the only symbols in the flowchart are rectangles (for actions) and a diamond (for a decision).
print “Failed”
false
grade >= 60
true
print “Passed”
Fig. 3.3 | Flowcharting the double-selection if…else statement.
C provides the conditional operator (?:), which is closely related to the if…else
statement. The conditional operator is C’s only ternary operator—it takes three operands.
These together with the conditional operator form a conditional expression. The first
operand is a condition. The second operand is the value for the entire conditional
expression if the condition is true and the operand is the value for the entire conditional
expression if the condition is false. For example, the puts statement
puts( grade >= 60 ? "Passed" : "Failed" );
contains as its second argument a conditional expression that evaluates to the string
"Passed" if the condition grade >= 60 is true and to the string "Failed" if the condition
is false. The puts statement performs in essentially the same way as the preceding
if…else statement.
The second and third operands in a conditional expression can also be actions to be
executed. For example, the conditional expression
grade >= 60 ? puts( "Passed" ) : puts( "Failed" );
is read, “If
grade
puts("Failed").”
is greater than or equal to 60, then puts("Passed"), otherwise
This, too, is comparable to the preceding if…else statement. We’ll
3.6 The if…else Selection Statement
77
see that conditional operators can be used in some places where if…else statements cannot.
Nested if...else Statements
Nested if…else statements test for multiple cases by placing if…else statements inside
if…else statements. For example, the following pseudocode statement will print A for
exam grades greater than or equal to 90, B for grades greater than or equal to 80 (but less
than 90), C for grades greater than or equal to 70 (but less than 80), D for grades greater
than or equal to 60 (but less than 70), and F for all other grades.
If student’s grade is greater than or equal to 90
Print “A”
else
If student’s grade is greater than or equal to 80
Print “B”
else
If student’s grade is greater than or equal to 70
Print “C”
else
If student’s grade is greater than or equal to 60
Print “D”
else
Print “F”
This pseudocode may be written in C as
if ( grade >= 90 ) {
puts( "A" );
} // end if
else {
if ( grade >= 80 ) {
puts("B");
} // end if
else {
if ( grade >= 70 ) {
puts("C");
} // end if
else {
if ( grade >= 60 ) {
puts( "D" );
} // end if
else {
puts( "F" );
} // end else
} // end else
} // end else
} // end else
If the variable grade is greater than or equal to 90, all four conditions will be true, but only
the puts statement after the first test will by executed. After that puts is executed, the else
part of the “outer” if…else statement is skipped.
78
Chapter 3
Structured Program Development in C
You may prefer to write the preceding if statement as
if ( grade >= 90 )
puts( "A" );
} // end if
else if ( grade >=
puts( "B" );
} // end else if
else if ( grade >=
puts( "C" );
} // end else if
else if ( grade >=
puts( "D" );
} // end else if
else {
puts( "F" );
} // end else
{
80 ) {
70 ) {
60 ) {
As far as the C compiler is concerned, both forms are equivalent. The latter form is popular
because it avoids the deep indentation of the code to the right. Such indentation often
leaves little room on a line, forcing lines to be split and decreasing program readability.
The if selection statement expects only one statement in its body—if you have only
one statement in the if’s body, you do not need the enclose it in braces. To include several
statements in the body of an if, you must enclose the set of statements in braces ({ and
}). A set of statements contained within a pair of braces is called a compound statement
or a block.
Software Engineering Observation 3.1
A compound statement can be placed anywhere in a program that a single statement can
be placed.
The following example includes a compound statement in the
statement.
else
part of an
if…else
if ( grade >= 60 )
puts( "Passed."
} // end if
else {
puts( "Failed."
puts( "You must
} // end else
{
);
);
take this course again." );
In this case, if grade is less than 60, the program executes both puts statements in the body
of the else and prints
Failed.
You must take this course again.
The braces surrounding the two statements in the
them, the statement
else
clause are important. Without
puts( "You must take this course again." );
would be outside the body of the else part of the
whether the grade was less than 60.
if
and would execute regardless of
3.7 The while Repetition Statement
79
A syntax error is caught by the compiler. A logic error has its effect at execution time.
A fatal logic error causes a program to fail and terminate prematurely. A nonfatal logic error
allows a program to continue executing but to produce incorrect results.
Just as a compound statement can be placed anywhere a single statement can be
placed, it’s also possible to have no statement at all, i.e., the empty statement. The empty
statement is represented by placing a semicolon (;) where a statement would normally be.
Common Programming Error 3.1
Placing a semicolon after the condition in an if statement as in if ( grade >= 60 ); leads
to a logic error in single-selection if statements and a syntax error in double-selection if
statements.
Error-Prevention Tip 3.1
Typing the beginning and ending braces of compound statements before typing the individual statements within the braces helps avoid omitting one or both of the braces, preventing syntax errors and logic errors (where both braces are indeed required).
3.7 The while Repetition Statement
A repetition statement (also called an iteration statement) allows you to specify that an
action is to be repeated while some condition remains true. The pseudocode statement
While there are more items on my shopping list
Purchase next item and cross it off my list
describes the repetition that occurs during a shopping trip. The condition, “there are more
items on my shopping list” may be true or false. If it’s true, then the action, “Purchase next
item and cross it off my list” is performed. This action will be performed repeatedly while
the condition remains true. The statement(s) contained in the while repetition statement
constitute the body of the while. The while statement body may be a single statement or a
compound statement.
Eventually, the condition will become false (when the last item on the shopping list
has been purchased and crossed off the list). At this point, the repetition terminates, and
the first pseudocode statement after the repetition structure is executed.
Common Programming Error 3.2
Not providing in the body of a while statement an action that eventually causes the condition in the while to become false. Normally, such a repetition structure will never terminate—an error called an “infinite loop.”
Common Programming Error 3.3
Spelling the keyword while with an uppercase W, as in While (remember that C is a casesensitive language).
As an example of a while statement, consider a program segment designed to find the
first power of 3 larger than 100. Suppose the integer variable product has been initialized
to 3. When the following while repetition statement finishes executing, product will contain the desired answer:
80
Chapter 3
Structured Program Development in C
product = 3;
while ( product <= 100 ) {
product = 3 * product;
} // end while
The flowchart of Fig. 3.4 illustrates the flow of control in the while repetition statement. Once again, note that (besides small circles and arrows) the flowchart contains only
a rectangle symbol and a diamond symbol. The flowchart clearly shows the repetition. The
flowline emerging from the rectangle wraps back to the decision, which is tested each time
through the loop until the decision eventually becomes false. At this point, the while statement is exited and control passes to the next statement in the program.
product <= 100
true
product = 3 * product
false
Fig. 3.4 | Flowcharting the while repetition statement.
When the while statement is entered, the value of product is 3. The variable product
is repeatedly multiplied by 3, taking on the values 9, 27 and 81 successively. When
product becomes 243, the condition in the while statement, product <= 100, becomes
false. This terminates the repetition, and the final value of product is 243. Program execution continues with the next statement after the while.
3.8 Formulating Algorithms Case Study 1: CounterControlled Repetition
To illustrate how algorithms are developed, we solve several variations of a class-averaging
problem. Consider the following problem statement:
A class of ten students took a quiz. The grades (integers in the range 0 to 100) for this
quiz are available to you. Determine the class average on the quiz.
The class average is equal to the sum of the grades divided by the number of students. The
algorithm for solving this problem on a computer must input each of the grades, perform
the averaging calculation, and print the result.
Let’s use pseudocode to list the actions to execute and specify the order in which these
actions should execute. We use counter-controlled repetition to input the grades one at a
time. This technique uses a variable called a counter to specify the number of times a set
of statements should execute. In this example, repetition terminates when the counter
exceeds 10. In this section we simply present the pseudocode algorithm (Fig. 3.5) and the
3.8 Counter-Controlled Repetition
81
corresponding C program (Fig. 3.6). In the next section we show how pseudocode algorithms are developed. Counter-controlled repetition is often called definite repetition
because the number of repetitions is known before the loop begins executing.
1
2
3
4
5
6
7
8
9
10
Set total to zero
Set grade counter to one
While grade counter is less than or equal to ten
Input the next grade
Add the grade into the total
Add one to the grade counter
Set the class average to the total divided by ten
Print the class average
Fig. 3.5 | Pseudocode algorithm that uses counter-controlled repetition to solve the class-average
problem.
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
// Fig. 3.6: fig03_06.c
// Class average program with counter-controlled repetition.
#include <stdio.h>
// function main begins program execution
int main( void )
{
unsigned int counter; // number of grade to be entered next
int grade; // grade value
int total; // sum of grades entered by user
int average; // average of grades
// initialization phase
total = 0; // initialize total
counter = 1; // initialize loop counter
// processing phase
while ( counter <= 10 ) { // loop 10 times
printf( "%s", "Enter grade: " ); // prompt for input
scanf( "%d", &grade ); // read grade from user
total = total + grade; // add grade to total
counter = counter + 1; // increment counter
} // end while
// termination phase
average = total / 10; // integer division
printf( "Class average is %d\n", average ); // display result
} // end function main
Fig. 3.6 | Class-average problem with counter-controlled repetition. (Part 1 of 2.)
82
Enter
Enter
Enter
Enter
Enter
Enter
Enter
Enter
Enter
Enter
Class
Chapter 3
Structured Program Development in C
grade: 98
grade: 76
grade: 71
grade: 87
grade: 83
grade: 90
grade: 57
grade: 79
grade: 82
grade: 94
average is 81
Fig. 3.6 | Class-average problem with counter-controlled repetition. (Part 2 of 2.)
The algorithm mentions a total and a counter. A total is a variable used to accumulate
the sum of a series of values. A counter is a variable (line 8) used to count—in this case, to
count the number of grades entered. Because the counter variable is used to count from 1
to 10 in this program (all positive values), we declared the variable as an unsigned int,
which can store only non-negative values (that is, 0 and higher). Variables used to store
totals should normally be initialized to zero before being used in a program; otherwise the
sum would include the previous value stored in the total’s memory location. Counter variables are normally initialized to zero or one, depending on their use (we’ll present examples
of each). An uninitialized variable contains a “garbage” value—the value last stored in the
memory location reserved for that variable.
Common Programming Error 3.4
If a counter or total isn’t initialized, the results of your program will probably be incorrect.
This is an example of a logic error.
Error-Prevention Tip 3.2
Initialize all counters and totals.
The averaging calculation in the program produced an integer result of 81. Actually,
the sum of the grades in this example is 817, which when divided by 10 should yield 81.7,
i.e., a number with a decimal point. We’ll see how to deal with such numbers (called
floating-point numbers) in the next section.
3.9 Formulating Algorithms with Top-Down, Stepwise
Refinement Case Study 2: Sentinel-Controlled
Repetition
Let’s generalize the class-average problem. Consider the following problem:
Develop a class-averaging program that will process an arbitrary number of grades
each time the program is run.
In the first class-average example, the number of grades (10) was known in advance. In this
example, no indication is given of how many grades are to be entered. The program must
3.9 Sentinel-Controlled Repetition
83
process an arbitrary number of grades. How can the program determine when to stop the
input of grades? How will it know when to calculate and print the class average?
One way to solve this problem is to use a special value called a sentinel value (also
called a signal value, a dummy value, or a flag value) to indicate “end of data entry.” The
user types in grades until all legitimate grades have been entered. The user then types the
sentinel value to indicate “the last grade has been entered.” Sentinel-controlled repetition
is often called indefinite repetition because the number of repetitions isn’t known before
the loop begins executing.
Clearly, the sentinel value must be chosen so that it cannot be confused with an
acceptable input value. Because grades on a quiz are normally nonnegative integers, –1 is
an acceptable sentinel value for this problem. Thus, a run of the class-average program
might process a stream of inputs such as 95, 96, 75, 74, 89 and –1. The program would
then compute and print the class average for the grades 95, 96, 75, 74, and 89 (–1 is the
sentinel value, so it should not enter into the averaging calculation).
Top-Down, Stepwise Refinement
We approach the class-average program with a technique called top-down, stepwise refinement, a technique that’s essential to the development of well-structured programs. We
begin with a pseudocode representation of the top:
Determine the class average for the quiz
The top is a single statement that conveys the program’s overall function. As such, the top
is, in effect, a complete representation of a program. Unfortunately, the top rarely conveys
a sufficient amount of detail for writing the C program. So we now begin the refinement
process. We divide the top into a series of smaller tasks and list these in the order in which
they need to be performed. This results in the following first refinement.
Initialize variables
Input, sum, and count the quiz grades
Calculate and print the class average
Here, only the sequence structure has been used—the steps listed are to be executed in order, one after the other.
Software Engineering Observation 3.2
Each refinement, as well as the top itself, is a complete specification of the algorithm; only
the level of detail varies.
Second Refinement
To proceed to the next level of refinement, i.e., the second refinement, we commit to specific variables. We need a running total of the numbers, a count of how many numbers
have been processed, a variable to receive the value of each grade as it’s input and a variable
to hold the calculated average. The pseudocode statement
Initialize variables
may be refined as follows:
Initialize total to zero
Initialize counter to zero
84
Chapter 3
Structured Program Development in C
Notice that only the total and counter need to be initialized; the variables average and
grade (for the calculated average and the user input, respectively) need not be initialized
because their values will be written over by the process of destructive read-in discussed in
Chapter 2. The pseudocode statement
Input, sum, and count the quiz grades
requires a repetition structure that successively inputs each grade. Because we do not know
in advance how many grades are to be processed, we’ll use sentinel-controlled repetition.
The user will enter legitimate grades one at a time. After the last legitimate grade is typed,
the user will type the sentinel value. The program will test for this value after each grade
is input and will terminate the loop when the sentinel is entered. The refinement of the
preceding pseudocode statement is then
Input the first grade
While the user has not as yet entered the sentinel
Add this grade into the running total
Add one to the grade counter
Input the next grade (possibly the sentinel)
Notice that in pseudocode, we do not use braces around the set of statements that
form the body of the while statement. We simply indent all these statements under the
while to show that they all belong to the while. Again, pseudocode is an informal program
development aid.
The pseudocode statement
Calculate and print the class average
may be refined as follows:
If the counter is not equal to zero
Set the average to the total divided by the counter
Print the average
else
Print “No grades were entered”
Notice that we’re being careful here to test for the possibility of division by zero—a fatal
error that if undetected would cause the program to fail (often called “crashing”). The
complete second refinement is shown in Fig. 3.7.
Common Programming Error 3.5
An attempt to divide by zero causes a fatal error.
Good Programming Practice 3.3
When performing division by an expression whose value could be zero, explicitly test for
this case and handle it appropriately in your program (such as printing an error message)
rather than allowing the fatal error to occur.
In Fig. 3.5 and Fig. 3.7, we include some completely blank lines in the pseudocode
for readability. Actually, the blank lines separate these programs into their various phases.
3.9 Sentinel-Controlled Repetition
1
2
3
4
5
6
7
8
9
10
11
12
13
14
85
Initialize total to zero
Initialize counter to zero
Input the first grade
While the user has not as yet entered the sentinel
Add this grade into the running total
Add one to the grade counter
Input the next grade (possibly the sentinel)
If the counter is not equal to zero
Set the average to the total divided by the counter
Print the average
else
Print “No grades were entered”
Fig. 3.7 | Pseudocode algorithm that uses sentinel-controlled repetition to solve the classaverage problem.
Software Engineering Observation 3.3
Many programs can be divided logically into three phases: an initialization phase that
initializes the program variables; a processing phase that inputs data values and adjusts
program variables accordingly; and a termination phase that calculates and prints the
final results.
The pseudocode algorithm in Fig. 3.7 solves the more general class-averaging
problem. This algorithm was developed after only two levels of refinement. Sometimes
more levels are necessary.
Software Engineering Observation 3.4
You terminate the top-down, stepwise refinement process when the pseudocode algorithm
is specified in sufficient detail for you to be able to convert the pseudocode to C.
Implementing the C program is then normally straightforward.
The C program and a sample execution are shown in Fig. 3.8. Although only integer
grades are entered, the averaging calculation is likely to produce a number with a decimal
point. The type int cannot represent such a number. The program introduces the data
type float to handle numbers with decimal points (called floating-point numbers) and
introduces a special operator called a cast operator to handle the averaging calculation.
These features are explained after the program is presented.
1
2
3
4
// Fig. 3.8: fig03_08.c
// Class-average program with sentinel-controlled repetition.
#include <stdio.h>
Fig. 3.8 | Class-average program with sentinel-controlled repetition. (Part 1 of 3.)
86
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
Chapter 3
Structured Program Development in C
// function main begins program execution
int main( void )
{
unsigned int counter; // number of grades entered
int grade; // grade value
int total; // sum of grades
float average; // number with decimal point for average
// initialization phase
total = 0; // initialize total
counter = 0; // initialize loop counter
// processing phase
// get first grade from user
printf( "%s", "Enter grade, -1 to end: " ); // prompt for input
scanf( "%d", &grade ); // read grade from user
// loop while sentinel value not yet read from user
while ( grade != -1 ) {
total = total + grade; // add grade to total
counter = counter + 1; // increment counter
// get next grade from user
printf( "%s", "Enter grade, -1 to end: " ); // prompt for input
scanf("%d", &grade); // read next grade
} // end while
// termination phase
// if user entered at least one grade
if ( counter != 0 ) {
// calculate average of all grades entered
average = ( float ) total / counter; // avoid truncation
// display average with two digits of precision
printf( "Class average is %.2f\n", average );
} // end if
else { // if no grades were entered, output message
puts( "No grades were entered" );
} // end else
} // end function main
Enter
Enter
Enter
Enter
Enter
Enter
Enter
Enter
Enter
Class
grade, -1 to end:
grade, -1 to end:
grade, -1 to end:
grade, -1 to end:
grade, -1 to end:
grade, -1 to end:
grade, -1 to end:
grade, -1 to end:
grade, -1 to end:
average is 82.50
75
94
97
88
70
64
83
89
-1
Fig. 3.8 | Class-average program with sentinel-controlled repetition. (Part 2 of 3.)
3.9 Sentinel-Controlled Repetition
87
Enter grade, -1 to end: -1
No grades were entered
Fig. 3.8 | Class-average program with sentinel-controlled repetition. (Part 3 of 3.)
Notice the compound statement in the while loop (line 24) in Fig. 3.8. Once again,
the braces are necessary to ensure that all four statements are executed within the loop.
Without the braces, the last three statements in the body of the loop would fall outside the
loop, causing the computer to interpret this code incorrectly as follows.
while ( grade != -1 )
total = total + grade; // add grade to total
counter = counter + 1; // increment counter
printf( "%s", "Enter grade, -1 to end: " ); // prompt for input
scanf( "%d", &grade ); // read next grade
This would cause an infinite loop if the user did not input -1 for the first grade.
Good Programming Practice 3.4
In a sentinel-controlled loop, the prompts requesting data entry should explicitly remind
the user what the sentinel value is.
Converting Between Types Explicitly and Implicitly
Averages do not always evaluate to integer values. Often, an average is a value such as 7.2
or –93.5 that contains a fractional part. These values are referred to as floating-point numbers and can be represented by the data type float. The variable average is defined to be
of type float (line 12) to capture the fractional result of our calculation. However, the
result of the calculation total / counter is an integer because total and counter are both
integer variables. Dividing two integers results in integer division in which any fractional
part of the calculation is truncated (i.e., lost). Because the calculation is performed first,
the fractional part is lost before the result is assigned to average. To produce a floatingpoint calculation with integer values, we must create temporary values that are floatingpoint numbers. C provides the unary cast operator to accomplish this task. Line 38
average = ( float ) total / counter;
includes the cast operator (float), which creates a temporary floating-point copy of its operand, total. The value stored in total is still an integer. Using a cast operator in this
manner is called explicit conversion. The calculation now consists of a floating-point value (the temporary float version of total) divided by the unsigned int value stored in
counter. C evaluates arithmetic expressions only in which the data types of the operands
are identical. To ensure that the operands are of the same type, the compiler performs an
operation called implicit conversion on selected operands. For example, in an expression
containing the data types unsigned int and float, copies of unsigned int operands are
made and converted to float. In our example, after a copy of counter is made and converted to float, the calculation is performed and the result of the floating-point division
is assigned to average. C provides a set of rules for convertion of operands of different
types. We discuss this further in Chapter 5.
88
Chapter 3
Structured Program Development in C
Cast operators are available for most data types—they’re formed by placing parentheses around a type name. Each cast operator is a unary operator, i.e., an operator that
takes only one operand. In Chapter 2, we studied the binary arithmetic operators. C also
supports unary versions of the plus (+) and minus (-) operators, so you can write expressions such as -7 or +5. Cast operators associate from right to left and have the same precedence as other unary operators such as unary + and unary -. This precedence is one level
higher than that of the multiplicative operators *, / and %.
Formatting Floating-Point Numbers
Figure 3.8 uses the printf conversion specifier %.2f (line 41) to print the value of average. The f specifies that a floating-point value will be printed. The .2 is the precision with
which the value will be displayed—with 2 digits to the right of the decimal point. If the
%f conversion specifier is used (without specifying the precision), the default precision of
6 is used—exactly as if the conversion specifier %.6f had been used. When floating-point
values are printed with precision, the printed value is rounded to the indicated number of
decimal positions. The value in memory is unaltered. When the following statements are
executed, the values 3.45 and 3.4 are printed.
printf( "%.2f\n", 3.446 ); // prints 3.45
printf( "%.1f\n", 3.446 ); // prints 3.4
Common Programming Error 3.6
Using precision in a conversion specification in the format control string of a scanf statement is wrong. Precisions are used only in printf conversion specifications.
Notes on Floating-Point Numbers
Although floating-point numbers are not always “100% precise,” they have numerous applications. For example, when we speak of a “normal” body temperature of 98.6, we do
not need to be precise to a large number of digits. When we view the temperature on a
thermometer and read it as 98.6, it may actually be 98.5999473210643. The point here
is that calling this number simply 98.6 is fine for most applications. We’ll say more about
this issue later.
Another way floating-point numbers develop is through division. When we divide 10
by 3, the result is 3.3333333… with the sequence of 3s repeating infinitely. The computer
allocates only a fixed amount of space to hold such a value, so the stored floating-point
value can be only an approximation.
Common Programming Error 3.7
Using floating-point numbers in a manner that assumes they’re represented precisely can
lead to incorrect results. Floating-point numbers are represented only approximately by
most computers.
Error-Prevention Tip 3.3
Do not compare floating-point values for equality.
3.10 Nested Control Statements
89
3.10 Formulating Algorithms with Top-Down, Stepwise
Refinement Case Study 3: Nested Control Statements
Let’s work another complete problem. We’ll once again formulate the algorithm using
pseudocode and top-down, stepwise refinement, and write a corresponding C program.
We’ve seen that control statements may be stacked on top of one another (in sequence) just
as a child stacks building blocks. In this case study we’ll see the only other structured way
control statements may be connected in C, namely through nesting of one control statement within another.
Consider the following problem statement:
A college offers a course that prepares students for the state licensing exam for real
estate brokers. Last year, 10 of the students who completed this course took the licensing examination. Naturally, the college wants to know how well its students did on the
exam. You’ve been asked to write a program to summarize the results. You’ve been
given a list of these 10 students. Next to each name a 1 is written if the student passed
the exam and a 2 if the student failed.
Your program should analyze the results of the exam as follows:
1. Input each test result (i.e., a 1 or a 2). Display the prompting message “Enter result” each
time the program requests another test result.
2. Count the number of test results of each type.
3. Display a summary of the test results indicating the number of students who passed and
the number who failed.
4. If more than eight students passed the exam, print the message “Bonus to instructor!”
After reading the problem statement carefully, we make the following observations:
1. The program must process 10 test results. A counter-controlled loop will be used.
2. Each test result is a number—either a 1 or a 2. Each time the program reads a test
result, the program must determine whether the number is a 1 or a 2. We test for
a 1 in our algorithm. If the number is not a 1, we assume that it’s a 2. (An exercise
at the end of the chapter considers the consequences of this assumption.)
3. Two counters are used—one to count the number of students who passed the
exam and one to count the number of students who failed the exam.
4. After the program has processed all the results, it must decide whether more than
8 students passed the exam.
Let’s proceed with top-down, stepwise refinement. We begin with a pseudocode representation of the top:
Analyze exam results and decide whether instructor should receive a bonus
Once again, it’s important to emphasize that the top is a complete representation of the
program, but several refinements are likely to be needed before the pseudocode can be naturally evolved into a C program. Our first refinement is
Initialize variables
Input the ten quiz grades and count passes and failures
Print a summary of the exam results and decide whether instructor should receive a bonus
90
Chapter 3
Structured Program Development in C
Here, too, even though we have a complete representation of the entire program, further
refinement is necessary. We now commit to specific variables. Counters are needed to record the passes and failures, a counter will be used to control the looping process, and a
variable is needed to store the user input. The pseudocode statement
Initialize variables
may be refined as follows:
Initialize passes to zero
Initialize failures to zero
Initialize student to one
Notice that only the counters and totals are initialized. The pseudocode statement
Input the ten quiz grades and count passes and failures
requires a loop that successively inputs the result of each exam. Here it’s known in advance
that there are precisely ten exam results, so counter-controlled looping is appropriate. Inside the loop (i.e., nested within the loop) a double-selection statement will determine
whether each exam result is a pass or a failure and will increment the appropriate counters
accordingly. The refinement of the preceding pseudocode statement is then
While student counter is less than or equal to ten
Input the next exam result
If the student passed
Add one to passes
else
Add one to failures
Add one to student counter
Notice the use of blank lines to set off the If…else to improve program readability.
The pseudocode statement
Print a summary of the exam results and decide whether instructor should receive a
bonus
may be refined as follows:
Print the number of passes4
Print the number of failures
If more than eight students passed
Print “Bonus to instructor!”
The complete second refinement appears in Fig. 3.9. We use blank lines to set off the
while statement for program readability.
This pseudocode is now sufficiently refined for conversion to C. The C program and
two sample executions are shown in Fig. 3.10. We’ve taken advantage of a feature of C
that allows initialization to be incorporated into definitions (lines 9–11). Such initialization occurs at compile time. Also, notice that when you output an unsigned int you use
the %u conversion specifier (lines 33–34).
3.10 Nested Control Statements
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
Initialize passes to zero
Initialize failures to zero
Initialize student to one
While student counter is less than or equal to ten
Input the next exam result
If the student passed
Add one to passes
else
Add one to failures
Add one to student counter
Print the number of passes
Print the number of failures
If more than eight students passed
Print “Bonus to instructor!”
Fig. 3.9 | Pseudocode for examination-results problem.
Software Engineering Observation 3.5
Experience has shown that the most difficult part of solving a problem on a computer is
developing the algorithm for the solution. Once a correct algorithm has been specified, the
process of producing a working C program is normally straightforward.
Software Engineering Observation 3.6
Many programmers write programs without ever using program development tools such
as pseudocode. They feel that their ultimate goal is to solve the problem on a computer and
that writing pseudocode merely delays the production of final outputs.
1
2
3
4
5
6
7
8
9
10
11
12
13
// Fig. 3.10: fig03_10.c
// Analysis of examination results.
#include <stdio.h>
// function main begins program execution
int main( void )
{
// initialize variables in definitions
unsigned int passes = 0; // number of passes
unsigned int failures = 0; // number of failures
unsigned int student = 1; // student counter
int result; // one exam result
Fig. 3.10 | Analysis of examination results. (Part 1 of 2.)
91
92
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
Chapter 3
Structured Program Development in C
// process 10 students using counter-controlled loop
while ( student <= 10 ) {
// prompt user for input and obtain value from user
printf( "%s", "Enter result ( 1=pass,2=fail ): " );
scanf( "%d", &result );
// if result 1, increment passes
if ( result == 1 ) {
passes = passes + 1;
} // end if
else { // otherwise, increment failures
failures = failures + 1;
} // end else
student = student + 1; // increment student counter
} // end while
// termination phase; display number of passes and failures
printf( "Passed %u\n", passes );
printf( "Failed %u\n", failures );
// if more than eight students passed, print "Bonus to instructor!"
if ( passes > 8 ) {
puts( "Bonus to instructor!" );
} // end if
} // end function main
Enter Result
Enter Result
Enter Result
Enter Result
Enter Result
Enter Result
Enter Result
Enter Result
Enter Result
Enter Result
Passed 6
Failed 4
(1=pass,2=fail):
(1=pass,2=fail):
(1=pass,2=fail):
(1=pass,2=fail):
(1=pass,2=fail):
(1=pass,2=fail):
(1=pass,2=fail):
(1=pass,2=fail):
(1=pass,2=fail):
(1=pass,2=fail):
1
2
2
1
1
1
2
1
1
2
Enter Result (1=pass,2=fail):
Enter Result (1=pass,2=fail):
Enter Result (1=pass,2=fail):
Enter Result (1=pass,2=fail):
Enter Result (1=pass,2=fail):
Enter Result (1=pass,2=fail):
Enter Result (1=pass,2=fail):
Enter Result (1=pass,2=fail):
Enter Result (1=pass,2=fail):
Enter Result (1=pass,2=fail):
Passed 9
Failed 1
Bonus to instructor!
1
1
1
2
1
1
1
1
1
1
Fig. 3.10 | Analysis of examination results. (Part 2 of 2.)
3.11 Assignment Operators
93
3.11 Assignment Operators
C provides several assignment operators for abbreviating assignment expressions. For example, the statement
c = c + 3;
can be abbreviated with the addition assignment operator += as
c += 3;
The += operator adds the value of the expression on the right of the operator to the value
of the variable on the left of the operator and stores the result in the variable on the left of
the operator. Any statement of the form
variable = variable operator expression;
where operator is one of the binary operators +, -, *,
Chapter 10), can be written in the form
/
or
%
(or others we’ll discuss in
variable operator = expression;
Thus the assignment c += 3 adds 3 to c. Figure 3.11 shows the arithmetic assignment
operators, sample expressions using these operators and explanations.
Assignment operator
Assume: int
Sample expression
Explanation
Assigns
c = 3, d = 5, e = 4, f = 6, g = 12;
to c
to d
20 to e
2 to f
3 to g
+=
c += 7
c = c + 7
10
-=
d -= 4
d = d - 4
1
*=
e *= 5
e = e * 5
/=
f /= 3
f = f / 3
%=
g %= 9
g = g % 9
Fig. 3.11 | Arithmetic assignment operators.
3.12 Increment and Decrement Operators
C also provides the unary increment operator, ++, and the unary decrement operator, --,
which are summarized in Fig. 3.12. If a variable c is to be incremented by 1, the increment
operator ++ can be used rather than the expressions c = c + 1 or c += 1. If increment or decrement operators are placed before a variable (i.e., prefixed), they’re referred to as the preincrement or predecrement operators, respectively. If increment or decrement operators are
placed after a variable (i.e., postfixed), they’re referred to as the postincrement or postdecrement operators, respectively. Preincrementing (predecrementing) a variable causes the
variable to be incremented (decremented) by 1, then its new value is used in the expression
in which it appears. Postincrementing (postdecrementing) the variable causes the current
value of the variable to be used in the expression in which it appears, then the variable value
is incremented (decremented) by 1.
94
Chapter 3
Structured Program Development in C
Operator
Sample expression
Explanation
++
++a
++
a++
--
--b
--
b--
Increment a by 1, then use the new value of
a in the expression in which a resides.
Use the current value of a in the expression
in which a resides, then increment a by 1.
Decrement b by 1, then use the new value
of b in the expression in which b resides.
Use the current value of b in the expression
in which b resides, then decrement b by 1.
Fig. 3.12 | Increment and decrement operators
Figure 3.13 demonstrates the difference between the preincrementing and the postincrementing versions of the ++ operator. Postincrementing the variable c causes it to be
incremented after it’s used in the printf statement. Preincrementing the variable c causes
it to be incremented before it’s used in the printf statement.
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
// Fig. 3.13: fig03_13.c
// Preincrementing and postincrementing.
#include <stdio.h>
// function main begins program execution
int main( void )
{
int c; // define variable
// demonstrate postincrement
c = 5; // assign 5 to c
printf( "%d\n", c ); // print 5
printf( "%d\n", c++ ); // print 5 then postincrement
printf( "%d\n\n", c ); // print 6
// demonstrate preincrement
c = 5; // assign 5 to c
printf( "%d\n", c ); // print 5
printf( "%d\n", ++c ); // preincrement then print 6
printf( "%d\n", c ); // print 6
} // end function main
5
5
6
5
6
6
Fig. 3.13
| Preincrementing and postincrementing.
3.12 Increment and Decrement Operators
95
The program displays the value of c before and after the ++ operator is used. The decrement operator (--) works similarly.
Good Programming Practice 3.5
Unary operators should be placed directly next to their operands with no intervening spaces.
The three assignment statements in Fig. 3.10
passes = passes + 1;
failures = failures + 1;
student = student + 1;
can be written more concisely with assignment operators as
passes += 1;
failures += 1;
student += 1;
with preincrement operators as
++passes;
++failures;
++student;
or with postincrement operators as
passes++;
failures++;
student++;
It’s important to note here that when incrementing or decrementing a variable in a
statement by itself, the preincrement and postincrement forms have the same effect. It’s
only when a variable appears in the context of a larger expression that preincrementing and
postincrementing have different effects (and similarly for predecrementing and postdecrementing). Of the expressions we’ve studied thus far, only a simple variable name may
be used as the operand of an increment or decrement operator.
Common Programming Error 3.8
Attempting to use the increment or decrement operator on an expression other than a simple variable name is a syntax error, e.g., writing ++(x + 1).
Error-Prevention Tip 3.4
C generally does not specify the order in which an operator’s operands will be evaluated
(although we’ll see exceptions to this for a few operators in Chapter 4). Therefore you
should use increment or decrement operators only in statements in which one variable is
incremented or decremented by itself.
Figure 3.14 lists the precedence and associativity of the operators introduced to this
point. The operators are shown top to bottom in decreasing order of precedence. The
second column indicates the associativity of the operators at each level of precedence.
Notice that the conditional operator (?:), the unary operators increment (++), decrement
(--), plus (+), minus (-) and casts, and the assignment operators =, +=, -=, *=, /= and %=
associate from right to left. The third column names the various groups of operators. All
other operators in Fig. 3.14 associate from left to right.
96
Chapter 3
Structured Program Development in C
Operators
++
+
*
(postfix) -- (postfix)
(type) ++ (prefix)
/
+
-
<
<=
==
!=
--
(prefix)
%
>
>=
-=
*=
?:
=
+=
/=
%=
Associativity
Type
right to left
right to left
left to right
left to right
left to right
left to right
right to left
right to left
postfix
unary
multiplicative
additive
relational
equality
conditional
assignment
Fig. 3.14 | Precedence and associativity of the operators encountered so far in the text.
3.13 Secure C Programming
Arithmetic Overflow
Figure 2.5 presented an addition program which calculated the sum of two int values (line
18) with the statement
sum = integer1 + integer2; // assign total to sum
Even this simple statement has a potential problem—adding the integers could result in a
value that’s too large to store in an int variable. This is known as arithmetic overflow and
can cause undefined behavior, possibly leaving a system open to attack.
The maximum and minimum values that can be stored in an int variable are represented by the constants INT_MAX and INT_MIN, respectively, which are defined in the header
<limits.h>. There are similar constants for the other integral types that we’ll be introducing
in Chapter 4. You can see your platform’s values for these constants by opening the header
<limits.h> in a text editor.
It’s considered a good practice to ensure that before you perform arithmetic calculations like the one in line 18 of Fig. 2.5, they will not overflow. The code for doing this is
shown on the CERT website www.securecoding.cert.org—just search for guideline
“INT32-C.” The code uses the && (logical AND) and || (logical OR) operators, which are
introduced in the Chapter 4. In industrial-strength code, you should perform checks like
these for all calculations. In later chapters, we’ll show other programming techniques for
handling such errors.
Unsigned Integers
In Fig. 3.6, line 8 declared as an unsigned int the variable counter because it’s used to
count only non-negative values. In general, counters that should store only non-negative
values should be declared with unsigned before the integer type. Variables of unsigned
types can represent values from 0 to approximately twice the positive range of the corresponding signed integer types. You can determine your platform’s maximum unsigned
int value with the constant UINT_MAX from <limits.h>.
The class-averaging program in Fig. 3.6 could have declared as unsigned int the variables grade, total and average. Grades are normally values from 0 to 100, so the total
3.13 Secure C Programming
97
and average should each be greater than or equal to 0. We declared those variables as ints
because we can’t control what the user actually enters—the user could enter negative
values. Worse yet, the user could enter a value that’s not even a number. (We’ll show how
to deal with such inputs later in the book.)
Sometimes sentinel-controlled loops use invalid values to terminate a loop. For
example, the class-averaging program of Fig. 3.8 terminates the loop when the user enters
the sentinel -1 (an invalid grade), so it would be improper to declare variable grade as an
unsigned int. As you’ll see, the end-of-file (EOF) indicator—which is introduced in the
next chapter and is often used to terminate sentinel-controlled loops—is also a negative
number. For more information, see Chapter 5, “Integer Security” of Robert Seacord’s
book Secure Coding in C and C++.
and printf_s
The C11 standard’s Annex K introduces more secure versions of printf and scanf called
printf_s and scanf_s. Annex K is designated as optional, so not every C vendor will implement it.
Microsoft implemented its own versions of printf_s and scanf_s prior to the publication of the C11 standard and immediately began issuing warnings for every scanf call.
The warnings say that scanf is deprecated—it should no longer be used—and that you
should consider using scanf_s instead.
Many organizations have coding standards that require code to compile without
warning messages. There are two ways to eliminate Visual C++’s scanf warnings—you can
use scanf_s instead of scanf or you can disable these warnings. For the input statements
we’ve used so far, Visual C++ users can simply replace scanf with scanf_s. You can disable the warning messages in Visual C++ as follows:
scanf_s
1. Type Alt F7 to display the Property Pages dialog for your project.
2. In the left column, expand Configuration Properties > C/C++ and select Preprocessor.
3. In the right column, at the end of the value for Preprocessor Definitions, insert
;_CRT_SECURE_NO_WARNINGS
4. Click OK to save the changes.
You’ll no longer receive warnings on scanf (or any other functions that Microsoft has deprecated for similar reasons). For industrial-strength coding, disabling the warnings is discouraged. We’ll say more about how to use scanf_s and printf_s in a later Secure C
Coding Guidelines section.
Summary
Section 3.1 Introduction
• Before writing a program to solve a particular problem, you must have a thorough understanding
of the problem and a carefully planned approach to solving the problem.
Section 3.2 Algorithms
• The solution to any computing problem involves executing a series of actions in a specific order.
98
Chapter 3
Structured Program Development in C
• A procedure for solving a problem in terms of the actions to be executed, and the order in which
these actions are to be executed, is called an algorithm.
• The order in which actions are to be executed is important.
Section 3.3 Pseudocode
•
•
•
•
•
•
Pseudocode is an artificial and informal language that helps you develop algorithms.
Pseudocode is similar to everyday English; it’s not an actual computer programming language.
Pseudocode programs help you “think out” a program.
Pseudocode consists purely of characters; you may type pseudocode using an editor.
Carefully prepared pseudocode programs may be converted easily to corresponding C programs.
Pseudocode consists only of action statements.
Section 3.4 Control Structures
• Normally, statements in a program execute one after the other in the order in which they’re written. This is called sequential execution.
• Various C statements enable you to specify that the next statement to execute may be other than
the next one in sequence. This is called transfer of control.
• Structured programming has become almost synonymous with “goto elimination.”
• Structured programs are clearer, easier to debug and modify and more likely to be bug free.
• All programs can be written in terms of sequence, selection and repetition control structures.
• Unless directed otherwise, the computer automatically executes C statements in sequence.
• A flowchart is a graphical representation of an algorithm. Flowcharts are drawn using rectangles,
diamonds, rounded rectangles and small circles, connected by arrows called flowlines.
• The rectangle (action) symbol indicates any type of action including a calculation or an input/
output operation.
• Flowlines indicate the order in which the actions are performed.
• When drawing a flowchart that represents a complete algorithm, a rounded rectangle containing
the word “Begin” is the first symbol used; a rounded rectangle symbol containing the word
“End” is the last symbol used. When drawing only a portion of an algorithm, we omit the rounded rectangle symbols in favor of using small circle symbols, also called connector symbols.
• The diamond (decision) symbol indicates that a decision is to be made.
• The if selection statement either performs (selects) an action if a condition is true or skips the
action if the condition is false. The if…else selection statement performs an action if a condition is true and performs a different action if the condition is false. The switch selection statement performs one of many different actions depending on the value of an expression.
• The if statement is called a single-selection statement because it selects or ignores a single action.
• The if…else statement is called a double-selection statement because it selects between two different actions.
• The switch statement is called a multiple-selection statement because it selects among many different actions.
• C provides three types of repetition statements (also called iteration statements), namely while,
do…while and for.
• Control-statement flowchart segments can be attached to one another with control-statement
stacking—connecting the exit point of one control statement to the entry point of the next.
• There’s only one other way control statements may be connected—control-statement nesting.
Summary
99
Section 3.5 The if Selection Statement
• Selection structures are used to choose among alternative courses of action.
• The decision symbol contains an expression, such as a condition, that can be either true or false.
The decision symbol has two flowlines emerging from it. One indicates the direction to be taken
when the expression is true; the other indicates the direction when the expression is false.
• A decision can be based on any expression—if the expression evaluates to zero, it’s treated as false,
and if it evaluates to nonzero, it’s treated as true.
• The if statement is a single-entry/single-exit structure.
Section 3.6 The if…else Selection Statement
• C provides the conditional operator (?:) which is closely related to the if…else statement.
• The conditional operator is C’s only ternary operator—it takes three operands. The first operand
is a condition. The second operand is the value for the conditional expression if the condition is
true, and the third operand is the value for the conditional expression if the condition is false.
• The values in a conditional expression can also be actions to execute.
• Nested if…else statements test for multiple cases by placing if…else statements inside
if…else statements.
• The if selection statement expects only one statement in its body. To include several statements
in the body of an if, you must enclose the set of statements in braces ({ and }).
• A set of statements contained within a pair of braces is called a compound statement or a block.
• A syntax error is caught by the compiler. A logic error has its effect at execution time. A fatal logic
error causes a program to fail and terminate prematurely. A nonfatal logic error allows a program
to continue executing but to produce incorrect results.
Section 3.7 The while Repetition Statement
• The while repetition statement specifies that an action is to be repeated while a condition is true.
Eventually, the condition will become false. At this point, the repetition terminates, and the first
statement after the repetition statement executes.
Section 3.8 Formulating Algorithms Case Study 1: Counter-Controlled Repetition
• Counter-controlled repetition uses a variable called a counter to specify the number of times a
set of statements should execute.
• Counter-controlled repetition is often called definite repetition because the number of repetitions is known before the loop begins executing.
• A total is a variable used to accumulate the sum of a series of values. Variables used to store totals
should normally be initialized to zero before being used in a program; otherwise the sum would
include the previous value stored in the total’s memory location.
• A counter is a variable used to count. Counter variables are normally initialized to zero or one,
depending on their use.
• An uninitialized variable contains a “garbage” value—the value last stored in the memory location reserved for that variable.
Section 3.9 Formulating Algorithms with Top-Down, Stepwise Refinement Case
Study 2: Sentinel-Controlled Repetition
• A sentinel value (also called a signal value, a dummy value, or a flag value) is used in a sentinelcontrolled loop to indicate the “end of data entry.”
100
Chapter 3 Structured Program Development in C
• Sentinel-controlled repetition is often called indefinite repetition because the number of repetitions is not known before the loop begins executing.
• The sentinel value must be chosen so that it cannot be confused with an acceptable input value.
• In top-down, stepwise refinement, the top is a statement that conveys the program’s overall function. It’s a complete representation of a program. In the refinement process, we divide the top
into smaller tasks and list these in execution order.
• The type float represents numbers with decimal points (called floating-point numbers).
• When two integers are divided any fractional part of the result is truncated.
• To produce a floating-point calculation with integer values, you must cast the integers to floating-point numbers. C provides the unary cast operator (float) to accomplish this task.
• Cast operators perform explicit conversions.
• Most computers can evaluate arithmetic expressions only in which the operands’ data types are
identical. To ensure this, the compiler performs an operation called implicit conversion on selected operands.
• Cast operators are available for most data types. A cast operator is formed by placing parentheses
around a type name. The cast operator is a unary operator—it takes only one operand.
• Cast operators associate from right to left and have the same precedence as other unary operators
such as unary + and unary -. This precedence is one level higher than that of *, / and %.
• The printf conversion specifier %.2f specifies that a floating-point value will be displayed with
two digits to the right of the decimal point. If the %f conversion specifier is used (without specifying the precision), the default precision of 6 is used.
• When floating-point values are printed with precision, the printed value is rounded to the indicated number of decimal positions for display purposes.
Section 3.11 Assignment Operators
• C provides several assignment operators for abbreviating assignment expressions.
• The += operator adds the value of the expression on the right of the operator to the value of the
variable on the left of the operator and stores the result in the variable on the left of the operator.
• Any statement of the form
variable
=
variable operator expression;
where operator is one of the binary operators +, -, *, / or % (or others we’ll discuss in Chapter 10),
can be written in the form
variable operator = expression;
Section 3.12 Increment and Decrement Operators
• C provides the unary increment operator, ++, and the unary decrement operator, --.
• If increment or decrement operators are placed before a variable, they’re referred to as the preincrement or predecrement operators, respectively. If increment or decrement operators are placed
after a variable, they’re referred to as the postincrement or postdecrement operators, respectively.
• Preincrementing (predecrementing) a variable causes it to be incremented (decremented) by 1,
then the new value of the variable is used in the expression in which it appears.
• Postincrementing (postdecrementing) a variable uses the current value of the variable in the expression in which it appears, then the variable value is incremented (decremented) by 1.
• When incrementing or decrementing a variable in a statement by itself, the preincrement and
postincrement forms have the same effect. When a variable appears in the context of a larger ex-
Terminology
101
pression, preincrementing and postincrementing have different effects (and similarly for predecrementing and postdecrementing).
Section 3.13 Secure C Programming
• Adding integers can result in a value that’s too large to store in an int variable. This is known as
arithmetic overflow and can cause unpredictable runtime behavior, possibly leaving a system
open to attack.
• The maximum and minimum values that can be stored in an int variable are represented by the
constants INT_MAX and INT_MIN, respectively, from the header <limits.h>.
• It’s considered a good practice to ensure that arithmetic calculations will not overflow before you
perform the calculation. In industrial-strength code, you should perform checks for all calculations that can result on overflow or underflow.
• In general, any integer variable that should store only non-negative values should be declared
with unsigned before the integer type. Variables of unsigned types can represent values from 0
to approximately double the positive range of the corresponding signed integer type.
• You can determine your platform’s maximum unsigned int value with the constant UINT_MAX
from <limits.h>.
• The C11 standard’s Annex K introduces more secure versions of printf and scanf called
printf_s and scanf_s. Annex K is designated as optional, so not every C compiler vendor will
implement it.
• Microsoft implemented its own versions of printf_s and scanf_s prior to the C11 standard’s publication and immediately began issuing warnings for every scanf call. The warnings say that scanf
is deprecated—it should no longer be used—and that you should consider using scanf_s instead.
• Many organizations have coding standards that require code to compile without warning messages. There are two ways to eliminate Visual C++’s scanf warnings. You can either start using
scanf_s immediately or disable this warning message.
Terminology
conditional operator 76
multiplication operator 88
*= multiplication assignment operator 93
/= division assignment operator 93
% remainder operator 88
%= remainder assignment operator 93
%u conversion specifier 90
-- decrement operator 94
++ increment operator 93
+= addition assignment operator 93
-= subtraction assignment operator 93
action 71
action symbol 73
addition assignment operator (+=) 93
algorithm 71
arithmetic overflow 96
block 78
cast operator 87
compound statement 78
conditional expression 76
?:
*
conditional operator (?:) 76
connector symbol 73
control-statement stacking 74
control structure 72
counter 80
counter-controlled repetition 80
decision symbol 73
default precision 88
definite repetition 81
diamond symbol 73
double-selection statement 73
dummy value 83
explicit conversion 87
fatal error 84
first refinement 83
flag value 83
float 85
floating-point number 85
flowchart 72
flowline 73
102
Chapter 3 Structured Program Development in C
“garbage value” 82
goto elimination 72
goto statement 72
implicit conversion 87
indefinite repetition 83
integer division 87
iteration statement 79
multiple-selection statement 73
multiplicative operator 88
nested statements 90
nested if...else statement 77
nesting statements 89
order 71
postdecrement operator (--) 93
postincrement operator (++) 93
precision 88
predecrement operator (--) 93
preincrement operator(++) 93
procedure 71
program control 71
pseudocode 71
rectangle symbol 73
repetition statement 79
repetition structure 72
rounded 88
rounded rectangle symbol 73
second refinement 83
selection structure 72
sentinel value 83
sequence structure 72
sequential execution 72
signal value 83
single-selection statement 73
single-entry/single-exit control statement 74
small circle symbols 73
top 83
top-down, stepwise refinement 83
total 82
transfer of control 72
truncated 87
unary operator 88
while repetition statement 79
white-space character 74
Self-Review Exercises
3.1
Fill in the blanks in each of the following questions.
a) A procedure for solving a problem in terms of the actions to be executed and the order
in which the actions should be executed is called a(n)
.
.
b) Specifying the execution order of statements by the computer is called
c) All programs can be written in terms of three types of control statements:
,
and
.
selection statement is used to execute one action when a condition is true
d) The
and another action when that condition is false.
e) Several statements grouped together in braces ({ and }) are called a(n)
.
repetition statement specifies that a statement or group of statements is
f) The
to be executed repeatedly while some condition remains true.
g) Repetition of a set of instructions a specific number of times is called
repetition.
h) When it’s not known in advance how many times a set of statements will be repeated,
value can be used to terminate the repetition.
a(n)
3.2
Write four different C statements that each add 1 to integer variable x.
3.3
Write a single C statement to accomplish each of the following:
a) Multiply the variable product by 2 using the *= operator.
b) Multiply the variable product by 2 using the = and * operators.
c) Test if the value of the variable count is greater than 10. If it is, print “Count is greater
than 10.”
d) Calculate the remainder after q is divided by divisor and assign the result to q. Write
this statement two different ways.
e) Print the value 123.4567 with two digits of precision. What value is printed?
f) Print the floating-point value 3.14159 with three digits to the right of the decimal point.
What value is printed?
Answers to Self-Review Exercises
3.4
103
Write a C statement to accomplish each of the following tasks.
a) Define variables sum and x to be of type int.
b) Set variable x to 1.
c) Set variable sum to 0.
d) Add variable x to variable sum and assign the result to variable sum.
e) Print "The sum is: " followed by the value of variable sum.
3.5
Combine the statements that you wrote in Exercise 3.4 into a program that calculates the
sum of the integers from 1 to 10. Use the while statement to loop through the calculation and increment statements. The loop should terminate when the value of x becomes 11.
3.6
Write single C statements that
a) Input unsigned integer variable x with scanf. Use the conversion specifier %u.
b) Input unsigned integer variable y with scanf. Use the conversion specifier %u.
c) Set unsigned integer variable i to 1.
d) Set unsigned integer variable power to 1.
e) Multiply unsigned integer variable power by x and assign the result to power.
f) Increment variable i by 1.
g) Test i to see if it’s less than or equal to y in the condition of a while statement.
h) Output unsigned integer variable power with printf. Use the conversion specifier %u.
3.7
Write a C program that uses the statements in Exercise 3.6 to calculate
power. The program should have a while repetition control statement.
3.8
x
raised to the
y
Identify and correct the errors in each of the following:
a) while ( c <= 5 ) {
product *= c;
++c;
b)
c)
scanf( "%.4f", &value );
if ( gender == 1 )
puts( "Woman" );
else;
puts( "Man" );
3.9
What’s wrong with the following while repetition statement (assume z has value 100),
which is supposed to calculate the sum of the integers from 100 down to 1?
while ( z >= 0 )
sum += z;
Answers to Self-Review Exercises
3.1
a) Algorithm. b) Program control. c) Sequence, selection, repetition. d) if…else. e) Compound statement or block. f) while. g) Counter-controlled or definite. h) Sentinel.
3.2
x = x + 1;
x += 1;
++x;
x++;
3.3
a)
b)
c)
product *= 2;
d)
q %= divisor;
product = product * 2;
if ( count > 10 )
puts( "Count is greater than 10." );
q = q % divisor;
104
Chapter 3 Structured Program Development in C
e)
printf( "%.2f", 123.4567 );
f)
printf( "%.3f\n", 3.14159 );
123.46
3.142
3.4
3.5
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
3.6
3.7
1
2
3
4
5
6
7
8
9
10
11
12
13
14
a)
b)
c)
d)
e)
is displayed.
is displayed.
int sum, x;
x = 1;
sum = 0;
sum += x; or sum = sum + x;
printf( "The sum is: %d\n", sum );
See below.
// Calculate the sum of the integers from 1 to 10
#include <stdio.h>
int main( void )
{
unsigned int sum, x; // define variables sum and x
]
x = 1; // set x
sum = 0; // set sum
while ( x <= 10 ) { // loop while x is less than or equal to 10
sum += x; // add x to sum
++x; // increment x
} // end while
printf( "The sum is: %u\n", sum ); // display sum
} // end main function
a)
b)
c)
d)
e)
f)
g)
h)
scanf( "%u", &x );
scanf( "%u", &y );
i = 1;
power = 1;
power *= x;
++i;
while ( i <= y )
printf( "%d", power );
See top of next page.
// raise x to the y power
#include <stdio.h>
int main( void )
{
unsigned int x, y, i, power; // define variables
i = 1; // set i
power = 1; // set power
printf( "%s", "Enter first integer: " );
scanf( "%u", &x ); // read value for x from user
printf( "%s", "Enter second integer: " );
scanf( "%u", &y ); // read value for y from user
Exercises
15
16
17
18
19
20
21
105
while ( i <= y ) { // loop while i is less than or equal to y
power *= x; // multiply power by x
++i; // increment i
} // end while
printf( "%u\n", power ); // display power
} // end main function
3.8
a) Error: Missing the closing right brace of the while body.
Correction: Add closing right brace after the statement ++c;.
b) Error: Precision used in a scanf conversion specification.
Correction: Remove .4 from the conversion specification.
c) Error: Semicolon after the else part of the if…else statement results in a logic error.
The second puts will always be executed.
Correction: Remove the semicolon after else.
3.9
The value of the variable z is never changed in the while statement. Therefore, an infinite
loop is created. To prevent the infinite loop, z must be decremented so that it eventually becomes 0.
Exercises
3.10 Identify and correct the errors in each of the following. [Note: There may be more than one
error in each piece of code.]
a) if ( age >= 65 );
puts( "Age is greater than or equal to 65" );
else
puts( "Age is less than 65" );
b)
int x = 1, total;
while ( x <= 10 ) {
total += x;
++x;
}
c)
While ( x <= 100 )
total += x;
++x;
d)
while ( y > 0 ) {
printf( "%d\n", y );
++y;
}
3.11
Fill in the blanks in each of the following:
a) The solution to any problem involves performing a series of actions in a specific
.
.
b) A synonym for procedure is
c) A variable that accumulates the sum of several numbers is a(n)
.
d) A special value used to indicate “end of data entry” is called a(n)
, a(n)
, a(n)
or a(n)
value.
e) A(n)
is a graphical representation of an algorithm.
f) In a flowchart, the order in which the steps should be performed is indicated by
symbols.
106
Chapter 3 Structured Program Development in C
g) Rectangle symbols correspond to calculations that are normally performed by
statements and input/output operations that are normally performed by calls to the
and
Standard Library functions.
h) The item written inside a decision symbol is called a(n)
.
3.12
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
What does the following program print?
#include <stdio.h>
int main( void )
{
unsigned int x = 1, total = 0, y;
while ( x <= 10 ) {
y = x * x;
printf( "%d\n", y );
total += y;
++x;
} // end while
printf( "Total is %d\n", total );
} // end main
3.13
Write a single pseudocode statement that indicates each of the following:
a) Display the message "Enter two numbers".
b) Assign the sum of variables x, y, and z to variable p.
c) The following condition is to be tested in an if…else selection statement: The current
value of variable m is greater than twice the current value of variable v.
d) Obtain values for variables s, r, and t from the keyboard.
3.14
Formulate a pseudocode algorithm for each of the following:
a) Obtain two numbers from the keyboard, compute their sum and display the result.
b) Obtain two numbers from the keyboard, and determine and display which (if either) is
the larger of the two numbers.
c) Obtain a series of positive numbers from the keyboard, and determine and display their
sum. Assume that the user types the sentinel value -1 to indicate “end of data entry.”
3.15
State which of the following are true and which are false. If a statement is false, explain why.
a) Experience has shown that the most difficult part of solving a problem on a computer
is producing a working C program.
b) A sentinel value must be a value that cannot be confused with a legitimate data value.
c) Flowlines indicate the actions to be performed.
d) Conditions written inside decision symbols always contain arithmetic operators (i.e., +,
-, *, /, and %).
e) In top-down, stepwise refinement, each refinement is a complete representation of the
algorithm.
For Exercises 3.16–3.20, perform each of these steps:
1. Read the problem statement.
2. Formulate the algorithm using pseudocode and top-down, stepwise refinement.
3. Write a C program.
4. Test, debug and execute the C program.
3.16 (Gas Mileage) Drivers are concerned with the mileage obtained by their automobiles. One
driver has kept track of several tankfuls of gasoline by recording miles driven and gallons used for
Exercises
107
each tankful. Develop a program that will input the miles driven and gallons used for each tankful.
The program should calculate and display the miles per gallon obtained for each tankful. After processing all input information, the program should calculate and print the combined miles per gallon
obtained for all tankfuls. Here is a sample input/output dialog:
Enter the gallons used (-1 to end): 12.8
Enter the miles driven: 287
The miles/gallon for this tank was 22.421875
Enter the gallons used (-1 to end): 10.3
Enter the miles driven: 200
The miles/gallon for this tank was 19.417475
Enter the gallons used (-1 to end): 5
Enter the miles driven: 120
The miles/gallon for this tank was 24.000000
Enter the gallons used (-1 to end): -1
The overall average miles/gallon was 21.601423
3.17 (Credit Limit Calculator) Develop a C program that will determine if a department store
customer has exceeded the credit limit on a charge account. For each customer, the following facts
are available:
a) Account number
b) Balance at the beginning of the month
c) Total of all items charged by this customer this month
d) Total of all credits applied to this customer's account this month
e) Allowed credit limit
The program should input each fact, calculate the new balance (= beginning balance +
charges – credits), and determine whether the new balance exceeds the customer's credit limit. For
those customers whose credit limit is exceeded, the program should display the customer's account
number, credit limit, new balance and the message “Credit limit exceeded.” Here is a sample
input/output dialog:
Enter account number (-1 to end): 100
Enter beginning balance: 5394.78
Enter total charges: 1000.00
Enter total credits: 500.00
Enter credit limit: 5500.00
Account:
100
Credit limit: 5500.00
Balance:
5894.78
Credit Limit Exceeded.
Enter
Enter
Enter
Enter
Enter
account number (-1 to end): 200
beginning balance: 1000.00
total charges: 123.45
total credits: 321.00
credit limit: 1500.00
Enter
Enter
Enter
Enter
Enter
account number (-1 to end): 300
beginning balance: 500.00
total charges: 274.73
total credits: 100.00
credit limit: 800.00
Enter account number (-1 to end): -1
108
Chapter 3 Structured Program Development in C
3.18 (Sales Commission Calculator) One large chemical company pays its salespeople on a commission basis. The salespeople receive $200 per week plus 9% of their gross sales for that week. For
example, a salesperson who sells $5000 worth of chemicals in a week receives $200 plus 9% of
$5000, or a total of $650. Develop a program that will input each salesperson’s gross sales for last
week and will calculate and display that salesperson’s earnings. Process one salesperson's figures at a
time. Here is a sample input/output dialog:
Enter sales in dollars (-1 to end): 5000.00
Salary is: $650.00
Enter sales in dollars (-1 to end): 1234.56
Salary is: $311.11
Enter sales in dollars (-1 to end): -1
3.19
(Interest Calculator) The simple interest on a loan is calculated by the formula
interest = principal * rate * days / 365;
The preceding formula assumes that rate is the annual interest rate, and therefore includes the
division by 365 (days). Develop a program that will input principal, rate and days for several
loans, and will calculate and display the simple interest for each loan, using the preceding formula.
Here is a sample input/output dialog:
Enter loan principal (-1 to end): 1000.00
Enter interest rate: .1
Enter term of the loan in days: 365
The interest charge is $100.00
Enter loan principal (-1 to end): 1000.00
Enter interest rate: .08375
Enter term of the loan in days: 224
The interest charge is $51.40
Enter loan principal (-1 to end): -1
3.20 (Salary Calculator) Develop a program that will determine the gross pay for each of several
employees. The company pays “straight time” for the first 40 hours worked by each employee and
pays “time-and-a-half” for all hours worked in excess of 40 hours. You’re given a list of the employees of the company, the number of hours each employee worked last week and the hourly rate of
each employee. Your program should input this information for each employee and should determine and display the employee's gross pay. Here is a sample input/output dialog:
Enter # of hours worked (-1 to end): 39
Enter hourly rate of the worker ($00.00): 10.00
Salary is $390.00
Enter # of hours worked (-1 to end): 40
Enter hourly rate of the worker ($00.00): 10.00
Salary is $400.00
Enter # of hours worked (-1 to end): 41
Enter hourly rate of the worker ($00.00): 10.00
Salary is $415.00
Enter # of hours worked (-1 to end): -1
Exercises
109
3.21 (Predecrementing vs. Postdecrementing) Write a program that demonstrates the difference
between predecrementing and postdecrementing using the decrement operator --.
3.22 (Printing Numbers from a Loop) Write a program that utilizes looping to print the numbers from 1 to 10 side by side on the same line with three spaces between numbers.
3.23 (Find the Largest Number) The process of finding the largest number (i.e., the maximum
of a group of numbers) is used frequently in computer applications. For example, a program that
determines the winner of a sales contest would input the number of units sold by each salesperson.
The salesperson who sells the most units wins the contest. Write a pseudocode program and then a
program that inputs a series of 10 non-negative numbers and determines and prints the largest of
the numbers. Hint: Your program should use three variables as follows:
counter:
number:
largest:
A counter to count to 10 (i.e., to keep track of how many numbers have
been input and to determine when all 10 numbers have been processed)
The current number input to the program
The largest number found so far
3.24 (Tabular Output) Write a program that uses looping to print the following table of values.
Use the tab escape sequence, \t, in the printf statement to separate the columns with tabs.
N
10*N
100*N
1000*N
1
2
3
4
5
6
7
8
9
10
10
20
30
40
50
60
70
80
90
100
100
200
300
400
500
600
700
800
900
1000
1000
2000
3000
4000
5000
6000
7000
8000
9000
10000
3.25 (Tabular Output) Write a program that utilizes looping to produce the following table of
values:
A
A+2
A+4
A+6
3
6
9
12
15
5
8
11
14
17
7
10
13
16
19
9
12
15
18
21
3.26 (Find the Two Largest Numbers) Using an approach similar to Exercise 3.23, find the two
largest values of the 10 numbers. [Note: You may input each number only once.]
3.27 (Validating User Input) Modify the program in Figure 3.10 to validate its inputs. On any
input, if the value entered is other than 1 or 2, keep looping until the user enters a correct value.
3.28
1
2
3
4
What does the following program print?
#include <stdio.h>
int main( void )
{
110
5
6
7
8
9
10
11
12
13
unsigned int count = 1; // initialize count
while ( count <= 10 ) { // loop 10 times
// output line of text
puts( count % 2 ? "****" : "++++++++" );
++count; // increment count
} // end while
} // end function main
3.29
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
Chapter 3 Structured Program Development in C
What does the following program print?
#include <stdio.h>
int main( void )
{
unsigned int row = 10; // initialize row
unsigned int column; // define column
while ( row >= 1 ) { // loop until row < 1
column = 1; // set column to 1 as iteration begins
while ( column <= 10 ) { // loop 10 times
printf( "%s", row % 2 ? "<": ">" ); // output
++column; // increment column
} // end inner while
--row; // decrement row
puts( "" ); // begin new output line
} // end outer while
} // end function main
3.30 (Dangling Else Problem) Determine the output for each of the following when x is 9 and y
is 11, and when x is 11 and y is 9. The compiler ignores the indentation in a C program. Also, the
compiler always associates an else with the previous if unless told to do otherwise by the placement
of braces {}. Because, on first glance, you may not be sure which if an else matches, this is referred
to as the “dangling else” problem. We eliminated the indentation from the following code to make
the problem more challenging. [Hint: Apply indentation conventions you have learned.]
a) if ( x < 10 )
if ( y > 10 )
puts( "*****" );
else
puts( "#####" );
puts( "$$$$$" );
b)
if ( x < 10 ) {
if ( y > 10 )
puts( "*****" );
}
else {
puts( "#####" );
puts( "$$$$$" );
}
3.31 (Another Dangling Else Problem) Modify the following code to produce the output shown.
Use proper indentation techniques. You may not make any changes other than inserting braces. The
Exercises
111
compiler ignores the indentation in a program. We eliminated the indentation from the following
code to make the problem more challenging. [Note: It’s possible that no modification is necessary.]
if ( y == 8 )
if ( x == 5 )
puts( "@@@@@"
else
puts( "#####"
puts( "$$$$$"
puts( "&&&&&"
);
);
);
);
a) Assuming x = 5 and y = 8, the following output is produced.
@@@@@
$$$$$
&&&&&
b) Assuming x = 5 and y = 8, the following output is produced.
@@@@@
c) Assuming x = 5 and y = 8, the following output is produced.
@@@@@
&&&&&
d) Assuming x = 5 and y = 7, the following output is produced. [Note: The last three puts
statements are all part of a compound statement.]
#####
$$$$$
&&&&&
3.32 (Square of Asterisks) Write a program that reads in the side of a square and then prints that
square out of asterisks. Your program should work for squares of all side sizes between 1 and 20. For
example, if your program reads a size of 4, it should print
****
****
****
****
3.33 (Hollow Square of Asterisks) Modify the program you wrote in Exercise 3.32 so that it
prints a hollow square. For example, if your program reads a size of 5, it should print
*****
*
*
*
*
*
*
*****
112
Chapter 3 Structured Program Development in C
3.34 (Palindrome Tester) A palindrome is a number or a text phrase that reads the same backward as forward. For example, each of the following five-digit integers is a palindrome: 12321,
55555, 45554 and 11611. Write a program that reads in a five-digit integer and determines whether
or not it’s a palindrome. [Hint: Use the division and remainder operators to separate the number
into its individual digits.]
3.35 (Printing the Decimal Equivalent of a Binary Number) Input an integer (5 digits or fewer)
containing only 0s and 1s (i.e., a “binary” integer) and print its decimal equivalent. [Hint: Use the
remainder and division operators to pick off the “binary” number’s digits one at a time from right
to left. Just as in the decimal number system, in which the rightmost digit has a positional value of
1, and the next digit left has a positional value of 10, then 100, then 1000, and so on, in the binary
number system the rightmost digit has a positional value of 1, the next digit left has a positional
value of 2, then 4, then 8, and so on. Thus the decimal number 234 can be interpreted as 4 * 1 + 3
* 10 + 2 * 100. The decimal equivalent of binary 1101 is 1 * 1 + 0 * 2 + 1 * 4 + 1 * 8 or 1 + 0 + 4
+ 8 or 13.]
3.36 (How Fast is Your Computer?) How can you determine how fast your own computer really
operates? Write a program with a while loop that counts from 1 to 1,000,000,000 by 1s. Every time
the count reaches a multiple of 100,000,000, print that number on the screen. Use your watch to
time how long each 100 million repetitions of the loop takes.
3.37 (Detecting Multiples of 10) Write a program that prints 100 asterisks, one at a time. After
every tenth asterisk, your program should print a newline character. [Hint: Count from 1 to 100.
Use the remainder operator to recognize each time the counter reaches a multiple of 10.]
3.38 (Counting 7s) Write a program that reads an integer (5 digits or fewer) and determines and
prints how many digits in the integer are 7s.
3.39 (Checkerboard Pattern of Asterisks) Write a program that displays the following checkerboard pattern:
* * * * * * * *
* * * * * * * *
* * * * * * * *
* * * * * * * *
* * * * * * * *
* * * * * * * *
* * * * * * * *
* * * * * * * *
Your program must use only three output statements, one of each of the following forms:
printf( "%s", "* " );
printf( "%s", " " );
puts( "" ); // outputs a newline
3.40 (Multiples of 2 with an Infinite Loop) Write a program that keeps printing the multiples of
the integer 2, namely 2, 4, 8, 16, 32, 64, and so on. Your loop should not terminate (i.e., you should
create an infinite loop). What happens when you run this program?
3.41 (Diameter, Circumference and Area of a Cirle) Write a program that reads the radius of a
circle (as a float value) and computes and prints the diameter, the circumference and the area. Use
the value 3.14159 for π.
3.42 What’s wrong with the following statement? Rewrite it to accomplish what the programmer
was probably trying to do.
printf( "%d", ++( x + y ) );
Making a Difference
113
3.43 (Sides of a Triangle) Write a program that reads three nonzero integer values and determines and prints whether they could represent the sides of a triangle.
3.44 (Sides of a Right Triangle) Write a program that reads three nonzero integers and determines and prints whether they could be the sides of a right triangle.
3.45 (Factorial) The factorial of a nonnegative integer n is written n! (pronounced “n factorial”)
and is defined as follows:
n! = n · (n - 1) · (n - 2) · … · 1 (for values of n greater than or equal to 1)
and
n! = 1 (for n = 0).
For example, 5! = 5 · 4 · 3 · 2 · 1, which is 120.
a) Write a program that reads a nonnegative integer and computes and prints its factorial.
b) Write a program that estimates the value of the mathematical constant e by using the
formula:
1- + ---1- + ---1- + …
e = 1 + ---1! 2! 3!
c) Write a program that computes the value of ex by using the formula
2
3
x
x- + ---x - + ---x- + …
e = 1 + ---1! 2! 3!
Making a Difference
3.46 (World-Population-Growth Calculator) Use the web to determine the current world population and the annual world population growth rate. Write an application that inputs these values,
then displays the estimated world population after one, two, three, four and five years.
3.47 (Target-Heart-Rate Calculator) While exercising, you can use a heart-rate monitor to see
that your heart rate stays within a safe range suggested by your trainers and doctors. According to
the American Heart Association (AHA), the formula for calculating your maximum heart rate in
beats per minute is 220 minus your age in years. Your target heart rate is a range that’s 50–85% of
your maximum heart rate. [Note: These formulas are estimates provided by the AHA. Maximum
and target heart rates may vary based on the health, fitness and gender of the individual. Always consult a physician or qualified health care professional before beginning or modifying an exercise program.]
Create a program that reads the user’s birthday and the current day (each consisting of the month,
day and year). Your program should calculate and display the person’s age (in years), the person’s
maximum heart rate and the person’s target-heart-rate range.
3.48 (Enforcing Privacy with Cryptography) The explosive growth of Internet communications
and data storage on Internet-connected computers has greatly increased privacy concerns. The field
of cryptography is concerned with coding data to make it difficult (and hopefully—with the most
advanced schemes—impossible) for unauthorized users to read. In this exercise you’ll investigate a
simple scheme for encrypting and decrypting data. A company that wants to send data over the Internet has asked you to write a program that will encrypt it so that it may be transmitted more securely. All the data is transmitted as four-digit integers. Your application should read a four-digit
integer entered by the user and encrypt it as follows: Replace each digit with the result of adding 7
to the digit and getting the remainder after dividing the new value by 10. Then swap the first digit
with the third, and swap the second digit with the fourth. Then print the encrypted integer. Write
a separate application that inputs an encrypted four-digit integer and decrypts it (by reversing the
encryption scheme) to form the original number. [Optional reading project: Research “public key
cryptography” in general and the PGP (Pretty Good Privacy) specific public key scheme. You may
also want to investigate the RSA scheme, which is widely used in industrial-strength applications.]
4
Who can control his fate?
—William Shakespeare
The used key is always bright.
—Benjamin Franklin
Every advantage in the past is
judged in the light of the final
issue.
—Demosthenes
Objectives
In this chapter, you’ll learn:
■
■
■
■
■
■
The essentials of countercontrolled repetition.
To use the for and
do…while repetition
statements to execute
statements repeatedly.
To understand multiple
selection using the switch
selection statement.
To use the break and
continue statements to
alter the flow of control.
To use the logical operators
to form complex conditional
expressions in control
statements.
To avoid the consequences of
confusing the equality and
assignment operators.
C Program Control
4.1 Introduction
4.1
4.2
4.3
4.4
4.5
Introduction
Repetition Essentials
Counter-Controlled Repetition
for Repetition Statement
for Statement: Notes and
Observations
4.6 Examples Using the for Statement
4.7 switch Multiple-Selection Statement
4.8
4.9
4.10
4.11
115
do…while Repetition Statement
break and continue Statements
Logical Operators
Confusing Equality (==) and
Assignment (=) Operators
4.12 Structured Programming Summary
4.13 Secure C Programming
Summary | Terminology | Self-Review Exercises | Answers to Self-Review Exercises | Exercises |
Making a Difference
4.1 Introduction
You should now be comfortable with writing simple, complete C programs. In this chapter, repetition is considered in greater detail, and additional repetition control statements,
namely the for and the do…while, are presented. The switch multiple-selection statement is introduced. We discuss the break statement for exiting immediately from certain
control statements, and the continue statement for skipping the remainder of the body of
a repetition statement and proceeding with the next iteration of the loop. The chapter discusses logical operators used for combining conditions, and summarizes the principles of
structured programming as presented in Chapters 3 and 4.
4.2 Repetition Essentials
Most programs involve repetition, or looping. A loop is a group of instructions the computer executes repeatedly while some loop-continuation condition remains true. We’ve
discussed two means of repetition:
1. Counter-controlled repetition
2. Sentinel-controlled repetition
Counter-controlled repetition is sometimes called definite repetition because we know in
advance exactly how many times the loop will be executed. Sentinel-controlled repetition
is sometimes called indefinite repetition because it’s not known in advance how many times
the loop will be executed.
In counter-controlled repetition, a control variable is used to count the number of
repetitions. The control variable is incremented (usually by 1) each time the group of
instructions is performed. When the value of the control variable indicates that the correct
number of repetitions has been performed, the loop terminates and execution continues
with the statement after the repetition statement.
Sentinel values are used to control repetition when:
1. The precise number of repetitions isn’t known in advance, and
2. The loop includes statements that obtain data each time the loop is performed.
The sentinel value indicates “end of data.” The sentinel is entered after all regular data items
have been supplied to the program. Sentinels must be distinct from regular data items.
116
Chapter 4 C Program Control
4.3 Counter-Controlled Repetition
Counter-controlled repetition requires:
1. The name of a control variable (or loop counter).
2. The initial value of the control variable.
3. The increment (or decrement) by which the control variable is modified each
time through the loop.
4. The condition that tests for the final value of the control variable (i.e., whether
looping should continue).
Consider the simple program shown in Fig. 4.1, which prints the numbers from 1 to
10. The definition
unsigned int counter = 1; // initialization
names the control variable (counter), defines it to be an integer, reserves memory space
for it, and sets it to an initial value of 1.
1
2
3
4
5
6
7
8
9
10
11
12
13
14
// Fig. 4.1: fig04_01.c
// Counter-controlled repetition.
#include <stdio.h>
// function main begins program execution
int main( void )
{
unsigned int counter = 1; // initialization
while ( counter <= 10 ) { // repetition condition
printf ( "%u\n", counter ); // display counter
++counter; // increment
} // end while
} // end function main
1
2
3
4
5
6
7
8
9
10
Fig. 4.1 | Counter-controlled repetition.
The definition and initialization of counter could also have been written as
unsigned int counter;
counter = 1;
The definition is not executable, but the assignment is. We use both methods of setting
the values of variables.
4.4 for Repetition Statement
117
The statement
++counter; // increment
increments the loop counter by 1 each time the loop is performed. The loop-continuation
condition in the while statement tests whether the value of the control variable is less than
or equal to 10 (the last value for which the condition is true). The body of this while is
performed even when the control variable is 10. The loop terminates when the control
variable exceeds 10 (i.e., counter becomes 11).
You could make the program in Fig. 4.1 more concise by initializing counter to 0 and
by replacing the while statement with
while ( ++counter <= 10 )
printf( "%u\n", counter );
This code saves a statement because the incrementing is done directly in the while condition before the condition is tested. Also, this code eliminates the need for the braces
around the body of the while because the while now contains only one statement. Coding
in such a condensed fashion takes some practice. Some programmers feel that this makes
the code too cryptic and error prone.
Common Programming Error 4.1
Floating-point values may be approximate, so controlling counting loops with floatingpoint variables may result in imprecise counter values and inaccurate termination tests.
Error-Prevention Tip 4.1
Control counting loops with integer values.
Good Programming Practice 4.1
Too many levels of nesting can make a program difficult to understand. As a rule, try to
avoid using more than three levels of nesting.
Good Programming Practice 4.2
The combination of vertical spacing before and after control statements and indentation
of the bodies of control statements within the control-statement headers gives programs a
two-dimensional appearance that greatly improves program readability.
4.4 for Repetition Statement
The for repetition statement handles all the details of counter-controlled repetition. To
illustrate its power, let’s rewrite the program of Fig. 4.1. The result is shown in Fig. 4.2.
The program operates as follows. When the for statement begins executing, the control variable counter is initialized to 1. Then, the loop-continuation condition counter <=
10 is checked. Because the initial value of counter is 1, the condition is satisfied, so the
printf statement (line 13) prints the value of counter, namely 1. The control variable
counter is then incremented by the expression ++counter, and the loop begins again with
the loop-continuation test. Because the control variable is now equal to 2, the final value is
not exceeded, so the program performs the printf statement again. This process continues
until the control variable counter is incremented to its final value of 11—this causes the
118
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
Chapter 4 C Program Control
// Fig. 4.2: fig04_02.c
// Counter-controlled repetition with the for statement.
#include <stdio.h>
// function main begins program execution
int main( void )
{
unsigned int counter; // define counter
// initialization, repetition condition, and increment
// are all included in the for statement header.
for ( counter = 1; counter <= 10; ++counter ) {
printf( "%u\n", counter );
} // end for
} // end function main
Fig. 4.2 | Counter-controlled repetition with the for statement.
loop-continuation test to fail, and repetition terminates. The program continues by performing the first statement after the for statement (in this case, the end of the program).
Statement Header Components
Figure 4.3 takes a closer look at the for statement of Fig. 4.2. Notice that the for statement “does it all”—it specifies each of the items needed for counter-controlled repetition
with a control variable. If there’s more than one statement in the body of the for, braces
are required to define the body of the loop.
The C standard allows you to declare the control variable in the initialization section
of the for header (as in int counter = 1). We show a complete code example of this in
Appendix F. This feature is not supported in Microsoft Visual C++.
for
for
keyword
Control
variable
name
Required Final value of control Required
semicolon variable for which
semicolon
separator the condition is true separator
for ( counter = 1; counter <= 10; ++counter )
Initial value of
control variable
Fig. 4.3 |
for
Loop-continuation
condition
Increment of
control variable
statement header components.
Off-By-One Errors
Notice that Fig. 4.2 uses the loop-continuation condition counter <= 10. If you incorrectly wrote counter < 10, then the loop would be executed only 9 times. This is a common
logic error called an off-by-one error.
4.4 for Repetition Statement
119
Error-Prevention Tip 4.2
Using the final value in the condition of a while or for statement and using the <= relational operator can help avoid off-by-one errors. For a loop used to print the values 1 to
10, for example, the loop-continuation condition should be counter <= 10 rather than
counter < 11 or counter < 10.
General Format of a for Statement
The general format of the for statement is
for ( expression1; expression2; expression3 ) {
statement
}
where expression1 initializes the loop-control variable, expression2 is the loop-continuation
condition, and expression3 increments the control variable. In most cases, the for statement can be represented with an equivalent while statement as follows:
expression1;
while ( expression2 ) {
statement
expression3;
}
There’s an exception to this rule, which we discuss in Section 4.9.
Comma-Separated Lists of Expressions
Often, expression1 and expression3 are comma-separated lists of expressions. The commas
as used here are actually comma operators that guarantee that lists of expressions evaluate
from left to right. The value and type of a comma-separated list of expressions are the value
and type of the rightmost expression in the list. The comma operator is most often used
in the for statement. Its primary use is to enable you to use multiple initialization and/or
multiple increment expressions. For example, there may be two control variables in a single for statement that must be initialized and incremented.
Software Engineering Observation 4.1
Place only expressions involving the control variables in the initialization and increment
sections of a for statement. Manipulations of other variables should appear either before
the loop (if they execute only once, like initialization statements) or in the loop body (if
they execute once per repetition, like incrementing or decrementing statements).
Expressions in the for Statement’s Header Are Optional
The three expressions in the for statement are optional. If expression2 is omitted, C assumes
that the condition is true, thus creating an infinite loop. You may omit expression1 if the control variable is initialized elsewhere in the program. expression3 may be omitted if the increment is calculated by statements in the body of the for statement or if no increment is
needed.
Increment Expression Acts Like a Standalone Statement
The increment expression in the for statement acts like a stand-alone C statement at the
end of the body of the for. Therefore, the expressions
120
Chapter 4 C Program Control
counter = counter + 1
counter += 1
++counter
counter++
are all equivalent in the increment part of the for statement. Some C programmers prefer
the form counter++ because the incrementing occurs after the loop body is executed, and
the postincrementing form seems more natural. Because the variable being preincremented or postincremented here does not appear in a larger expression, both forms of incrementing have the same effect. The two semicolons in the for statement are required.
Common Programming Error 4.2
Using commas instead of semicolons in a for header is a syntax error.
Common Programming Error 4.3
Placing a semicolon immediately to the right of a for header makes the body of that for
statement an empty statement. This is normally a logic error.
4.5 for Statement: Notes and Observations
1. The initialization, loop-continuation condition and increment can contain arithmetic expressions. For example, if x = 2 and y = 10, the statement
for ( j = x; j <= 4 * x * y; j += y / x )
is equivalent to the statement
for ( j = 2; j <= 80; j += 5 )
2. The “increment” may be negative (in which case it’s really a decrement and the
loop actually counts downward).
3. If the loop-continuation condition is initially false, the loop body does not execute. Instead, execution proceeds with the statement following the for statement.
4. The control variable is frequently printed or used in calculations in the body of a
loop, but it need not be. It’s common to use the control variable for controlling
repetition while never mentioning it in the body of the loop.
5. The for statement is flowcharted much like the while statement. For example,
Fig. 4.4 shows the flowchart of the for statement
for ( counter = 1; counter <= 10; ++counter )
printf( "%u", counter );
This flowchart makes it clear that the initialization occurs only once and that incrementing occurs after the body statement is performed.
Error-Prevention Tip 4.3
Although the value of the control variable can be changed in the body of a for loop, this
can lead to subtle errors. It’s best not to change it.
4.6 Examples Using the for Statement
121
Establish initial
value of control
variable
counter = 1
counter <= 10
Determine if final
value of control
variable has been
reached
false
true
printf( "%u", counter );
Body of loop
(this may be many
statements)
++counter
Increment
the control
variable
Fig. 4.4 | Flowcharting a typical for repetition statement.
4.6 Examples Using the for Statement
The following examples show methods of varying the control variable in a for statement.
1. Vary the control variable from 1 to 100 in increments of 1.
for ( i = 1; i <= 100; ++i )
2. Vary the control variable from 100 to 1 in increments of -1 (decrements of 1).
for ( i = 100; i >= 1; --i )
3. Vary the control variable from 7 to 77 in steps of 7.
for ( i = 7; i <= 77; i += 7 )
4. Vary the control variable from 20 to 2 in steps of -2.
for ( i = 20; i >= 2; i -= 2 )
5. Vary the control variable over the following sequence of values: 2, 5, 8, 11, 14, 17.
for ( j = 2; j <= 17; j += 3 )
6. Vary the control variable over the following sequence of values: 44, 33, 22, 11, 0.
for ( j = 44; j >= 0; j -= 11 )
Application: Summing the Even Integers from 2 to 100
Figure 4.5 uses the for statement to sum all the even integers from 2 to 100. Each iteration
of the loop (lines 11–13) adds control variable number’s value to variable sum.
1
2
3
4
// Fig. 4.5: fig04_05.c
// Summation with for.
#include <stdio.h>
Fig. 4.5 | Summation with for. (Part 1 of 2.)
122
5
6
7
8
9
10
11
12
13
14
15
16
Chapter 4 C Program Control
// function main begins program execution
int main( void )
{
unsigned int sum = 0; // initialize sum
unsigned int number; // number to be added to sum
for ( number = 2; number <= 100; number += 2 ) {
sum += number; // add number to sum
} // end for
printf( "Sum is %u\n", sum ); // output sum
} // end function main
Sum is 2550
Fig. 4.5 | Summation with for. (Part 2 of 2.)
The body of the for statement in Fig. 4.5 could actually be merged into the rightmost
portion of the for header by using the comma operator as follows:
for ( number = 2; number <= 100; sum += number, number += 2 )
; // empty statement
The initialization sum = 0 could also be merged into the initialization section of the for.
Good Programming Practice 4.3
Although statements preceding a for and statements in the body of a for can often be
merged into the for header, avoid doing so, because it makes the program more difficult
to read.
Good Programming Practice 4.4
Limit the size of control-statement headers to a single line if possible.
Application: Compound-Interest Calculations
The next example computes compound interest using the for statement. Consider the following problem statement:
A person invests $1000.00 in a savings account yielding 5% interest. Assuming that
all interest is left on deposit in the account, calculate and print the amount of money
in the account at the end of each year for 10 years. Use the following formula for
determining these amounts:
a = p(1 + r) n
where
p is the original amount invested (i.e., the principal)
r is the annual interest rate
n is the number of years
a is the amount on deposit at the end of the nth year.
This problem involves a loop that performs the indicated calculation for each of the
10 years the money remains on deposit. The solution is shown in Fig. 4.6.
4.6 Examples Using the for Statement
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
123
// Fig. 4.6: fig04_06.c
// Calculating compound interest.
#include <stdio.h>
#include <math.h>
// function main begins program execution
int main( void )
{
double amount; // amount on deposit
double principal = 1000.0; // starting principal
double rate = .05; // annual interest rate
unsigned int year; // year counter
// output table column heads
printf( "%4s%21s\n", "Year", "Amount on deposit" );
// calculate amount on deposit for each of ten years
for ( year = 1; year <= 10; ++year ) {
// calculate new amount for specified year
amount = principal * pow( 1.0 + rate, year );
// output one table row
printf( "%4u%21.2f\n", year, amount );
} // end for
} // end function main
Year
1
2
3
4
5
6
7
8
9
10
Amount on deposit
1050.00
1102.50
1157.63
1215.51
1276.28
1340.10
1407.10
1477.46
1551.33
1628.89
Fig. 4.6 | Calculating compound interest.
The for statement executes the body of the loop 10 times, varying a control variable
from 1 to 10 in increments of 1. Although C does not include an exponentiation operator,
we can use the Standard Library function pow for this purpose. The function pow(x, y)
calculates the value of x raised to the yth power. It takes two arguments of type double
and returns a double value. Type double is a floating-point type like float, but typically
a variable of type double can store a value of much greater magnitude with greater precision
than float. The header <math.h> (line 4) should be included whenever a math function
such as pow is used. Actually, this program would malfunction without the inclusion of
math.h, as the linker would be unable to find the pow function.1 Function pow requires
1.
On many Linux/UNIX C compilers, you must include the -lm option (e.g., gcc -lm fig04_06.c)
when compiling Fig. 4.6. This links the math library to the program.
124
Chapter 4 C Program Control
two double arguments, but variable year is an integer. The math.h file includes information that tells the compiler to convert the value of year to a temporary double representation before calling the function. This information is contained in something called pow’s
function prototype. Function prototypes are explained in Chapter 5. We also provide a
summary of the pow function and other math library functions in Chapter 5.
A Caution about Using Type float or double for Monetary Amounts
Notice that we defined the variables amount, principal and rate to be of type double.
We did this for simplicity because we’re dealing with fractional parts of dollars.
Error-Prevention Tip 4.4
Do not use variables of type float or double to perform monetary calculations. The impreciseness of floating-point numbers can cause errors that will result in incorrect monetary values. [In this chapter’s exercises, we explore the use of integers to perform precise
monetary calculations.]
Here is a simple explanation of what can go wrong when using float or double to
represent dollar amounts. Two float dollar amounts stored in the machine could be
14.234 (which with %.2f prints as 14.23) and 18.673 (which with %.2f prints as 18.67).
When these amounts are added, they produce the sum 32.907, which with %.2f prints as
32.91. Thus your printout could appear as
14.23
+ 18.67
32.91
Clearly the sum of the individual numbers as printed should be 32.90! You’ve been
warned!
Formatting Numeric Output
The conversion specifier %21.2f is used to print the value of the variable amount in the program. The 21 in the conversion specifier denotes the field width in which the value will be
printed. A field width of 21 specifies that the value printed will appear in 21 print positions.
The 2 specifies the precision (i.e., the number of decimal positions). If the number of characters displayed is less than the field width, then the value will automatically be right justified
in the field. This is particularly useful for aligning floating-point values with the same precision (so that their decimal points align vertically). To left justify a value in a field, place a
- (minus sign) between the % and the field width. The minus sign may also be used to left
justify integers (such as in %-6d) and character strings (such as in %-8s). We’ll discuss the
powerful formatting capabilities of printf and scanf in detail in Chapter 9.
4.7 switch Multiple-Selection Statement
In Chapter 3, we discussed the if single-selection statement and the if…else doubleselection statement. Occasionally, an algorithm will contain a series of decisions in which a
variable or expression is tested separately for each of the constant integral values it may assume, and different actions are taken. This is called multiple selection. C provides the
switch multiple-selection statement to handle such decision making.
4.7 switch Multiple-Selection Statement
125
The switch statement consists of a series of case labels, an optional default case and
statements to execute for each case. Figure 4.7 uses switch to count the number of each
different letter grade students earned on an exam.
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
// Fig. 4.7: fig04_07.c
// Counting letter grades with switch.
#include <stdio.h>
// function main begins program execution
int main( void )
{
int grade; // one grade
unsigned int aCount = 0; // number of As
unsigned int bCount = 0; // number of Bs
unsigned int cCount = 0; // number of Cs
unsigned int dCount = 0; // number of Ds
unsigned int fCount = 0; // number of Fs
puts( "Enter the letter grades." );
puts( "Enter the EOF character to end input."
);
// loop until user types end-of-file key sequence
while ( ( grade = getchar() ) != EOF ) {
// determine which grade was input
switch ( grade ) { // switch nested in while
case 'A': //
case 'a': //
++aCount;
break; //
grade was uppercase A
or lowercase a
// increment aCount
necessary to exit switch
case 'B': //
case 'b': //
++bCount;
break; //
grade was uppercase B
or lowercase b
// increment bCount
exit switch
case 'C': //
case 'c': //
++cCount;
break; //
grade was uppercase C
or lowercase c
// increment cCount
exit switch
case 'D': //
case 'd': //
++dCount;
break; //
grade was uppercase D
or lowercase d
// increment dCount
exit switch
case 'F': //
case 'f': //
++fCount;
break; //
grade was uppercase F
or lowercase f
// increment fCount
exit switch
Fig. 4.7 | Counting letter grades with switch. (Part 1 of 2.)
126
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
Chapter 4 C Program Control
case '\n': // ignore newlines,
case '\t': // tabs,
case ' ': // and spaces in input
break; // exit switch
default: // catch all other characters
printf( "%s", "Incorrect letter grade entered." );
puts( " Enter a new grade." );
break; // optional; will exit switch anyway
} // end switch
} // end while
// output summary of results
puts( "\nTotals for each letter
printf( "A: %u\n", aCount ); //
printf( "B: %u\n", bCount ); //
printf( "C: %u\n", cCount ); //
printf( "D: %u\n", dCount ); //
printf( "F: %u\n", fCount ); //
} // end function main
Enter the
Enter the
a
b
c
C
A
d
f
C
E
Incorrect
D
A
b
^Z
grade are:" );
display number
display number
display number
display number
display number
of
of
of
of
of
A
B
C
D
F
grades
grades
grades
grades
grades
letter grades.
EOF character to end input.
letter grade entered. Enter a new grade.
Not all systems display a representation of the EOF character
Totals for each letter grade are:
A: 3
B: 2
C: 3
D: 2
F: 1
Fig. 4.7 | Counting letter grades with switch. (Part 2 of 2.)
Reading Character Input
In the program, the user enters letter grades for a class. In the while header (line 19),
while ( ( grade = getchar() ) != EOF )
the parenthesized assignment (grade = getchar()) executes first. The getchar function
(from <stdio.h>) reads one character from the keyboard and stores that character in the
4.7 switch Multiple-Selection Statement
127
integer variable grade. Characters are normally stored in variables of type char. However,
an important feature of C is that characters can be stored in any integer data type because
they’re usually represented as one-byte integers in the computer. Thus, we can treat a character as either an integer or a character, depending on its use. For example, the statement
printf( "The character (%c) has the value %d.\n", 'a', 'a' );
uses the conversion specifiers %c and %d to print the character a and its integer value, respectively. The result is
The character (a) has the value 97.
The integer 97 is the character’s numerical representation in the computer. Many
computers today use the ASCII (American Standard Code for Information Interchange)
character set in which 97 represents the lowercase letter 'a'. A list of the ASCII characters
and their decimal values is presented in Appendix B. Characters can be read with scanf
by using the conversion specifier %c.
Assignments as a whole actually have a value. This value is assigned to the variable on
the left side of the =. The value of the assignment expression grade = getchar() is the character that’s returned by getchar and assigned to the variable grade.
The fact that assignments have values can be useful for setting several variables to the
same value. For example,
a = b = c = 0;
first evaluates the assignment c = 0 (because the = operator associates from right to left).
The variable b is then assigned the value of the assignment c = 0 (which is 0). Then, the
variable a is assigned the value of the assignment b = (c = 0) (which is also 0). In the program, the value of the assignment grade = getchar() is compared with the value of EOF
(a symbol whose acronym stands for “end of file”). We use EOF (which normally has the
value -1) as the sentinel value. The user types a system-dependent keystroke combination
to mean “end of file”—i.e., “I have no more data to enter.” EOF is a symbolic integer constant defined in the <stdio.h> header (we’ll see in Chapter 6 how symbolic constants are
defined). If the value assigned to grade is equal to EOF, the program terminates. We’ve
chosen to represent characters in this program as ints because EOF has an integer value
(again, normally -1).
Portability Tip 4.1
The keystroke combinations for entering EOF (end of file) are system dependent.
Portability Tip 4.2
Testing for the symbolic constant EOF [rather than –1 makes programs more portable. The
C standard states that EOF is a negative integral value (but not necessarily –1). Thus, EOF
could have different values on different systems.
Entering the EOF Indicator
On Linux/UNIX/Mac OS X systems, the EOF indicator is entered by typing
<Ctrl> d
128
Chapter 4 C Program Control
on a line by itself. This notation <Ctrl> d means to press the Enter key and then simultaneously press both the Ctrl key and the d key. On other systems, such as Microsoft Windows, the EOF indicator can be entered by typing
<Ctrl> z
You may also need to press Enter on Windows.
The user enters grades at the keyboard. When the Enter key is pressed, the characters
are read by function getchar one character at a time. If the character entered is not equal
to EOF, the switch statement (line 22–58) is entered.
Statement Details
Keyword switch is followed by the variable name grade in parentheses. This is called the
controlling expression. The value of this expression is compared with each of the case labels. Assume the user has entered the letter C as a grade. C is automatically compared to
each case in the switch. If a match occurs (case 'C':), the statements for that case are
executed. In the case of the letter C, cCount is incremented by 1 (line 36), and the switch
statement is exited immediately with the break statement.
The break statement causes program control to continue with the first statement after
the switch statement. The break statement is used because the cases in a switch statement would otherwise run together. If break is not used anywhere in a switch statement,
then each time a match occurs in the statement, the statements for all the remaining cases
will be executed. (This feature is rarely useful, although it’s perfect for programming
Exercise 4.38—the iterative song The Twelve Days of Christmas!) If no match occurs, the
default case is executed, and an error message is printed.
switch
Statement Flowchart
Each case can have one or more actions. The switch statement is different from all other
control statements in that braces are not required around multiple actions in a case of a
switch. The general switch multiple-selection statement (using a break in each case) is
flowcharted in Fig. 4.8. The flowchart makes it clear that each break statement at the end
of a case causes control to immediately exit the switch statement.
switch
Common Programming Error 4.4
Forgetting a break statement when one is needed in a switch statement is a logic error.
Software Engineering Observation 4.2
Provide a default case in switch statements. Cases not explicitly tested in a switch are
ignored. The default case helps prevent this by focusing you on the need to process
exceptional conditions. Sometimes no default processing is needed.
Good Programming Practice 4.5
Although the case clauses and the default case clause in a switch statement can occur
in any order, it’s common to place the default clause last.
Good Programming Practice 4.6
In a switch statement when the default clause is last, the break statement isn’t required.
You may prefer to include this break for clarity and symmetry with other cases.
4.7 switch Multiple-Selection Statement
case a
true
case a actions(s)
break
case b actions(s)
break
case z actions(s)
break
129
false
case b
true
false
...
case z
true
false
default actions(s)
Fig. 4.8 |
switch
multiple-selection statement with breaks.
Ignoring Newline, Tab and Blank Characters in Input
In the switch statement of Fig. 4.7, the lines
case '\n': // ignore newlines,
case '\t': // tabs,
case ' ': // and spaces in input
break; // exit switch
cause the program to skip newline, tab and blank characters. Reading characters one at a
time can cause some problems. To have the program read the characters, you must send
them to the computer by pressing the Enter key. This causes the newline character to be
placed in the input after the character we wish to process. Often, this newline character
must be specially processed to make the program work correctly. By including the preceding cases in our switch statement, we prevent the error message in the default case from
being printed each time a newline, tab or space is encountered in the input.
Error-Prevention Tip 4.5
Remember to provide processing capabilities for newline (and possibly other white-space)
characters in the input when processing characters one at a time.
Listing several case labels together (such as case 'D': case 'd': in Fig. 4.7) simply
means that the same set of actions is to occur for either of these cases.
130
Chapter 4 C Program Control
Constant Integral Expressions
When using the switch statement, remember that each individual case can test only a
constant integral expression—i.e., any combination of character constants and integer
constants that evaluates to a constant integer value. A character constant can be represented as the specific character in single quotes, such as 'A'. Characters must be enclosed within single quotes to be recognized as character constants—characters in double quotes are
recognized as strings. Integer constants are simply integer values. In our example, we’ve
used character constants. Remember that characters are represented as small integer values.
Notes on Integral Types
Portable languages like C must have flexible data type sizes. Different applications may need
integers of different sizes. C provides several data types to represent integers. In addition to
int and char, C provides types short int (which can be abbreviated as short) and long int
(which can be abbreviated as long). The C standard specifies the minimum range of values
for each integer type, but the actual range may be greater and depends on the implementation. For short ints the minimum range is –32767 to +32767. For most integer calculations, long ints are sufficient. The minimum range of values for long ints is –2147483647
to +2147483647. The range of values for an int greater than or equal to that of a short int
and less than or equal to that of a long int. On many of today’s platforms, ints and long
ints represent the same range of values. The data type signed char can be used to represent
integers in the range –127 to +127 or any of the characters in the computer’s character set.
See section 5.2.4.2 of the C standard document for the complete list of signed and unsigned integer-type ranges.
4.8 do…while Repetition Statement
The do…while repetition statement is similar to the while statement. In the while statement, the loop-continuation condition is tested at the beginning of the loop before the
body of the loop is performed. The do…while statement tests the loop-continuation condition after the loop body is performed. Therefore, the loop body will be executed at least
once. When a do…while terminates, execution continues with the statement after the
while clause. It’s not necessary to use braces in the do…while statement if there’s only
one statement in the body. However, the braces are usually included to avoid confusion
between the while and do…while statements. For example,
while ( condition )
is normally regarded as the header to a while statement. A
around the single-statement body appears as
do…while
with no braces
do
statement
while ( condition );
which can be confusing. The last line—while( condition );—may be misinterpreted as a
while statement containing an empty statement. Thus, to avoid confusion, the do…while
with one statement is often written as follows:
do {
statement
} while ( condition );
4.8 do…while Repetition Statement
131
Good Programming Practice 4.7
To eliminate the potential for ambiguity, you may want to include braces in do…while
statements, even if they’re not necessary.
Common Programming Error 4.5
Infinite loops are caused when the loop-continuation condition in a repetition statement
never becomes false. To prevent this, make sure there’s not a semicolon immediately after
a while or for statement’s header. In a counter-controlled loop, make sure the control
variable is incremented (or decremented) in the loop. In a sentinel-controlled loop, make
sure the sentinel value is eventually input.
Figure 4.9 uses a do…while statement to print the numbers from 1 to 10. The control variable counter is preincremented in the loop-continuation test.
1
2
3
4
5
6
7
8
9
10
11
12
13
1
// Fig. 4.9: fig04_09.c
// Using the do...while repetition statement.
#include <stdio.h>
// function main begins program execution
int main( void )
{
unsigned int counter = 1; // initialize counter
do {
printf( "%u ", counter ); // display counter
} while ( ++counter <= 10 ); // end do...while
} // end function main
2
3
4
5
6
7
8
9
10
Fig. 4.9 | Using the do…while repetition statement.
do…while
Statement Flowchart
Figure 4.10 shows the do…while statement flowchart, which makes it clear that the loopcontinuation condition does not execute until after the action is performed at least once.
action(s)
condition
true
false
Fig. 4.10 | Flowcharting the do…while repetition statement.
132
Chapter 4 C Program Control
4.9 break and continue Statements
The break and continue statements are used to alter the flow of control. Section 4.7
showed how break can be used to terminate a switch statement’s execution. This section
discusses how to use break in a repetition statement.
Statement
The break statement, when executed in a while, for, do…while or switch statement,
causes an immediate exit from that statement. Program execution continues with the next
statement. Common uses of the break statement are to escape early from a loop or to skip
the remainder of a switch statement (as in Fig. 4.7). Figure 4.11 demonstrates the break
statement in a for repetition statement. When the if statement detects that x has become
5, break is executed. This terminates the for statement, and the program continues with
the printf after the for. The loop fully executes only four times.
break
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
// Fig. 4.11: fig04_11.c
// Using the break statement in a for statement.
#include <stdio.h>
// function main begins program execution
int main( void )
{
unsigned int x; // counter
// loop 10 times
for ( x = 1; x <= 10; ++x ) {
// if x is 5, terminate loop
if ( x == 5 ) {
break; // break loop only if x is 5
} // end if
printf( "%u ", x ); // display value of x
} // end for
printf( "\nBroke out of loop at x == %u\n", x );
} // end function main
1 2 3 4
Broke out of loop at x == 5
Fig. 4.11 | Using the break statement in a for statement.
Statement
The continue statement, when executed in a while, for or do…while statement, skips
the remaining statements in the body of that control statement and performs the next iteration of the loop. In while and do…while statements, the loop-continuation test is evaluated immediately after the continue statement is executed. In the for statement, the
increment expression is executed, then the loop-continuation test is evaluated. Earlier, we
said that the while statement could be used in most cases to represent the for statement.
continue
4.9 break and continue Statements
133
The one exception occurs when the increment expression in the while statement follows
the continue statement. In this case, the increment is not executed before the repetitioncontinuation condition is tested, and the while does not execute in the same manner as
the for. Figure 4.12 uses the continue statement in a for statement to skip the printf
statement and begin the next iteration of the loop.
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
// Fig. 4.12: fig04_12.c
// Using the continue statement in a for statement.
#include <stdio.h>
// function main begins program execution
int main( void )
{
unsigned int x; // counter
// loop 10 times
for ( x = 1; x <= 10; ++x ) {
// if x is 5, continue with next iteration of loop
if ( x == 5 ) {
continue; // skip remaining code in loop body
} // end if
printf( "%u ", x ); // display value of x
} // end for
puts( "\nUsed continue to skip printing the value 5" );
} // end function main
1 2 3 4 6 7 8 9 10
Used continue to skip printing the value 5
Fig. 4.12 | Using the continue statement in a for statement.
Software Engineering Observation 4.3
Some programmers feel that break and continue violate the norms of structured
programming. The effects of these statements can be achieved by structured programming
techniques we’ll soon learn, so these programmers do not use break and continue.
Performance Tip 4.1
The break and continue statements, when used properly, perform faster than the corresponding structured techniques that we’ll soon learn.
Software Engineering Observation 4.4
There’s a tension between achieving quality software engineering and achieving the bestperforming software. Often one of these goals is achieved at the expense of the other. For
all but the most performance-intensive situations, apply the following guidelines: First,
make your code simple and correct; then make it fast and small, but only if necessary.
134
Chapter 4 C Program Control
4.10 Logical Operators
So far we’ve studied only simple conditions, such as counter <= 10, total > 1000, and number != sentinelValue. We’ve expressed these conditions in terms of the relational operators,
>, <, >=
and <=, and the equality operators, == and !=. Each decision tested precisely one condition. To test multiple conditions in the process of making a decision, we had to perform
these tests in separate statements or in nested if or if…else statements. C provides logical
operators that may be used to form more complex conditions by combining simple conditions. The logical operators are && (logical AND), || (logical OR) and ! (logical NOT
also called logical negation). We’ll consider examples of each of these operators.
Logical AND (&&) Operator
Suppose we wish to ensure that two conditions are both true before we choose a certain
path of execution. In this case, we can use the logical operator && as follows:
if ( gender == 1 && age >= 65 )
++seniorFemales;
This if statement contains two simple conditions. The condition gender == 1 might be evaluated, for example, to determine whether a person is a female. The condition age >= 65 is
evaluated to determine whether a person is a senior citizen. The two simple conditions are
evaluated first because == and >= are have higher precedence than &&. The if statement then
considers the combined condition gender == 1 && age >= 65, which is true if and only if both
of the simple conditions are true. Finally, if this combined condition is true, then the count
of seniorFemales is incremented by 1. If either or both of the simple conditions are false, then
the program skips the incrementing and proceeds to the statement following the if.
Figure 4.13 summarizes the && operator. The table shows all four possible combinations of zero (false) and nonzero (true) values for expression1 and expression2. Such
tables are often called truth tables. C evaluates all expressions that include relational operators, equality operators, and/or logical operators to 0 or 1. Although C sets a true value to 1,
it accepts any nonzero value as true.
expression1
expression2
expression1 && expression2
0
0
nonzero
nonzero
0
nonzero
0
nonzero
0
0
0
1
Fig. 4.13 | Truth table for the logical AND (&&) operator.
Logical OR (||) Operator
Now let’s consider the || (logical OR) operator. Suppose we wish to ensure at some point
in a program that either or both of two conditions are true before we choose a certain path
of execution. In this case, we use the || operator, as in the following program segment:
if ( semesterAverage >= 90 || finalExam >= 90 )
puts( "Student grade is A" );
4.10 Logical Operators
135
This statement also contains two simple conditions. The condition semesterAverage >=
90 is evaluated to determine whether the student deserves an “A” in the course because of
a solid performance throughout the semester. The condition finalExam >= 90 is evaluated
to determine whether the student deserves an “A” in the course because of an outstanding
performance on the final exam. The if statement then considers the combined condition
semesterAverage >= 90 || finalExam >= 90
and awards the student an “A” if either or both of the simple conditions are true. The message “Student grade is A” is not printed only when both of the simple conditions are false
(zero). Figure 4.14 is a truth table for the logical OR operator (||).
expression1
expression2
expression1 || expression2
0
0
nonzero
nonzero
0
nonzero
0
nonzero
0
1
1
1
Fig. 4.14 | Truth table for the logical OR (||) operator.
The && operator has a higher precedence than ||. Both operators associate from left to
right. An expression containing && or || operators is evaluated only until truth or falsehood
is known. Thus, evaluation of the condition
gender == 1 && age >= 65
will stop if gender is not equal to 1 (i.e., the entire expression is false), and continue if gender is equal to 1 (i.e., the entire expression could still be true if age >= 65). This performance feature for the evaluation of logical AND and logical OR expressions is called
short-circuit evaluation.
Performance Tip 4.2
In expressions using operator &&, make the condition that’s most likely to be false the leftmost condition. In expressions using operator ||, make the condition that’s most likely to
be true the leftmost condition. This can reduce a program’s execution time.
Logical Negation (!) Operator
C provides ! (logical negation) to enable you to “reverse” the meaning of a condition. Unlike operators && and ||, which combine two conditions (and are therefore binary operators), the logical negation operator has only a single condition as an operand (and is
therefore a unary operator). The logical negation operator is placed before a condition
when we’re interested in choosing a path of execution if the original condition (without
the logical negation operator) is false, such as in the following program segment:
if ( !( grade == sentinelValue ) )
printf( "The next grade is %f\n", grade );
The parentheses around the condition grade == sentinelValue are needed because the
logical negation operator has a higher precedence than the equality operator. Figure 4.15
is a truth table for the logical negation operator.
136
Chapter 4 C Program Control
expression
!expression
0
nonzero
1
0
Fig. 4.15 | Truth table for operator ! (logical negation).
In most cases, you can avoid using logical negation by expressing the condition differently with an appropriate relational operator. For example, the preceding statement
may also be written as follows:
if ( grade != sentinelValue )
printf( "The next grade is %f\n", grade );
Summary of Operator Precedence and Associativity
Figure 4.16 shows the precedence and associativity of the operators introduced to this
point. The operators are shown from top to bottom in decreasing order of precedence.
Operators
++
(postfix)
+
-
!
*
/
%
+
-
<
<=
==
!=
(postfix)
++ (prefix)
--
>
>=
-=
*=
&&
||
?:
=
,
+=
/=
%=
--
(prefix)
(type)
Associativity
Type
right to left
right to left
left to right
left to right
left to right
left to right
left to right
left to right
right to left
right to left
left to right
postfix
unary
multiplicative
additive
relational
equality
logical AND
logical OR
conditional
assignment
comma
Fig. 4.16 | Operator precedence and associativity.
The _Bool Data Type
The C standard includes a boolean type—represented by the keyword _Bool—which can
hold only the values 0 or 1. Recall C’s convention of using zero and nonzero values to represent false and true—the value 0 in a condition evaluates to false, while any nonzero value
evaluates to true. Assigning any non-zero value to a _Bool sets it to 1. The standard also
includes the <stdbool.h> header, which defines bool as a shorthand for the type _Bool,
and true and false as named representations of 1 and 0, respectively. At preprocessor
time, bool, true and false are replaced with _Bool, 1 and 0. Section F.8 presents an example that uses bool, true and false. The example uses a programmer-defined function,
a concept we introduce in Chapter 5. You can study the example now, but might wish to
revisit it after reading Chapter 5. Microsoft Visual C++ does not implement the _Bool data
type.
4.11 Confusing Equality (==) and Assignment (=) Operators
137
4.11 Confusing Equality (==) and Assignment (=)
Operators
There’s one type of error that C programmers, no matter how experienced, tend to make
so frequently that we felt it was worth a separate section. That error is accidentally swapping the operators == (equality) and = (assignment). What makes these swaps so damaging
is the fact that they do not ordinarily cause compilation errors. Rather, statements with these
errors ordinarily compile correctly, allowing programs to run to completion while likely
generating incorrect results through runtime logic errors.
Two aspects of C cause these problems. One is that any expression in C that produces
a value can be used in the decision portion of any control statement. If the value is 0, it’s
treated as false, and if the value is nonzero, it’s treated as true. The second is that assignments in C produce a value, namely the value that’s assigned to the variable on the left side
of the assignment operator. For example, suppose we intend to write
if ( payCode == 4 )
printf( "%s", "You get a bonus!" );
but we accidentally write
if ( payCode = 4 )
printf( "%s", "You get a bonus!" );
The first if statement properly awards a bonus to the person whose paycode is equal to 4.
The second if statement—the one with the error—evaluates the assignment expression in
the if condition. This expression is a simple assignment whose value is the constant 4. Because any nonzero value is interpreted as “true,” the condition in this if statement is always true, and not only is the value of payCode inadvertantly set to 4, but the person always
receives a bonus regardless of what the actual paycode is!
Common Programming Error 4.6
Using operator == for assignment or using operator = for equality is a logic error.
lvalues and rvalues
You’ll probably be inclined to write conditions such as x == 7 with the variable name on
the left and the constant on the right. By reversing these terms so that the constant is on
the left and the variable name is on the right, as in 7 == x, then if you accidentally replace
the == operator with =, you’ll be protected by the compiler. The compiler will treat this as
a syntax error, because only a variable name can be placed on the left-hand side of an assignment expression. This will prevent the potential devastation of a runtime logic error.
Variable names are said to be lvalues (for “left values”) because they can be used on
the left side of an assignment operator. Constants are said to be rvalues (for “right values”)
because they can be used on only the right side of an assignment operator. lvalues can also
be used as rvalues, but not vice versa.
Error-Prevention Tip 4.6
When an equality expression has a variable and a constant, as in x == 1, you may prefer to
write it with the constant on the left and the variable name on the right (e.g., 1 == x as protection against the logic error that occurs when you accidentally replace operator == with =).
138
Chapter 4 C Program Control
Confusing == and = in Standalone Statements
The other side of the coin can be equally unpleasant. Suppose you want to assign a value
to a variable with a simple statement such as
x = 1;
but instead write
x == 1;
Here, too, this is not a syntax error. Rather the compiler simply evaluates the conditional
expression. If x is equal to 1, the condition is true and the expression returns the value 1.
If x is not equal to 1, the condition is false and the expression returns the value 0. Regardless of what value is returned, there’s no assignment operator, so the value is simply lost,
and the value of x remains unaltered, probably causing an execution-time logic error. Unfortunately, we do not have a handy trick available to help you with this problem! Many
compilers, however, will issue a warning on such a statement.
Error-Prevention Tip 4.7
After you write a program, text search it for every = and check that it’s used properly.
4.12 Structured Programming Summary
Just as architects design buildings by employing the collective wisdom of their profession,
so should programmers design programs. Our field is younger than architecture is, and our
collective wisdom is considerably sparser. We’ve learned a great deal in a mere six decades.
Perhaps most important, we’ve learned that structured programming produces programs
that are easier (than unstructured programs) to understand and therefore are easier to test,
debug, modify, and even prove correct in a mathematical sense.
Chapters 3 and 4 have concentrated on C’s control statements. Each statement has
been presented, flowcharted and discussed separately with examples. Now, we summarize
the results of Chapters 3 and 4 and introduce a simple set of rules for the formation and
properties of structured programs.
Figure 4.17 summarizes the control statements discussed in Chapters 3 and 4. Small
circles are used in the figure to indicate the single entry point and the single exit point of each
statement. Connecting individual flowchart symbols arbitrarily can lead to unstructured
programs. Therefore, the programming profession has chosen to combine flowchart symbols to form a limited set of control statements, and to build only structured programs by
properly combining control statements in two simple ways. For simplicity, only singleentry/single-exit control statements are used—there’s only one way to enter and only one
way to exit each control statement. Connecting control statements in sequence to form
structured programs is simple—the exit point of one control statement is connected
directly to the entry point of the next—i.e., the control statements are simply placed one
after another in a program—we’ve called this “control-statement stacking.” The rules for
forming structured programs also allow for control statements to be nested.
Figure 4.18 shows the rules for forming structured programs. The rules assume that
the rectangle flowchart symbol may be used to indicate any action including input/output.
Figure 4.19 shows the simplest flowchart.
4.12 Structured Programming Summary
Sequence
Selection
if statement
(single selection)
if...else statement
(double selection)
T
F
T
F
...
switch statement
(multiple selection)
T
break
F
T
break
T
break
F
...
F
Repetition
while statement
do...while statement
for statement
T
F
T
F
T
F
Fig. 4.17 | C’s single-entry/single-exit sequence, selection and repetition statements.
139
140
Chapter 4 C Program Control
Rules for forming structured programs
1) Begin with the “simplest flowchart” (Fig. 4.19).
2) Any rectangle (action) can be replaced by two rectangles
(actions) in sequence.
3) Any rectangle (action) can be replaced by any control statement
(sequence, if, if…else, switch, while, do…while or for).
4) Rules 2 and 3 may be applied as often as you like and in any
order.
Fig. 4.18 | Rules for forming structured programs.
Fig. 4.19 | Simplest flowchart.
Applying the rules of Fig. 4.18 always results in a structured flowchart with a neat,
building-block appearance. Repeatedly applying Rule 2 to the simplest flowchart (Fig. 4.19)
results in a structured flowchart containing many rectangles in sequence (Fig. 4.20). Rule 2
generates a stack of control statements; so we call Rule 2 the stacking rule.
Rule 3 is called the nesting rule. Repeatedly applying Rule 3 to the simplest flowchart
results in a flowchart with neatly nested control statements. For example, in Fig. 4.21, the
rectangle in the simplest flowchart is first replaced with a double-selection (if…else)
statement. Then Rule 3 is applied again to both of the rectangles in the double-selection
statement, replacing each of these rectangles with double-selection statements. The dashed
box around each of the double-selection statements represents the rectangle that was
replaced in the original flowchart.
Rule 4 generates larger, more involved, and more deeply nested structures. The flowcharts that emerge from applying the rules in Fig. 4.18 constitute the set of all possible
structured flowcharts and hence the set of all possible structured programs.
It’s because of the elimination of the goto statement that these building blocks never
overlap one another. The beauty of the structured approach is that we use only a small
number of simple single-entry/single-exit pieces, and we assemble them in only two simple
ways. Figure 4.22 shows the kinds of stacked building blocks that emerge from applying
Rule 2 and the kinds of nested building blocks that emerge from applying Rule 3. The
figure also shows the kind of overlapped building blocks that cannot appear in structured
flowcharts (because of the elimination of the goto statement).
4.12 Structured Programming Summary
Rule 2
Rule 2
141
Rule 2
...
Fig. 4.20 | Repeatedly applying Rule 2 of Fig. 4.18 to the simplest flowchart.
Rule 3
Rule 3
Rule 3
Fig. 4.21 | Applying Rule 3 of Fig. 4.18 to the simplest flowchart.
142
Chapter 4 C Program Control
Stacked building blocks
Nested building blocks
Overlapping building blocks
(Illegal in structured programs)
Fig. 4.22 | Stacked, nested and overlapped building blocks.
If the rules in Fig. 4.18 are followed, an unstructured flowchart (such as that in
Fig. 4.23) cannot be created. If you’re uncertain whether a particular flowchart is structured, apply the rules of Fig. 4.18 in reverse to try to reduce the flowchart to the simplest
flowchart. If you succeed, the original flowchart is structured; otherwise, it’s not.
Fig. 4.23 | An unstructured flowchart.
Structured programming promotes simplicity. Bohm and Jacopini showed that only
three forms of control are needed:
•
Sequence
•
Selection
•
Repetition
4.13 Secure C Programming
143
Sequence is straightforward. Selection is implemented in one of three ways:
statement (single selection)
•
if
•
if…else
•
switch
statement (double selection)
statement (multiple selection)
It’s straightforward to prove that the simple if statement is sufficient to provide any form
of selection—everything that can be done with the if…else statement and the switch
statement can be implemented with one or more if statements.
Repetition is implemented in one of three ways:
statement
•
while
•
do…while
•
for
statement
statement
It’s straightforward to prove that the while statement is sufficient to provide any form
of repetition. Everything that can be done with the do…while statement and the for
statement can be done with the while statement.
Combining these results illustrates that any form of control ever needed in a C program can be expressed in terms of only three forms of control:
•
sequence
•
if
•
while
statement (selection)
statement (repetition)
And these control statements can be combined in only two ways—stacking and
nesting. Indeed, structured programming promotes simplicity.
In Chapters 3 and 4, we discussed how to compose programs from control statements
containing actions and decisions. In Chapter 5, we introduce another program structuring
unit called the function. We’ll learn to compose large programs by combining functions,
which, in turn, are composed of control statements. We’ll also discuss how using functions
promotes software reusability.
4.13 Secure C Programming
Checking Function scanf’s Return Value
Figure 4.6 used the math library function pow, which calculates the value of its first argument raised to the power of its second argument and returns the result as a double value.
The calculation’s result was then used in the statement that called pow.
Many functions return values indicating whether they executed successfully. For
example, function scanf returns an int indicating whether the input operation was successful. If an input failure occurs, scanf returns the value EOF (defined in <stdio.h>); otherwise, it returns the number of items that were read. If this value does not match the number
you intended to read, then scanf was unable to complete the input operation.
Consider the following statement from Fig. 3.6
scanf( "%d", &grade ); // read grade from user
which expects to read one int value. If the user enters an integer, scanf returns 1 indicating that one value was indeed read. If the user enters a string, such as "hello", scanf re-
144
Chapter 4 C Program Control
turns 0 indicating that it was unable to read the input as an integer. In this case, the
variable grade does not receive a value.
Function scanf can read multiple inputs, as in
scanf( "%d%d", &number1, &number2 ); // read two integers
If the input is successful, scanf will return 2 indicating that two values were read. If the
user enters a string for the first value, scanf will return 0 and neither number1 nor number2
will receive values. If the user enters an integer followed by a string, scanf will return 1
and only number1 will receive a value.
To make your input processing more robust, check scanf’s return value to ensure that
the number of inputs read matches the number of inputs expected. Otherwise, your program will use the values of the variables as if scanf completed successfully. This could lead
to logic errors, program crashes or even attacks.
Range Checking
Even if a scanf operates successfully, the values read might still be invalid. For example,
grades are typically integers in the range 0–100. In a program that inputs such grades, you
should validate the grades by using range checking to ensure that they are values from 0
to 100. You can then ask the user to reenter any value that’s out of range. If a program
requires inputs from a specific set of values (e.g., non-sequential product codes), you can
ensure that each input matches a value in the set. For more information, see Chapter 5,
“Integer Security” of Robert Seacord’s book Secure Coding in C and C++.
Summary
Section 4.2 Repetition Essentials
• Most programs involve repetition, or looping. A loop is a group of instructions the computer executes repeatedly while some loop-continuation condition remains true.
• Counter-controlled repetition is sometimes called definite repetition because we know in advance exactly how many times the loop will execute.
• Sentinel-controlled repetition is sometimes called indefinite repetition because it’s not known in
advance how many times the loop will execute; the loop includes statements that obtain data each
time the loop is performed.
• In counter-controlled repetition, a control variable is used to count the number of repetitions.
The control variable is incremented (or decremented) each time the group of instructions is performed. When the correct number of repetitions has been performed, the loop terminates, and
the program resumes execution with the statement after the repetition statement.
• The sentinel value indicates “end of data.” The sentinel is entered after all regular data items have
been supplied to the program. Sentinels must be distinct from regular data items.
Section 4.3 Counter-Controlled Repetition
• Counter-controlled repetition requires the name of a control variable (or loop counter), the initial value of the control variable, the increment (or decrement) by which the control variable is
modified each time through the loop, and the condition that tests for the final value of the control variable (i.e., whether looping should continue).
Section 4.4 for Repetition Statement
• The for repetition statement handles all the details of counter-controlled repetition.
Summary
145
• When the for statement begins executing, its control variable is initialized. Then, the loop-continuation condition is checked. If the condition is true, the loop’s body executes. The control
variable is then incremented, and the loop begins again with the loop-continuation condition.
This process continues until the loop-continuation condition fails.
• The general format of the for statement is
expression1; expression2; expression3
statement
for (
)
where expression1 initializes the loop-control variable, expression2 is the loop-continuation condition, and expression3 increments the control variable.
• In most cases, the for statement can be represented with an equivalent while statement as in:
expression1;
while ( expression2 )
statement
expression3;
{
}
• The comma operator guarantees that lists of expressions evaluate from left to right. The value of
the entire expression is that of the rightmost expression.
• The three expressions in the for statement are optional. If expression2 is omitted, C assumes that
the condition is true, thus creating an infinite loop. One might omit expression1 if the control
variable is initialized elsewhere in the program. expression3 might be omitted if the increment is
calculated by statements in the body of the for statement or if no increment is needed.
• The increment expression in the for statement acts like a stand-alone C statement at the end of
the body of the for.
• The two semicolons in the for statement are required.
Section 4.5 for Statement: Notes and Observations
• The initialization, loop-continuation condition and increment can contain arithmetic expressions.
• The “increment” may be negative (in which case it’s really a decrement and the loop actually
counts downward).
• If the loop-continuation condition is initially false, the body portion of the loop isn’t performed.
Instead, execution proceeds with the statement following the for statement.
Section 4.6 Examples Using the for Statement
• Function pow performs exponentiation. The function pow(x, y) calculates the value of x raised
to the yth power. It takes two arguments of type double and returns a double value.
• Type double is a floating-point type much like float, but typically a variable of type double can
store a value of much greater magnitude with greater precision than float.
• The header <math.h> should be included whenever a math function such as pow is used.
• The conversion specifier %21.2f denotes that a floating-point value will be displayed right justified in a field of 21 characters with two digits to the right of the decimal point.
• To left justify a value in a field, place a - (minus sign) between the % and the field width.
Section 4.7 switch Multiple-Selection Statement
• Occasionally, an algorithm will contain a series of decisions in which a variable or expression is
tested separately for each of the constant integral values it may assume, and different actions are
taken. This is called multiple selection. C provides the switch statement to handle this.
146
Chapter 4 C Program Control
• The switch statement consists of a series of case labels, an optional default case and statements
to execute for each case.
• The getchar function (from the standard input/output library) reads and returns one character
from the keyboard.
• Characters are normally stored in variables of type char. Characters can be stored in any integer
data type because they’re usually represented as one-byte integers in the computer. Thus, we can
treat a character as either an integer or a character, depending on its use.
• Many computers today use the ASCII (American Standard Code for Information Interchange)
character set in which 97 represents the lowercase letter 'a'.
• Characters can be read with scanf by using the conversion specifier %c.
• Assignment expressions as a whole actually have a value. This value is assigned to the variable on
the left side of the =.
• The fact that assignment statements have values can be useful for setting several variables to the
same value, as in a = b = c = 0;.
•
EOF
is often used as a sentinel value. EOF is a symbolic integer constant defined in <stdio.h>.
• On Linux/UNIX systems and many others, the EOF indicator is entered by typing <Ctrl> d . On
other systems, such as Microsoft Windows, the EOF indicator can be entered by typing <Ctrl> z.
• Keyword switch is followed by the controlling expression in parentheses. The value of this expression is compared with each of the case labels. If a match occurs, the statements for that case
execute. If no match occurs, the default case executes.
• The break statement causes program control to continue with the statement after the
The break statement prevents the cases in a switch statement from running together.
switch.
• Each case can have one or more actions. The switch statement is different from all other control
statements in that braces are not required around multiple actions in a case of a switch.
• Listing several case labels together simply means that the same set of actions is to occur for any
of these cases.
• Remember that the switch statement can be used only for testing a constant integral expression—i.e., any combination of character constants and integer constants that evaluates to a constant integer value. A character constant can be represented as the specific character in single
quotes, such as 'A'. Characters must be enclosed within single quotes to be recognized as character constants. Integer constants are simply integer values.
• C provides several data types to represent integers. In addition to int and char, C provides types
short int (which can be abbreviated as short) and long int (which can be abbreviated as long).
The C standard specifies the minimum range of values for each integer type, but the actual range
may be greater and depends on the implementation. For short ints the minimum range is –32767
to +32767. The minimum range of values for long ints is –2147483647 to +2147483647. The
range of values for an int greater than or equal to that of a short int and less than or equal to that
of a long int. On many of today’s platforms, ints and long ints represent the same range of values.
The data type signed char can be used to represent integers in the range –127 to +127 or any of
the characters in the computer’s character set. See section 5.2.4.2 of the C standard document for
the complete list of signed and unsigned integer-type ranges.
Section 4.8 do…while Repetition Statement
• The do…while statement tests the loop-continuation condition after the loop body is performed.
Therefore, the loop body will be executed at least once. When a do…while terminates, execution
continues with the statement after the while clause.
Terminology
147
Section 4.9 break and continue Statements
• The break statement, when executed in a while, for, do…while or switch statement, causes immediate exit from that statement. Program execution continues with the next statement.
• The continue statement, when executed in a while, for or do…while statement, skips the remaining statements in the body of that control statement and performs the next iteration of the
loop. In while and do…while statements, the loop-continuation test is evaluated immediately
after the continue statement is executed. In the for statement, the increment expression is executed, then the loop-continuation test is evaluated.
Section 4.10 Logical Operators
• Logical operators may be used to form complex conditions by combining simple conditions. The
logical operators are && (logical AND), || (logical OR) and ! (logical NOT, or logical negation).
• A condition containing the && (logical AND) operator is true if and only if both of the simple
conditions are true.
• C evaluates all expressions that include relational operators, equality operators, and/or logical operators to 0 or 1. Although C sets a true value to 1, it accepts any nonzero value as true.
• A condition containing the || (logical OR) operator is true if either or both of the simple conditions are true.
• The && operator has a higher precedence than ||. Both operators associate from left to right.
• An expression containing && or || operators is evaluated only until truth or falsehood is known.
• C provides ! (logical negation) to enable you to “reverse” the meaning of a condition. Unlike the
binary operators && and ||, which combine two conditions, the unary logical negation operator
has only a single condition as an operand.
• The logical negation operator is placed before a condition when we’re interested in choosing a
path of execution if the original condition (without the logical negation operator) is false.
• In most cases, you can avoid using logical negation by expressing the condition differently with
an appropriate relational operator.
Section 4.11 Confusing Equality (==) and Assignment (=) Operators
• Programmers often accidentally swap the operators == (equality) and = (assignment). What
makes these swaps so damaging is that they do not ordinarily cause syntax errors. Rather, statements with these errors ordinarily compile correctly, allowing programs to run to completion
while likely generating incorrect results through runtime logic errors.
• You may be inclined to write conditions such as x == 7 with the variable name on the left and the
constant on the right. By reversing these terms so that the constant is on the left and the variable
name is on the right, as in 7 == x, then if you accidentally replace the == operator with =, you’ll
be protected by the compiler. The compiler will treat this as a syntax error, because only a variable name can be placed on the left-hand side of an assignment statement.
• Variable names are said to be lvalues (for “left values”) because they can be used on the left side
of an assignment operator.
• Constants are said to be rvalues (for “right values”) because they can be used only on the right
side of an assignment operator. lvalues can also be used as rvalues, but not vice versa.
Terminology
ASCII (American Standard Code for Information Interchange) character set 127
case label 128
char primitive type 127
comma operator 119
constant integral expression 130
148
Chapter 4 C Program Control
control variable 115
controlling expression in a switch 128
decrement a control variable 116
definite repetition 115
final value of a control variable 116
function prototype 124
increment a control variable 116
indefinite repetition 115
initial value of a control variable 116
logical AND operator (&&) 134
logical negation operator (!) 134
logical OR operator (||) 134
logical NOT operator (!) 134
loop-continuation condition 115
lvalue (“left value”) 137
<math.h> header 124
name of a control variable 116
nesting rule 140
off-by-one error 118
pow (power) function 124
rvalue (“right value”) 137
short-circuit evaluation 135
stacking rule 140
truth table 134
Self-Review Exercises
4.1
Fill in the blanks in each of the following statements.
repetition because it’s known
a) Counter-controlled repetition is also known as
in advance how many times the loop will be executed.
repetition because it’s not
b) Sentinel-controlled repetition is also known as
known in advance how many times the loop will be executed.
c) In counter-controlled repetition, a(n)
is used to count the number of times a
group of instructions should be repeated.
statement, when executed in a repetition statement, causes the next itd) The
eration of the loop to be performed immediately.
statement, when executed in a repetition statement or a switch, causes
e) The
an immediate exit from the statement.
f) The
is used to test a particular variable or expression for each of the constant
integral values it may assume.
4.2
State whether the following are true or false. If the answer is false, explain why.
a) The default case is required in the switch selection statement.
b) The break statement is required in the default case of a switch selection statement.
c) The expression (x > y && a < b) is true if either x > y is true or a < b is true.
d) An expression containing the || operator is true if either or both of its operands is true.
4.3
Write a statement or a set of statements to accomplish each of the following tasks:
a) Sum the odd integers between 1 and 99 using a for statement. Assume the integer variables sum and count have been defined.
b) Print the value 333.546372 in a field width of 15 characters with precisions of 1, 2, 3, 4
and 5. Left justify the output. What are the five values that print?
c) Calculate the value of 2.5 raised to the power of 3 using the pow function. Print the result with a precision of 2 in a field width of 10 positions. What is the value that prints?
d) Print the integers from 1 to 20 using a while loop and the counter variable x. Assume
that the variable x has been defined, but not initialized. Print only five integers per line.
[Hint: Use the calculation x % 5. When the value of this is 0, print a newline character,
otherwise print a tab character.]
e) Repeat Exercise 4.3(d) using a for statement.
4.4
Find the error in each of the following code segments and explain how to correct it.
a) x = 1;
while ( x <= 10 );
++x;
}
Answers to Self-Review Exercises
b)
for ( y = .1; y != 1.0; y += .1 )
c)
switch ( n ) {
149
printf( "%f\n", y );
case 1:
puts( "The number is 1" );
case 2:
puts( "The number is 2" );
break;
default:
puts( "The number is not 1 or 2" );
break;
}
d) The following code should print the values 1 to 10.
n = 1;
while ( n < 10 )
printf( "ud ", n++ );
Answers to Self-Review Exercises
4.1
a) definite. b) indefinite. c) control variable or counter. d) continue. e) break. f) switch
selection statement.
4.2
a) False. The default case is optional. If no default action is needed, then there’s no need
for a default case.
b) False. The break statement is used to exit the switch statement. The break statement
is not required in any case.
c) False. Both of the relational expressions must be true in order for the entire expression
to be true when using the && operator.
d) True.
4.3
a)
sum = 0;
for ( count = 1; count <= 99; count += 2 ) {
sum += count;
}
b)
printf( "%-15.1f\n", 333.546372 ); // prints 333.5
printf( "%-15.2f\n", 333.546372 ); // prints 333.55
printf( "%-15.3f\n", 333.546372 ); // prints 333.546
printf( "%-15.4f\n", 333.546372 ); // prints 333.5464
printf( "%-15.5f\n", 333.546372 ); // prints 333.54637
c)
d)
printf( "%10.2f\n", pow( 2.5, 3 ) ); // prints 15.63
x = 1;
while ( x <= 20 ) {
printf( "%d", x );
if ( x % 5 == 0 ) {
puts( "" );
}
else {
printf( "%s", "\t" );
}
++x;
}
150
Chapter 4 C Program Control
or
x = 1;
while ( x <= 20 ) {
if ( x % 5 == 0 ) {
printf( "%u\n", x++ );
}
else {
printf( "%u\t", x++ );
}
}
or
e)
x = 0;
while ( ++x <= 20 ) {
if ( x % 5 == 0 ) {
printf( "%u\n", x );
}
else {
printf( "%u\t", x );
}
}
for ( x = 1; x <= 20; ++x ) {
printf( "%u", x );
if ( x % 5 == 0 ) {
puts( "" );
}
else {
printf( "%s", "\t" );
}
}
or
for ( x = 1; x <= 20; ++x ) {
if ( x % 5 == 0 ) {
printf( "%u\n", x );
}
else {
printf( "%u\t", x );
}
}
4.4
a) Error: The semicolon after the while header causes an infinite loop.
Correction: Replace the semicolon with a { or remove both the ; and the }.
b) Error: Using a floating-point number to control a for repetition statement.
Correction: Use an integer, and perform the proper calculation to get the values you desire.
for ( y = 1; y != 10; ++y )
printf( "%f\n", ( float ) y / 10 );
c) Error: Missing break statement in the statements for the first case.
Correction: Add a break statement at the end of the statements for the first case. This
is not necessarily an error if you want the statement of case 2: to execute every time the
case 1: statement executes.
Exercises
151
d) Error: Improper relational operator used in the while repetition-continuation condition.
Correction: Use <= rather than <.
Exercises
4.5
Find the error in each of the following. (Note: There may be more than one error.)
a) For ( x = 100, x >= 1, ++x )
printf( "%d\n", x );
b) The following code should print whether a given integer is odd or even:
switch ( value % 2 ) {
case 0:
puts( "Even integer" );
case 1:
puts( "Odd integer" );
}
c) The following code should input an integer and a character and print them. Assume the
user types as input 100 A.
scanf( "%d", &intVal );
charVal = getchar();
printf( "Integer: %d\nCharacter: %c\n", intVal, charVal );
d)
for ( x = .000001; x == .0001; x += .000001 ) {
printf( "%.7f\n", x );
}
e) The following code should output the odd integers from 999 to 1:
for ( x = 999; x >= 1; x += 2 ) {
printf( "%d\n", x );
}
f) The following code should output the even integers from 2 to 100:
counter = 2;
Do {
if ( counter % 2 == 0 ) {
printf( "%u\n", counter );
}
counter += 2;
} While ( counter < 100 );
g) The following code should sum the integers from 100 to 150 (assume total is initialized to 0):
for ( x = 100; x <= 150; ++x ); {
total += x;
}
4.6
State which values of the control variable x are printed by each of the following for statements:
a) for ( x = 2; x <= 13; x += 2 ) {
printf( "%u\n", x );
}
b)
for ( x = 5; x <= 22; x += 7 ) {
printf( "%u\n", x );
}
c)
for ( x = 3; x <= 15; x += 3 ) {
printf( "%u\n", x );
}
152
Chapter 4 C Program Control
d)
for ( x = 1; x <= 5; x += 7 ) {
printf( "%u\n", x );
}
e)
for ( x = 12; x >= 2; x -= 3 ) {
printf( "%d\n", x );
}
4.7
Write for statements that print the following sequences of values:
a) 1, 2, 3, 4, 5, 6, 7
b) 3, 8, 13, 18, 23
c) 20, 14, 8, 2, –4, –10
d) 19, 27, 35, 43, 51
4.8
What does the following program do?
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
#include <stdio.h>
// function main begins program execution
int main( void )
{
unsigned int x;
unsigned int y;
unsigned int i;
unsigned int j;
// prompt user for input
printf( "%s", "Enter two unsigned integers in the range 1-20: " );
scanf( "%u%u", &x, &y ); // read values for x and y
for ( i = 1; i <= y; ++i ) { // count from 1 to y
for ( j = 1; j <= x; ++j ) { // count from 1 to x
printf( "%s", "@" ); // output @
} // end inner for
puts( "" ); // begin new line
} // end outer for
} // end function main
4.9
(Sum a Sequence of Integers) Write a program that sums a sequence of integers. Assume that
the first integer read with scanf specifies the number of values remaining to be entered. Your program should read only one value each time scanf is executed. A typical input sequence might be
5 100 200 300 400 500
where the 5 indicates that the subsequent five values are to be summed.
4.10 (Average a Sequence of Integers) Write a program that calculates and prints the average of
several integers. Assume the last value read with scanf is the sentinel 9999. A typical input sequence
might be
10 8 11 7 9 9999
indicating that the average of all the values preceding 9999 is to be calculated.
4.11 (Find the Smallest) Write a program that finds the smallest of several integers. Assume that
the first value read specifies the number of values remaining.
4.12 (Calculating the Sum of Even Integers) Write a program that calculates and prints the sum
of the even integers from 2 to 30.
Exercises
153
4.13 (Calculating the Product of Odd Integers) Write a program that calculates and prints the
product of the odd integers from 1 to 15.
4.14 (Factorials) The factorial function is used frequently in probability problems. The factorial of
a positive integer n (written n! and pronounced “n factorial”) is equal to the product of the positive
integers from 1 to n. Write a program that evaluates the factorials of the integers from 1 to 5. Print the
results in tabular format. What difficulty might prevent you from calculating the factorial of 20?
4.15 (Modified Compound-Interest Program) Modify the compound-interest program of
Section 4.6 to repeat its steps for interest rates of 5%, 6%, 7%, 8%, 9%, and 10%. Use a for loop
to vary the interest rate.
4.16 (Triangle-Printing Program) Write a program that prints the following patterns separately,
one below the other. Use for loops to generate the patterns. All asterisks (*) should be printed by a
single printf statement of the form printf( "%s", "*" ); (this causes the asterisks to print side by
side). [Hint: The last two patterns require that each line begin with an appropriate number of blanks.]
(A)
*
**
***
****
*****
******
*******
********
*********
**********
(B)
**********
*********
********
*******
******
*****
****
***
**
*
(C)
**********
*********
********
*******
******
*****
****
***
**
*
(D)
*
**
***
****
*****
******
*******
********
*********
**********
4.17 (Calculating Credit Limits) Collecting money becomes increasingly difficult during periods of recession, so companies may tighten their credit limits to prevent their accounts receivable
(money owed to them) from becoming too large. In response to a prolonged recession, one company
has cut its customers’ credit limits in half. Thus, if a particular customer had a credit limit of $2000,
it’s now $1000. If a customer had a credit limit of $5000, it’s now $2500. Write a program that
analyzes the credit status of three customers of this company. For each customer you’re given:
a) The customer’s account number
b) The customer’s credit limit before the recession
c) The customer’s current balance (i.e., the amount the customer owes the company).
Your program should calculate and print the new credit limit for each customer and should
determine (and print) which customers have current balances that exceed their new credit limits.
4.18 (Bar Chart Printing Program) One interesting application of computers is drawing graphs
and bar charts (sometimes called “histograms”). Write a program that reads five numbers (each between 1 and 30). For each number read, your program should print a line containing that number
of adjacent asterisks. For example, if your program reads the number seven, it should print *******.
4.19 (Calculating Sales) An online retailer sells five different products whose retail prices are
shown in the following table:
Product number
Retail price
1
2
3
4
5
$ 2.98
$ 4.50
$ 9.98
$ 4.49
$ 6.87
154
Chapter 4 C Program Control
Write a program that reads a series of pairs of numbers as follows:
a) Product number
b) Quantity sold for one day
Your program should use a switch statement to help determine the retail price for each product.
Your program should calculate and display the total retail value of all products sold last week.
4.20
(Truth Tables) Complete the following truth tables by filling in each blank with 0 or 1.
Condition1
Condition2
Condition1 && Condition2
0
0
nonzero
nonzero
0
nonzero
0
nonzero
0
0
Condition1
Condition2
Condition1 || Condition2
0
0
nonzero
nonzero
0
nonzero
0
nonzero
0
1
_____
_____
_____
_____
Condition1
!Condition1
0
nonzero
1
_____
4.21 Rewrite the program of Fig. 4.2 so that the initialization of the variable counter is done in
the definition rather than in the for statement.
4.22 (Average Grade) Modify the program of Fig. 4.7 so that it calculates the average grade for
the class.
4.23 (Calculating the Compound Interest with Integers) Modify the program of Fig. 4.6 so that
it uses only integers to calculate the compound interest. [Hint: Treat all monetary amounts as integral numbers of pennies. Then “break” the result into its dollar portion and cents portion by using
the division and remainder operations, respectively. Insert a period.]
4.24
Assume i = 1, j = 2, k = 3 and m = 2. What does each of the following statements print?
a) printf( "%d", i == 1 );
b) printf( "%d", j == 3 );
c) printf( "%d", i >= 1 && j < 4 );
d) printf( "%d", m < = 99 && k < m );
e) printf( "%d", j >= i || k == m );
f) printf( "%d", k + m < j || 3 - j >= k );
g) printf( "%d", !m );
h) printf( "%d", !( j - m ) );
i) printf( "%d", !( k > m ) );
j) printf( "%d", !( j > k ) );
Exercises
155
4.25 (Table of Decimal, Binary, Octal and Hexadecimal Equivalents) Write a program that
prints a table of the binary, octal and hexadecimal equivalents of the decimal numbers in the range
1 through 256. If you’re not familiar with these number systems, read Appendix C before you attempt this exercise. [Note: You can display an integer as an octal or hexadecimal value with the conversion specifiers %o and %X, respectively.]
(Calculating the Value of π) Calculate the value of π from the infinite series
4- …
π = 4 – 4--- + 4--- – 4--- + 4--- – ----+
3 5 7 9 11
Print a table that shows the value of π approximated by one term of this series, by two terms, by
three terms, and so on. How many terms of this series do you have to use before you first get 3.14?
3.141? 3.1415? 3.14159?
4.26
4.27 (Pythagorean Triples) A right triangle can have sides that are all integers. The set of three
integer values for the sides of a right triangle is called a Pythagorean triple. These three sides must
satisfy the relationship that the sum of the squares of two of the sides is equal to the square of the
hypotenuse. Find all Pythagorean triples for side1, side2, and the hypotenuse all no larger than 500.
Use a triple-nested for loop that simply tries all possibilities. This is an example of “brute-force”
computing. It’s not aesthetically pleasing to many people. But there are many reasons why these
techniques are important. First, with computing power increasing at such a phenomenal pace, solutions that would have taken years or even centuries of computer time to produce with the technology of just a few years ago can now be produced in hours, minutes or even seconds. Recent
microprocessor chips can process a billion instructions per second! Second, as you’ll learn in more
advanced computer science courses, there are large numbers of interesting problems for which
there’s no known algorithmic approach other than sheer brute force. We investigate many kinds of
problem-solving methodologies in this book. We’ll consider many brute-force approaches to various interesting problems.
4.28 (Calculating Weekly Pay) A company pays its employees as managers (who receive a fixed
weekly salary), hourly workers (who receive a fixed hourly wage for up to the first 40 hours they
work and “time-and-a-half”—i.e., 1.5 times their hourly wage—for overtime hours worked), commission workers (who receive $250 plus 5.7% of their gross weekly sales), or pieceworkers (who receive a fixed amount of money for each of the items they produce—each pieceworker in this
company works on only one type of item). Write a program to compute the weekly pay for each
employee. You do not know the number of employees in advance. Each type of employee has its
own pay code: Managers have paycode 1, hourly workers have code 2, commission workers have
code 3 and pieceworkers have code 4. Use a switch to compute each employee’s pay based on that
employee’s paycode. Within the switch, prompt the user (i.e., the payroll clerk) to enter the appropriate facts your program needs to calculate each employee’s pay based on that employee’s paycode.
[Note: You can input values of type double using the conversion specifier %lf with scanf.]
4.29 (De Morgan’s Laws) In this chapter, we discussed the logical operators &&, ||, and !. De
Morgan’s Laws can sometimes make it more convenient for us to express a logical expression. These
laws state that the expression !(condition1 && condition2) is logically equivalent to the expression
(!condition1 || !condition2). Also, the expression !(condition1 || condition2) is logically equivalent
to the expression (!condition1 && !condition2). Use De Morgan’s Laws to write equivalent expressions for each of the following, and then write a program to show that both the original expression
and the new expression in each case are equivalent.
a) !( x < 5 ) && !( y >= 7 )
b) !( a == b ) || !( g != 5 )
c) !( ( x <= 8 ) && ( y > 4 ) )
d) !( ( i > 4 ) || ( j <= 6 ) )
156
Chapter 4 C Program Control
4.30 (Replacing switch with if…else) Rewrite the program of Fig. 4.7 by replacing the switch
statement with a nested if…else statement; be careful to deal with the default case properly. Then
rewrite this new version by replacing the nested if…else statement with a series of if statements;
here, too, be careful to deal with the default case properly (this is more difficult than in the nested
if…else version). This exercise demonstrates that switch is a convenience and that any switch
statement can be written with only single-selection statements.
4.31 (Diamond-Printing Program) Write a program that prints the following diamond shape.
You may use printf statements that print either a single asterisk (*) or a single blank. Maximize
your use of repetition (with nested for statements) and minimize the number of printf statements.
*
***
*****
*******
*********
*******
*****
***
*
4.32 (Modified Diamond-Printing Program) Modify the program you wrote in Exercise 4.31 to
read an odd number in the range 1 to 19 to specify the number of rows in the diamond. Your program should then display a diamond of the appropriate size.
4.33 (Roman-Numeral Equivalent of Decimal Values) Write a program that prints a table of all
the Roman numeral equivalents of the decimal numbers in the range 1 to 100.
4.34 Describe the process you would use to replace a do…while loop with an equivalent while
loop. What problem occurs when you try to replace a while loop with an equivalent do…while
loop? Suppose you have been told that you must remove a while loop and replace it with a
do…while. What additional control statement would you need to use and how would you use it to
ensure that the resulting program behaves exactly as the original?
4.35 A criticism of the break statement and the continue statement is that each is unstructured.
Actually, break statements and continue statements can always be replaced by structured statements, although doing so can be awkward. Describe in general how you would remove any break
statement from a loop in a program and replace that statement with some structured equivalent.
[Hint: The break statement leaves a loop from within the body of the loop. The other way to leave
is by failing the loop-continuation test. Consider using in the loop-continuation test a second test
that indicates “early exit because of a ‘break’ condition.”] Use the technique you developed here to
remove the break statement from the program of Fig. 4.11.
4.36
1
2
3
4
5
6
7
8
What does the following program segment do?
for ( i = 1; i <= 5; ++i ) {
for ( j = 1; j <= 3; ++j ) {
for ( k = 1; k <= 4; ++k )
printf( "%s", "*" );
puts( "" );
}
puts( "" );
}
Making a Difference
157
4.37 Describe in general how you would remove any continue statement from a loop in a program and replace that statement with some structured equivalent. Use the technique you developed
here to remove the continue statement from the program of Fig. 4.12.
4.38 (“The Twelve Days of Christmas” Song) Write a program that uses repetition and switch
statements to print the song “The Twelve Days of Christmas.” One switch statement should be
used to print the day (i.e., “first,” “second,” etc.). A separate switch statement should be used to
print the remainder of each verse.
Making a Difference
4.39 (World Population Growth) World population has grown considerably over the centuries.
Continued growth could eventually challenge the limits of breathable air, drinkable water, arable
cropland and other limited resources. There’s evidence that growth has been slowing in recent years
and that world population could peak some time this century, then start to decline.
For this exercise, research world population growth issues online. Be sure to investigate various
viewpoints. Get estimates for the current world population and its growth rate (the percentage by
which it’s likely to increase this year). Write a program that calculates world population growth
each year for the next 75 years, using the simplifying assumption that the current growth rate will stay
constant. Print the results in a table. The first column should display the year from year 1 to year
75. The second column should display the anticipated world population at the end of that year.
The third column should display the numerical increase in the world population that would occur
that year. Using your results, determine the year in which the population would be double what it
is today, if this year’s growth rate were to persist.
4.40 (Tax Plan Alternatives; The “FairTax”) There are many proposals to make taxation fairer.
Check out the FairTax initiative in the United States at
www.fairtax.org/site/PageServer?pagename=calculator
Research how the proposed FairTax works. One suggestion is to eliminate income taxes and most
other taxes in favor of a 23% consumption tax on all products and services that you buy. Some
FairTax opponents question the 23% figure and say that because of the way the tax is calculated, it
would be more accurate to say the rate is 30%—check this carefully. Write a program that prompts
the user to enter expenses in various categories (e.g., housing, food, clothing, transportation, education, health care, vacations), then prints the estimated FairTax that person would pay.
5
Form ever follows function.
—Louis Henri Sullivan
O! call back yesterday, bid time
return.
—William Shakespeare
Answer me in one word.
—William Shakespeare
There is a point at which
methods devour themselves.
—Frantz Fanon
Objectives
In this chapter, you’ll:
■
Construct programs
modularly from small pieces
called functions.
■
Use common math functions
in the C standard library.
■
Create new functions.
■
Use the mechanisms that
pass information between
functions.
■
Learn how the function call/
return mechanism is
supported by the function
call stack and stack frames.
■
Use simulation techniques
based on random number
generation.
■
Write and use functions that
call themselves.
C Functions
5.1 Introduction
5.1
5.2
5.3
5.4
5.5
5.6
5.7
5.8
5.9
Introduction
Program Modules in C
Math Library Functions
Functions
Function Definitions
Function Prototypes: A Deeper Look
Function Call Stack and Stack Frames
Headers
Passing Arguments By Value and By
Reference
159
5.10
5.11
5.12
5.13
5.14
5.15
Random Number Generation
Example: A Game of Chance
Storage Classes
Scope Rules
Recursion
Example Using Recursion: Fibonacci
Series
5.16 Recursion vs. Iteration
5.17 Secure C Programming
Summary | Terminology | Self-Review Exercises | Answers to Self-Review Exercises | Exercises |
Making a Difference
5.1 Introduction
Most computer programs that solve real-world problems are much larger than the programs presented in the first few chapters. Experience has shown that the best way to develop and maintain a large program is to construct it from smaller pieces or modules, each
of which is more manageable than the original program. This technique is called divide
and conquer. This chapter describes some key features of the C language that facilitate the
design, implementation, operation and maintenance of large programs.
5.2 Program Modules in C
Modules in C are called functions. C programs are typically written by combining new
functions you write with prepackaged functions available in the C standard library. We discuss both kinds of functions in this chapter. The C standard library provides a rich collection of functions for performing common mathematical calculations, string manipulations,
character manipulations, input/output, and many other useful operations. This makes your
job easier, because these functions provide many of the capabilities you need.
Good Programming Practice 5.1
Familiarize yourself with the rich collection of functions in the C standard library.
Software Engineering Observation 5.1
Avoid reinventing the wheel. When possible, use C standard library functions instead of
writing new functions. This can reduce program development time.
Portability Tip 5.1
Using the functions in the C standard library helps make programs more portable.
The C language and the standard library are both specified by the C standard, and
they’re both provided with standard C systems (with the exception that some of the
160
Chapter 5 C Functions
libraries are designated as optional). The functions printf, scanf and pow that we’ve used
in previous chapters are standard library functions.
You can write functions to define specific tasks that may be used at many points in a
program. These are sometimes referred to as programmer-defined functions. The actual
statements defining the function are written only once, and the statements are hidden
from other functions.
Functions are invoked by a function call, which specifies the function name and provides information (as arguments) that the function needs to perform its designated task. A
common analogy for this is the hierarchical form of management. A boss (the calling function or caller) asks a worker (the called function) to perform a task and report back when
the task is done (Fig. 5.1). For example, a function needing to display information on the
screen calls the worker function printf to perform that task, then printf displays the information and reports back—or returns—to the calling function when its task is completed.
The boss function does not know how the worker function performs its designated tasks. The
worker may call other worker functions, and the boss will be unaware of this. We’ll soon see
how this “hiding” of implementation details promotes good software engineering. Figure 5.1
shows a boss function communicating with several worker functions in a hierarchical
manner. Note that Worker1 acts as a boss function to worker4 and worker5. Relationships
among functions may differ from the hierarchical structure shown in this figure.
Boss
Worker1
Worker4
Worker2
Worker3
Worker5
Fig. 5.1 | Hierarchical boss-function/worker-function relationship.
5.3 Math Library Functions
Math library functions allow you to perform certain common mathematical calculations.
We use some of them here to introduce the concept of functions. Later in the book, we’ll
discuss many of the other functions in the C standard library.
Functions are normally used in a program by writing the name of the function
followed by a left parenthesis followed by the argument (or a comma-separated list of
arguments) of the function followed by a right parenthesis. For example, to calculate and
print the square root of 900.0 you might write
printf( "%.2f", sqrt( 900.0 ) );
When this statement executes, the math library function sqrt is called to calculate the square
root of the number contained in the parentheses (900.0). The number 900.0 is the argument
of the sqrt function. The preceding statement would print 30.00. The sqrt function takes
5.3 Math Library Functions
161
an argument of type double and returns a result of type double. All functions in the math
library that return floating-point values return the data type double. Note that double values, like float values, can be output using the %f conversion specification.
Error-Prevention Tip 5.1
Include the math header by using the preprocessor directive
using functions in the math library.
#include <math.h>
when
Function arguments may be constants, variables, or expressions. If c1 = 13.0, d = 3.0
and f = 4.0, then the statement
printf( "%.2f", sqrt( c1 + d * f ) );
calculates and prints the square root of 13.0 + 3.0 * 4.0 = 25.0, namely 5.00.
Figure 5.2 summarizes a small sample of the C math library functions. In the figure,
the variables x and y are of type double. The C11 standard adds a wide range of floatingpoint and complex-number capabilities.
Function
Description
Example
sqrt( x )
square root of x
is 30.0
is 3.0
cbrt( 27.0 ) is 3.0
cbrt( -8.0 ) is -2.0
exp( 1.0 ) is 2.718282
exp( 2.0 ) is 7.389056
log( 2.718282 ) is 1.0
log( 7.389056 ) is 2.0
log10( 1.0 ) is 0.0
log10( 10.0 ) is 1.0
log10( 100.0 ) is 2.0
fabs( 13.5 ) is 13.5
fabs( 0.0 ) is 0.0
fabs( -13.5 ) is 13.5
ceil( 9.2 ) is 10.0
ceil( -9.8 ) is -9.0
floor( 9.2 ) is 9.0
floor( -9.8 ) is -10.0
pow( 2, 7 ) is 128.0
pow( 9, .5 ) is 3.0
sqrt( 900.0 )
sqrt( 9.0 )
cbrt( x )
cube root of x (C99 and C11 only)
exp( x )
exponential function ex
log( x )
natural logarithm of x (base e)
log10( x )
logarithm of x (base 10)
fabs( x )
absolute value of x as a floating-point
number
ceil( x )
rounds x to the smallest integer not less
than x
rounds x to the largest integer not
greater than x
x raised to power y (x y )
floor( x )
pow( x, y )
fmod( x, y )
sin( x )
cos( x )
tan( x )
remainder of x/y as a floating-point
number
trigonometric sine of x (x in radians)
trigonometric cosine of x (x in radians)
trigonometric tangent of x (x in radians)
Fig. 5.2 | Commonly used math library functions.
fmod( 13.657, 2.333 )
is 0.0
is 1.0
) is 0.0
sin( 0.0 )
cos( 0.0 )
tan( 0.0
is 1.992
162
Chapter 5 C Functions
5.4 Functions
Functions allow you to modularize a program. All variables defined in function definitions
are local variables—they can be accessed only in the function in which they’re defined.
Most functions have a list of parameters that provide the means for communicating information between functions. A function’s parameters are also local variables of that function.
Software Engineering Observation 5.2
In programs containing many functions, main is often implemented as a group of calls to
functions that perform the bulk of the program’s work.
There are several motivations for “functionalizing” a program. The divide-and-conquer approach makes program development more manageable. Another motivation is
software reusability—using existing functions as building blocks to create new programs.
Software reusability is a major factor in the object-oriented programming movement that
you’ll learn more about when you study languages derived from C, such as C++, Java and
C# (pronounced “C sharp”). With good function naming and definition, programs can
be created from standardized functions that accomplish specific tasks, rather than being
built by using customized code. This is known as abstraction. We use abstraction each
time we use standard library functions like printf, scanf and pow. A third motivation is
to avoid repeating code in a program. Packaging code as a function allows the code to be
executed from other locations in a program simply by calling the function.
Software Engineering Observation 5.3
Each function should be limited to performing a single, well-defined task, and the
function name should express that task. This facilitates abstraction and promotes software
reusability.
Software Engineering Observation 5.4
If you cannot choose a concise name that expresses what the function does, it’s possible that
your function is attempting to perform too many diverse tasks. It’s usually best to break
such a function into several smaller functions—this is sometimes called decomposition.
5.5 Function Definitions
Each program we’ve presented has consisted of a function called main that called standard
library functions to accomplish its tasks. We now consider how to write custom functions.
Consider a program that uses a function square to calculate and print the squares of the
integers from 1 to 10 (Fig. 5.3).
1
2
3
4
5
6
// Fig. 5.3: fig05_03.c
// Creating and using a programmer-defined function.
#include <stdio.h>
int square( int y ); // function prototype
Fig. 5.3 | Creating and using a programmer-defined function. (Part 1 of 2.)
5.5 Function Definitions
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
1
163
// function main begins program execution
int main( void )
{
int x; // counter
// loop 10 times and calculate and output square of x each time
for ( x = 1; x <= 10; ++x ) {
printf( "%d ", square( x ) ); // function call
} // end for
puts( "" );
} // end main
// square function definition returns the square of its parameter
int square( int y ) // y is a copy of the argument to the function
{
return y * y; // returns the square of y as an int
} // end function square
4
9
16
25
36
49
64
81
100
Fig. 5.3 | Creating and using a programmer-defined function. (Part 2 of 2.)
Function square is invoked or called in main within the printf statement (line 14)
printf( "%d
", square( x ) ); // function call
Function square receives a copy of the value of x in the parameter y (line 21). Then square
calculates y * y. The result is passed back returned to function printf in main where
square was invoked (line 14), and printf displays the result. This process is repeated 10
times using the for statement.
The definition of function square (lines 21–24) shows that square expects an integer
parameter y. The keyword int preceding the function name (line 21) indicates that
square returns an integer result. The return statement in square passes the value of the
expression y * y (that is, the result of the calculation) back to the calling function.
Line 5
int square( int y ); // function prototype
is a function prototype. The int in parentheses informs the compiler that square expects
to receive an integer value from the caller. The int to the left of the function name square
informs the compiler that square returns an integer result to the caller. The compiler refers
to the function prototype to check that any calls to square (line 14) contain the correct
return type, the correct number of arguments and the correct argument types, and that the arguments are in the correct order. Function prototypes are discussed in detail in Section 5.6.
The format of a function definition is
return-value-type function-name( parameter-list )
{
definitions
statements
}
164
Chapter 5 C Functions
The function-name is any valid identifier. The return-value-type is the data type of the result returned to the caller. The return-value-type void indicates that a function does not
return a value. Together, the return-value-type, function-name and parameter-list are sometimes referred to as the function header.
Error-Prevention Tip 5.2
Check that your functions that are supposed to return values do so. Check that your functions that are not supposed to return values do not.
The parameter-list is a comma-separated list that specifies the parameters received by
the function when it’s called. If a function does not receive any values, parameter-list is
void. A type must be listed explicitly for each parameter.
Common Programming Error 5.1
Specifying function parameters of the same type as double x, y instead of double x, double y results in a compilation error.
Common Programming Error 5.2
Placing a semicolon after the right parenthesis enclosing the parameter list of a function
definition is a syntax error.
Common Programming Error 5.3
Defining a parameter again as a local variable in a function is a compilation error.
Good Programming Practice 5.2
Although it’s not incorrect to do so, do not use the same names for a function’s arguments
and the corresponding parameters in the function definition. This helps avoid ambiguity.
The definitions and statements within braces form the function body, which is also
referred to as a block. Variables can be declared in any block, and blocks can be nested.
Common Programming Error 5.4
Defining a function inside another function is a syntax error.
Good Programming Practice 5.3
Choosing meaningful function names and meaningful parameter names makes programs
more readable and helps avoid excessive use of comments.
Software Engineering Observation 5.5
Small functions promote software reusability.
Software Engineering Observation 5.6
Programs should be written as collections of small functions. This makes programs easier
to write, debug, maintain and modify.
5.5 Function Definitions
165
Software Engineering Observation 5.7
A function requiring a large number of parameters may be performing too many tasks.
Consider dividing the function into smaller functions that perform the separate tasks. The
function header should fit on one line if possible.
Software Engineering Observation 5.8
The function prototype, function header and function calls should all agree in the number,
type, and order of arguments and parameters, and in the type of return value.
There are three ways to return control from a called function to the point at which a
function was invoked. If the function does not return a result, control is returned simply
when the function-ending right brace is reached, or by executing the statement
return;
If the function does return a result, the statement
return expression;
returns the value of expression to the caller.
main’s
Return Type
Notice that main has an int return type. The return value of main is used to indicate
whether the program executed correctly. In earlier versions of C, we’d explicitly place
return 0;
at the end of main—0 indicates that a program ran successfully. The C standard indicates
that main implicitly returns 0 if you to omit the preceding statement—as we’ve done
throughout this book. You can explicitly return non-zero values from main to indicate that
a problem occured during your program’s execution. For information on how to report a
program failure, see the documentation for your particular operating-system environment.
Function maximum
Our second example uses a programmer-defined function maximum to determine and return the largest of three integers (Fig. 5.4). The integers are input with scanf (line 15).
Next, they’re passed to maximum (line 19), which determines the largest integer. This value
is returned to main by the return statement in maximum (line 36). The value returned is
then printed in the printf statement (line 19).
1
2
3
4
5
6
7
8
9
// Fig. 5.4: fig05_04.c
// Finding the maximum of three integers.
#include <stdio.h>
int maximum( int x, int y, int z ); // function prototype
// function main begins program execution
int main( void )
{
Fig. 5.4 | Finding the maximum of three integers. (Part 1 of 2.)
166
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
Chapter 5 C Functions
int number1; // first integer entered by the user
int number2; // second integer entered by the user
int number3; // third integer entered by the user
printf( "%s", "Enter three integers: " );
scanf( "%d%d%d", &number1, &number2, &number3 );
// number1, number2 and number3 are arguments
// to the maximum function call
printf( "Maximum is: %d\n", maximum( number1, number2, number3 ) );
} // end main
// Function maximum definition
// x, y and z are parameters
int maximum( int x, int y, int z )
{
int max = x; // assume x is largest
if ( y > max ) { // if y is larger than max,
max = y; // assign y to max
} // end if
if ( z > max ) { // if z is larger than max,
max = z; // assign z to max
} // end if
return max; // max is largest value
} // end function maximum
Enter three integers: 22 85 17
Maximum is: 85
Enter three integers: 47 32 14
Maximum is: 47
Enter three integers: 35 8 79
Maximum is: 79
Fig. 5.4 | Finding the maximum of three integers. (Part 2 of 2.)
5.6 Function Prototypes: A Deeper Look
An important feature of C is the function prototype. This feature was borrowed from C++.
The compiler uses function prototypes to validate function calls. Early versions of C did
not perform this kind of checking, so it was possible to call functions improperly without
the compiler detecting the errors. Such calls could result in fatal execution-time errors or
nonfatal errors that caused subtle, difficult-to-detect problems. Function prototypes correct this deficiency.
5.6 Function Prototypes: A Deeper Look
167
Good Programming Practice 5.4
Include function prototypes for all functions to take advantage of C’s type-checking capabilities. Use #include preprocessor directives to obtain function prototypes for the standard library functions from the headers for the appropriate libraries, or to obtain headers
containing function prototypes for functions developed by you and/or your group members.
The function prototype for maximum in Fig. 5.4 (line 5) is
int maximum( int x, int y, int z ); // function prototype
It states that maximum takes three arguments of type int and returns a result of type int.
Notice that the function prototype is the same as the first line of maximum’s definition.
Good Programming Practice 5.5
Parameter names are sometimes included in function prototypes (our preference) for documentation purposes. The compiler ignores these names.
Common Programming Error 5.5
Forgetting the semicolon at the end of a function prototype is a syntax error.
Compilation Errors
A function call that does not match the function prototype is a compilation error. An error
is also generated if the function prototype and the function definition disagree. For example, in Fig. 5.4, if the function prototype had been written
void maximum( int x, int y, int z );
the compiler would generate an error because the void return type in the function prototype would differ from the int return type in the function header.
Argument Coercion and “Usual Arithmetic Conversion Rules”
Another important feature of function prototypes is the coercion of arguments, i.e., the
forcing of arguments to the appropriate type. For example, the math library function sqrt
can be called with an integer argument even though the function prototype in <math.h>
specifies a double parameter, and the function will still work correctly. The statement
printf( "%.3f\n", sqrt( 4 ) );
correctly evaluates sqrt(4) and prints the value 2.000. The function prototype causes the
compiler to convert a copy of the integer value 4 to the double value 4.0 before the copy is
passed to sqrt. In general, argument values that do not correspond precisely to the parameter
types in the function prototype are converted to the proper type before the function is called.
These conversions can lead to incorrect results if C’s usual arithmetic conversion rules are
not followed. These rules specify how values can be converted to other types without losing data. In our sqrt example above, an int is automatically converted to a double without changing its value (because double can represent a much larger range of values than
int). However, a double converted to an int truncates the fractional part of the double
value, thus changing the original value. Converting large integer types to small integer
types (e.g., long to short) may also result in changed values.
The usual arithmetic conversion rules automatically apply to expressions containing
values of two data types (also referred to as mixed-type expressions), and are handled for
168
Chapter 5 C Functions
you by the compiler. In a mixed-type expression, the compiler makes a temporary copy of
the value that needs to be converted then converts the copy to the “highest” type in the
expression—the original value remains unchanged. The usual arithmetic conversion rules
for a mixed-type expression containing at least one floating-point value are:
•
If one of the values is a long double, the other is converted to a long double.
•
If one of the values is a double, the other is converted to a double.
•
If one of the values is a float, the other is converted to a float.
If the mixed-type expression contains only integer types, then the usual arithmetic conversions specify a set of integer promotion rules. In most cases, the integer types lower in
Fig. 5.5 are converted to types higher in the figure. Section 6.3.1 of the C standard document specifies the complete details of arithmetic operands and the usual arithmetic conversion rules. Figure 5.5 lists the floating-point and integer data types with each type’s
printf and scanf conversion specifications.
printf conversion
specification
scanf conversion
specification
long double
%Lf
%Lf
double
%f
%lf
float
%f
%f
unsigned long long int
%llu
%llu
long long int
%lld
%lld
unsigned long int
%lu
%lu
long int
%ld
%ld
unsigned int
%u
%u
int
%d
%d
unsigned short
%hu
%hu
short
%hd
%hd
char
%c
%c
Data type
Floating-point types
Integer types
Fig. 5.5 | Arithmetic data types and their conversion specifications.
Converting values to lower types in Fig. 5.5 can result in incorrect values, so the compiler typically issues warnings for such cases. A value can be converted to a lower type only
by explicitly assigning the value to a variable of lower type or by using a cast operator. Arguments in a function call are converted to the parameter types specified in a function prototype as if the arguments were being assigned directly to variables of those types. If our square
function that uses an int parameter (Fig. 5.3) is called with a floating-point argument, the
argument is converted to int (a lower type), and square usually returns an incorrect value.
For example, square(4.5) returns 16, not 20.25.
5.7 Function Call Stack and Stack Frames
169
Common Programming Error 5.6
Converting from a higher data type in the promotion hierarchy to a lower type can change
the data value. Many compilers issue warnings in such cases.
If there’s no function prototype for a function, the compiler forms its own function
prototype using the first occurrence of the function—either the function definition or a
call to the function. This typically leads to warnings or errors, depending on the compiler.
Error-Prevention Tip 5.3
Always include function prototypes for the functions you define or use in your program to
help prevent compilation errors and warnings.
Software Engineering Observation 5.9
A function prototype placed outside any function definition applies to all calls to the
function appearing after the function prototype in the file. A function prototype placed in
a function applies only to calls made in that function.
5.7 Function Call Stack and Stack Frames
To understand how C performs function calls, we first need to consider a data structure
(i.e., collection of related data items) known as a stack. Think of a stack as analogous to a
pile of dishes. When a dish is placed on the pile, it’s normally placed at the top (referred to
as pushing the dish onto the stack). Similarly, when a dish is removed from the pile, it’s
normally removed from the top (referred to as popping the dish off the stack). Stacks are
known as last-in, first-out (LIFO) data structures—the last item pushed (inserted) on the
stack is the first item popped (removed) from the stack.
An important mechanism for computer science students to understand is the function
call stack (sometimes referred to as the program execution stack). This data structure—
working “behind the scenes”—supports the function call/return mechanism. It also supports the creation, maintenance and destruction of each called function’s automatic variables. We explained the last-in, first-out (LIFO) behavior of stacks with our dish-stacking
example. As we’ll see in Figs. 5.7–5.9, this LIFO behavior is exactly what a function does
when returning to the function that called it.
As each function is called, it may call other functions, which may call other functions—all before any function returns. Each function eventually must return control to the
function that called it. So, we must keep track of the return addresses that each function
needs to return control to the function that called it. The function call stack is the perfect
data structure for handling this information. Each time a function calls another function,
an entry is pushed onto the stack. This entry, called a stack frame, contains the return
address that the called function needs in order to return to the calling function. It also contains some additional information we’ll soon discuss. If the called function returns, instead
of calling another function before returning, the stack frame for the function call is popped,
and control transfers to the return address in the popped stack frame.
Each called function always finds the information it needs to return to its caller at the
top of the call stack. And, if a function makes a call to another function, a stack frame for the
new function call is simply pushed onto the call stack. Thus, the return address required by
the newly called function to return to its caller is now located at the top of the stack.
170
Chapter 5 C Functions
The stack frames have another important responsibility. Most functions have automatic variables—parameters and some or all of their local variables. Automatic variables
need to exist while a function is executing. They need to remain active if the function
makes calls to other functions. But when a called function returns to its caller, the called
function’s automatic variables need to “go away.” The called function’s stack frame is a
perfect place to reserve the memory for automatic variables. That stack frame exists only
as long as the called function is active. When that function returns—and no longer needs
its local automatic variables—its stack frame is popped from the stack, and those local automatic variables are no longer known to the program.
Of course, the amount of memory in a computer is finite, so only a certain amount
of memory can be used to store stack frames on the function call stack. If more function
calls occur than can have their stack frames stored on the function call stack, a fatal error
known as stack overflow occurs.
Function Call Stack in Action
Now let’s consider how the call stack supports the operation of a square function called
by main (lines 8–13 of Fig. 5.6). First the operating system calls main—this pushes a stack
frame onto the stack (shown in Fig. 5.7). The stack frame tells main how to return to the
operating system (i.e., transfer to return address R1) and contains the space for main’s automatic variable (i.e., a, which is initialized to 10).
Function main—before returning to the operating system—now calls function
square in line 12 of Fig. 5.6. This causes a stack frame for square (lines 16–19) to be
pushed onto the function call stack (Fig. 5.8). This stack frame contains the return address
that square needs to return to main (i.e., R2) and the memory for square’s automatic variable (i.e., x).
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
// Fig. 5.6: fig05_06.c
// Demonstrating the function call stack
// and stack frames using a function square.
#include <stdio.h>
int square( int ); // prototype for function square
int main()
{
int a = 10; // value to square (local automatic variable in main)
printf( "%d squared: %d\n", a, square( a ) ); // display a squared
} // end main
// returns the square of an integer
int square( int x ) // x is a local variable
{
return x * x; // calculate square and return result
} // end function square
10 squared: 100
Fig. 5.6 | Demonstrating the function call stack and stack frames using a function square.
5.7 Function Call Stack and Stack Frames
171
Step 1: Operating system invokes main to execute application
int main()
{
Operating system
int a = 10;
printf("%d squared: %d\n",
a, square( a ) );
}
Return location R1
Function call stack after Step 1
Top of stack
Return location: R1
Stack frame
for function main
Automatic variables:
a
10
Key
Lines that represent the operating
system executing instructions
Fig. 5.7 | Function call stack after the operating system invokes main to execute the program.
Step 2: main invokes function square to perform calculation
int main()
int square( int x )
{
int a = 10;
printf("%d squared: %d\n",
a, square( a ) );
Return location R2
{
return x * x;
}
}
Function call stack after Step 2
Top of stack
Return location: R2
Stack frame for
function square
Automatic variables:
x
10
Return location: R1
Stack frame
for function main
Automatic variables:
a
10
Fig. 5.8 | Function call stack after main invokes square to perform the calculation.
172
Chapter 5 C Functions
After square calculates the square of its argument, it needs to return to main—and no
longer needs the memory for its automatic variable x. So the stack is popped—giving
square the return location in main (i.e., R2) and losing square’s automatic variable.
Figure 5.9 shows the function call stack after square’s stack frame has been popped.
Step 3: square returns its result to main
int main()
int square( int x )
{
int a = 10;
printf("%d squared: %d\n",
a, square( a ) );
Return location R2
{
return x * x;
}
}
Function call stack after Step 3
Top of stack
Return location: R1
Stack frame
for function main
Automatic variables:
a
10
Fig. 5.9 | Function call stack after function square returns to main.
Function main now displays the result of calling square (line 12). Reaching the
closing right brace of main causes its stack frame to be popped from the stack, gives main
the address it needs to return to the operating system (i.e., R1 in Fig. 5.7) and causes the
memory for main’s automatic variable (i.e., a) to become unavailable.
You’ve now seen how valuable the stack data structure is in implementing a key mechanism that supports program execution. Data structures have many important applications in computer science. We discuss stacks, queues, lists, trees and other data structures
in Chapter 12.
5.8 Headers
Each standard library has a corresponding header containing the function prototypes for all
the functions in that library and definitions of various data types and constants needed by
those functions. Figure 5.10 lists alphabetically some of the standard library headers that may
be included in programs. The C standard includes additional headers. The term “macros”
that’s used several times in Fig. 5.10 is discussed in detail in Chapter 13.
You can create custom headers. Programmer-defined headers should also use the .h
filename extension. A programmer-defined header can be included by using the #include
preprocessor directive. For example, if the prototype for our square function was located
5.9 Passing Arguments By Value and By Reference
173
Header
Explanation
<assert.h>
Contains information for adding diagnostics that aid program debugging.
Contains function prototypes for functions that test characters for certain properties, and function prototypes for functions that can be used to convert lowercase letters to uppercase letters and vice versa.
Defines macros that are useful for reporting error conditions.
Contains the floating-point size limits of the system.
Contains the integral size limits of the system.
Contains function prototypes and other information that enables a program to
be modified for the current locale on which it’s running. The notion of locale
enables the computer system to handle different conventions for expressing data
such as dates, times, currency amounts and large numbers throughout the world.
<ctype.h>
<errno.h>
<float.h>
<limits.h>
<locale.h>
<math.h>
<setjmp.h>
<signal.h>
<stdarg.h>
<stddef.h>
<stdio.h>
<stdlib.h>
<string.h>
<time.h>
Contains function prototypes for math library functions.
Contains function prototypes for functions that allow bypassing of the usual
function call and return sequence.
Contains function prototypes and macros to handle various conditions that may
arise during program execution.
Defines macros for dealing with a list of arguments to a function whose number
and types are unknown.
Contains common type definitions used by C for performing calculations.
Contains function prototypes for the standard input/output library functions,
and information used by them.
Contains function prototypes for conversions of numbers to text and text to
numbers, memory allocation, random numbers, and other utility functions.
Contains function prototypes for string-processing functions.
Contains function prototypes and types for manipulating the time and date.
Fig. 5.10 | Some of the standard library headers.
in the header square.h, we’d include that header in our program by using the following
directive at the top of the program:
#include "square.h"
Section 13.2 presents additional information on including headers.
5.9 Passing Arguments By Value and By Reference
In many programming languages, there are two ways to pass arguments—pass-by-value
and pass-by-reference. When arguments are passed by value, a copy of the argument’s value
is made and passed to the called function. Changes to the copy do not affect an original
variable’s value in the caller. When an argument is passed by reference, the caller allows the
called function to modify the original variable’s value.
Pass-by-value should be used whenever the called function does not need to modify
the value of the caller’s original variable. This prevents the accidental side effects (variable
174
Chapter 5 C Functions
modifications) that so greatly hinder the development of correct and reliable software systems. Pass-by-reference should be used only with trusted called functions that need to
modify the original variable.
In C, all arguments are passed by value. As we’ll see in Chapter 7, it’s possible to simulate pass-by-reference by using the address operator and the indirection operator. In
Chapter 6, we’ll see that array arguments are automatically passed by reference for performance reasons. We’ll see in Chapter 7 that this is not a contradiction. For now, we concentrate on pass-by-value.
5.10 Random Number Generation
We now take a brief and, hopefully, entertaining diversion into simulation and game playing. In this and the next section, we’ll develop a nicely structured game-playing program
that includes multiple functions. The program uses most of the control statements we’ve
studied. The element of chance can be introduced into computer applications by using the
C standard library function rand from the <stdlib.h> header.
Consider the following statement:
i = rand();
The rand function generates an integer between 0 and RAND_MAX (a symbolic constant defined in the <stdlib.h> header). Standard C states that the value of RAND_MAX must be at
least 32767, which is the maximum value for a two-byte (i.e., 16-bit) integer. The programs in this section were tested on Microsoft Visual C++ with a maximum RAND_MAX value of 32767 and on GNU gcc with a maximum RAND_MAX value of 2147483647. If rand
truly produces integers at random, every number between 0 and RAND_MAX has an equal
chance (or probability) of being chosen each time rand is called.
The range of values produced directly by rand is often different from what’s needed
in a specific application. For example, a program that simulates coin tossing might require
only 0 for “heads” and 1 for “tails.” A dice-rolling program that simulates a six-sided die
would require random integers from 1 to 6.
Rolling a Six-Sided Die
To demonstrate rand, let’s develop a program to simulate 20 rolls of a six-sided die and
print the value of each roll. The function prototype for function rand is in <stdlib.h>.
We use the remainder operator (%) in conjunction with rand as follows
rand() % 6
to produce integers in the range 0 to 5. This is called scaling. The number 6 is called the
scaling factor. We then shift the range of numbers produced by adding 1 to our previous
result. The output of Fig. 5.11 confirms that the results are in the range 1 to 6—the actual
random values chosen might vary by compiler.
1
2
3
4
// Fig. 5.11: fig05_11.c
// Shifted, scaled random integers produced by 1 + rand() % 6.
#include <stdio.h>
#include <stdlib.h>
Fig. 5.11 | Shifted, scaled random integers produced by 1 + rand() % 6. (Part 1 of 2.)
5.10 Random Number Generation
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
175
// function main begins program execution
int main( void )
{
unsigned int i; // counter
// loop 20 times
for ( i = 1; i <= 20; ++i ) {
// pick random number from 1 to 6 and output it
printf( "%10d", 1 + ( rand() % 6 ) );
// if counter is divisible by 5, begin new line of output
if ( i % 5 == 0 ) {
puts( "" );
} // end if
} // end for
} // end main
6
5
6
6
6
1
6
2
5
1
2
3
5
5
4
4
6
3
2
1
Fig. 5.11 | Shifted, scaled random integers produced by 1 + rand() % 6. (Part 2 of 2.)
Rolling a Six-Sided Die 6,000,000 Times
To show that these numbers occur approximately with equal likelihood, let’s simulate
6,000,000 rolls of a die with the program of Fig. 5.12. Each integer from 1 to 6 should
appear approximately 1,000,000 times.
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
// Fig. 5.12: fig05_12.c
// Rolling a six-sided die 6,000,000 times.
#include <stdio.h>
#include <stdlib.h>
// function main begins program
int main( void )
{
unsigned int frequency1 = 0;
unsigned int frequency2 = 0;
unsigned int frequency3 = 0;
unsigned int frequency4 = 0;
unsigned int frequency5 = 0;
unsigned int frequency6 = 0;
execution
//
//
//
//
//
//
rolled
rolled
rolled
rolled
rolled
rolled
1
2
3
4
5
6
counter
counter
counter
counter
counter
counter
unsigned int roll; // roll counter, value 1 to 6000000
int face; // represents one roll of the die, value 1 to 6
Fig. 5.12 | Rolling a six-sided die 6,000,000 times. (Part 1 of 2.)
176
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Chapter 5 C Functions
// loop 6000000 times and summarize results
for ( roll = 1; roll <= 6000000; ++roll ) {
face = 1 + rand() % 6; // random number from 1 to 6
// determine face value and increment appropriate counter
switch ( face ) {
case 1: // rolled 1
++frequency1;
break;
case 2: // rolled 2
++frequency2;
break;
case 3: // rolled 3
++frequency3;
break;
case 4: // rolled 4
++frequency4;
break;
case 5: // rolled 5
++frequency5;
break;
case 6: // rolled 6
++frequency6;
break; // optional
} // end switch
} // end for
// display results in tabular format
printf( "%s%13s\n", "Face", "Frequency" );
printf( "
1%13u\n", frequency1 );
printf( "
2%13u\n", frequency2 );
printf( "
3%13u\n", frequency3 );
printf( "
4%13u\n", frequency4 );
printf( "
5%13u\n", frequency5 );
printf( "
6%13u\n", frequency6 );
} // end main
Face
1
2
3
4
5
6
Frequency
999702
1000823
999378
998898
1000777
1000422
Fig. 5.12 | Rolling a six-sided die 6,000,000 times. (Part 2 of 2.)
5.10 Random Number Generation
177
As the program output shows, by scaling and shifting we’ve used the rand function to
realistically simulate the rolling of a six-sided die. Note the use of the %s conversion specifier to print the character strings "Face" and "Frequency" as column headers (line 53).
After we study arrays in Chapter 6, we’ll show how to replace this 26-line switch statement elegantly with a single-line statement.
Randomizing the Random Number Generator
Executing the program of Fig. 5.11 again produces
6
5
6
6
6
1
6
2
5
1
2
3
5
5
4
4
6
3
2
1
Notice that exactly the same sequence of values was printed. How can these be random numbers? Ironically, this repeatability is an important characteristic of function rand. When debugging a program, this repeatability is essential for proving that corrections to a program
work properly.
Function rand actually generates pseudorandom numbers. Calling rand repeatedly
produces a sequence of numbers that appears to be random. However, the sequence repeats
itself each time the program is executed. Once a program has been thoroughly debugged,
it can be conditioned to produce a different sequence of random numbers for each execution. This is called randomizing and is accomplished with the standard library function
srand. Function srand takes an unsigned integer argument and seeds function rand to
produce a different sequence of random numbers for each execution of the program.
We demonstrate function srand in Fig. 5.13. Function srand takes an unsigned int
value as an argument. The conversion specifier %u is used to read an unsigned int value
with scanf. The function prototype for srand is found in <stdlib.h>.
Let’s run the program several times and observe the results. Notice that a different
sequence of random numbers is obtained each time the program is run, provided that a
different seed is supplied.
To randomize without entering a seed each time, use a statement like
srand( time( NULL ) );
This causes the computer to read its clock to obtain the value for the seed automatically.
Function time returns the number of seconds that have passed since midnight on January
1, 1970. This value is converted to an unsigned integer and used as the seed to the random
number generator. The function prototype for time is in <time.h>. We’ll say more about
NULL in Chapter 7.
1
2
3
4
5
// Fig. 5.13: fig05_13.c
// Randomizing the die-rolling program.
#include <stdlib.h>
#include <stdio.h>
Fig. 5.13 | Randomizing the die-rolling program. (Part 1 of 2.)
178
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
Chapter 5 C Functions
// function main begins program execution
int main( void )
{
unsigned int i; // counter
unsigned int seed; // number used to seed the random number generator
printf( "%s", "Enter seed: " );
scanf( "%u", &seed ); // note %u for unsigned int
srand( seed );
// seed the random number generator
// loop 10 times
for ( i = 1; i <= 10; ++i ) {
// pick a random number from 1 to 6 and output it
printf( "%10d", 1 + ( rand() % 6 ) );
// if counter is divisible by 5, begin a new line of output
if ( i % 5 == 0 ) {
puts( "" );
} // end if
} // end for
} // end main
Enter seed: 67
6
1
1
6
4
1
6
6
2
4
Enter seed: 867
2
1
4
1
6
3
1
6
6
2
Enter seed: 67
6
1
1
6
4
1
6
6
2
4
Fig. 5.13 | Randomizing the die-rolling program. (Part 2 of 2.)
Generalized Scaling and Shifting of Random Numbers
The values produced directly by rand are always in the range:
0
≤ rand() ≤ RAND_MAX
As you know, the following statement simulates rolling a six-sided die:
face = 1 + rand() % 6;
This statement always assigns an integer value (at random) to the variable face in the
range 1 ≤ face ≤ 6. The width of this range (i.e., the number of consecutive integers in
the range) is 6 and the starting number in the range is 1. Referring to the preceding statement, we see that the width of the range is determined by the number used to scale rand
5.11 Example: A Game of Chance
179
with the remainder operator (i.e., 6), and the starting number of the range is equal to the
number (i.e., 1) that’s added to rand % 6. We can generalize this result as follows:
n = a + rand() % b;
where a is the shifting value (which is equal to the first number in the desired range of
consecutive integers) and b is the scaling factor (which is equal to the width of the desired
range of consecutive integers). In the exercises, we’ll see that it’s possible to choose integers
at random from sets of values other than ranges of consecutive integers.
Common Programming Error 5.7
Using srand in place of rand to generate random numbers.
5.11 Example: A Game of Chance
One of the most popular games of chance is a dice game known as “craps,” which is played
in casinos and back alleys throughout the world. The rules of the game are straightforward:
A player rolls two dice. Each die has six faces. These faces contain 1, 2, 3, 4, 5, and 6
spots. After the dice have come to rest, the sum of the spots on the two upward faces is
calculated. If the sum is 7 or 11 on the first throw, the player wins. If the sum is 2, 3,
or 12 on the first throw (called “craps”), the player loses (i.e., the “house” wins). If the
sum is 4, 5, 6, 8, 9, or 10 on the first throw, then that sum becomes the player’s
“point.” To win, you must continue rolling the dice until you “make your point.” The
player loses by rolling a 7 before making the point.
Figure 5.14 simulates the game of craps and Fig. 5.15 shows several sample executions.
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
// Fig. 5.14: fig05_14.c
// Simulating the game of craps.
#include <stdio.h>
#include <stdlib.h>
#include <time.h> // contains prototype for function time
// enumeration constants represent game status
enum Status { CONTINUE, WON, LOST };
int rollDice( void ); // function prototype
// function main begins program execution
int main( void )
{
int sum; // sum of rolled dice
int myPoint; // player must make this point to win
enum Status gameStatus; // can contain CONTINUE, WON, or LOST
// randomize random number generator using current time
srand( time( NULL ) );
Fig. 5.14 | Simulating the game of craps. (Part 1 of 3.)
180
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
Chapter 5 C Functions
sum = rollDice(); // first roll of the dice
// determine game status based on sum of dice
switch( sum ) {
// win on first roll
case 7: // 7 is a winner
case 11: // 11 is a winner
gameStatus = WON; // game has been won
break;
// lose on first roll
case 2: // 2 is a loser
case 3: // 3 is a loser
case 12: // 12 is a loser
gameStatus = LOST; // game has been lost
break;
// remember point
default:
gameStatus = CONTINUE; // player should keep rolling
myPoint = sum; // remember the point
printf( "Point is %d\n", myPoint );
break; // optional
} // end switch
// while game not complete
while ( CONTINUE == gameStatus ) { // player should keep rolling
sum = rollDice(); // roll dice again
// determine game status
if ( sum == myPoint ) { // win by making point
gameStatus = WON; // game over, player won
} // end if
else {
if ( 7 == sum ) { // lose by rolling 7
gameStatus = LOST; // game over, player lost
} // end if
} // end else
} // end while
// display won or lost message
if ( WON == gameStatus ) { // did player win?
puts( "Player wins" );
} // end if
else { // player lost
puts( "Player loses" );
} // end else
} // end main
// roll dice, calculate sum and display results
int rollDice( void )
{
Fig. 5.14 | Simulating the game of craps. (Part 2 of 3.)
5.11 Example: A Game of Chance
76
77
78
79
80
81
82
83
84
85
86
87
181
int die1; // first die
int die2; // second die
int workSum; // sum of dice
die1 = 1 + ( rand() % 6 ); // pick random die1 value
die2 = 1 + ( rand() % 6 ); // pick random die2 value
workSum = die1 + die2; // sum die1 and die2
// display results of this roll
printf( "Player rolled %d + %d = %d\n", die1, die2, workSum );
return workSum; // return sum of dice
} // end function rollRice
Fig. 5.14 | Simulating the game of craps. (Part 3 of 3.)
Player wins on the first roll
Player rolled 5 + 6 = 11
Player wins
Player wins on a subsequent roll
Player rolled
Point is 5
Player rolled
Player rolled
Player rolled
Player wins
4 + 1 = 5
6 + 2 = 8
2 + 1 = 3
3 + 2 = 5
Player loses on the first roll
Player rolled 1 + 1 = 2
Player loses
Player loses on a subsequent roll
Player rolled 6 + 4 = 10
Point is 10
Player rolled 3 + 4 = 7
Player loses
Fig. 5.15 | Sample runs for the game of craps.
In the rules of the game, notice that the player must roll two dice on the first roll, and
must do so later on all subsequent rolls. We define a function rollDice to roll the dice
and compute and print their sum. Function rollDice is defined once, but it’s called from
two places in the program (lines 23 and 51). Interestingly, rollDice takes no arguments,
so we’ve indicated void in the parameter list (line 74). Function rollDice does return the
sum of the two dice, so a return type of int is indicated in its function header and in its
function prototype.
182
Chapter 5 C Functions
Enumerations
The game is reasonably involved. The player may win or lose on the first roll, or may win
or lose on any subsequent roll. Variable gameStatus, defined to be of a new type—enum
Status—stores the current status. Line 8 creates a programmer-defined type called an
enumeration. An enumeration, introduced by the keyword enum, is a set of integer constants represented by identifiers. Enumeration constants are sometimes called symbolic
constants. Values in an enum start with 0 and are incremented by 1. In line 8, the constant
CONTINUE has the value 0, WON has the value 1 and LOST has the value 2. It’s also possible to
assign an integer value to each identifier in an enum (see Chapter 10). The identifiers in an
enumeration must be unique, but the values may be duplicated.
Common Programming Error 5.8
Assigning a value to an enumeration constant after it has been defined is a syntax error.
Good Programming Practice 5.6
Use only uppercase letters in the names of enumeration constants to make these constants
stand out in a program and to indicate that enumeration constants are not variables.
When the game is won, either on the first roll or on a subsequent roll, gameStatus is
set to WON. When the game is lost, either on the first roll or on a subsequent roll, gameStatus is set to LOST. Otherwise gameStatus is set to CONTINUE and the game continues.
Game Ends on First Roll
After the first roll, if the game is over, the while statement (lines 50–62) is skipped because
gameStatus is not CONTINUE. The program proceeds to the if…else statement at lines
65–70, which prints "Player wins" if gameStatus is WON and "Player loses" otherwise.
Game Ends on a Subsequent Roll
After the first roll, if the game is not over, then sum is saved in myPoint. Execution proceeds
with the while statement because gameStatus is CONTINUE. Each time through the while,
rollDice is called to produce a new sum. If sum matches myPoint, gameStatus is set to WON
to indicate that the player won, the while-test fails, the if…else statement prints "Player wins" and execution terminates. If sum is equal to 7 (line 58), gameStatus is set to LOST
to indicate that the player lost, the while-test fails, the if…else statement prints "Player
loses" and execution terminates.
Control Architecture
Note the program’s interesting control architecture. We’ve used two functions—main and
rollDice—and the switch, while, nested if…else and nested if statements. In the exercises, we’ll investigate various interesting characteristics of the game of craps.
5.12 Storage Classes
In Chapters 2–4, we used identifiers for variable names. The attributes of variables include
name, type, size and value. In this chapter, we also use identifiers as names for user-defined
functions. Actually, each identifier in a program has other attributes, including storage
class, storage duration, scope and linkage.
5.12 Storage Classes
183
C provides the storage class specifiers auto, register1, extern and static.2 An
identifier’s storage class determines its storage duration, scope and linkage. An identifier’s
storage duration is the period during which the identifier exists in memory. Some exist
briefly, some are repeatedly created and destroyed, and others exist for the program’s entire
execution. An identifier’s scope is where the identifier can be referenced in a program.
Some can be referenced throughout a program, others from only portions of a program.
An identifier’s linkage determines for a multiple-source-file program whether the identifier is known only in the current source file or in any source file with proper declarations.
This section discusses storage classes and storage duration. Section 5.13 discusses scope.
Chapter 14 discusses identifier linkage and programming with multiple source files.
The storage-class specifiers can be split automatic storage duration and static storage
duration. Keyword auto is used to declare variables of automatic storage duration. Variables
with automatic storage duration are created when the block in which they’re defined is
entered; they exist while the block is active, and they’re destroyed when the block is exited.
Local Variables
Only variables can have automatic storage duration. A function’s local variables (those declared in the parameter list or function body) normally have automatic storage duration.
Keyword auto explicitly declares variables of automatic storage duration. Local variables
have automatic storage duration by default, so keyword auto is rarely used. For the remainder of the text, we’ll refer to variables with automatic storage duration simply as automatic
variables.
Performance Tip 5.1
Automatic storage is a means of conserving memory, because automatic variables exist
only when they’re needed. They’re created when a function is entered and destroyed when
the function is exited.
Static Storage Class
Keywords extern and static are used in the declarations of identifiers for variables and
functions of static storage duration. Identifiers of static storage duration exist from the time
at which the program begins execution until the program terminates. For static variables,
storage is allocated and initialized only once, before the program begins execution. For functions, the name of the function exists when the program begins execution. However, even
though the variables and the function names exist from the start of program execution, this
does not mean that these identifiers can be accessed throughout the program. Storage duration and scope (where a name can be used) are separate issues, as we’ll see in Section 5.13.
There are several types of identifiers with static storage duration: external identifiers
(such as global variables and function names) and local variables declared with the storageclass specifier static. Global variables and function names are of storage class extern by
default. Global variables are created by placing variable declarations outside any function
definition, and they retain their values throughout the execution of the program. Global
variables and functions can be referenced by any function that follows their declarations
1.
2.
Keyword register is archaic and should not be used.
The new C standard adds the storage class specifier _Thread_local, which is beyond this book’s
scope.
184
Chapter 5 C Functions
or definitions in the file. This is one reason for using function prototypes—when we
include stdio.h in a program that calls printf, the function prototype is placed at the
start of our file to make the name printf known to the rest of the file.
Software Engineering Observation 5.10
Defining a variable as global rather than local allows unintended side effects to occur
when a function that does not need access to the variable accidentally or maliciously
modifies it. In general, global variables should be avoided except in certain situations
with unique performance requirements (as discussed in Chapter 14).
Software Engineering Observation 5.11
Variables used only in a particular function should be defined as local variables in that
function rather than as external variables.
Local variables declared with the keyword static are still known only in the function
in which they’re defined, but unlike automatic variables, static local variables retain their
value when the function is exited. The next time the function is called, the static local
variable contains the value it had when the function last exited. The following statement
declares local variable count to be static and initializes it to 1.
static int count = 1;
All numeric variables of static storage duration are initialized to zero by default if you do
not explicitly initialize them.
Keywords extern and static have special meaning when explicitly applied to
external identifiers. In Chapter 14 we discuss the explicit use of extern and static with
external identifiers and multiple-source-file programs.
5.13 Scope Rules
The scope of an identifier is the portion of the program in which the identifier can be referenced. For example, when we define a local variable in a block, it can be referenced only
following its definition in that block or in blocks nested within that block. The four identifier scopes are function scope, file scope, block scope, and function-prototype scope.
Labels (identifiers followed by a colon such as start:) are the only identifiers with function scope. Labels can be used anywhere in the function in which they appear, but cannot
be referenced outside the function body. Labels are used in switch statements (as case labels) and in goto statements (see Chapter 14). Labels are implementation details that
functions hide from one another. This hiding—more formally called information hiding—is a means of implementing the principle of least privilege—a fundamental principle of good software engineering. In the context of an application, the principle states that
code should be granted only the amount of privilege and access that it needs to accomplish
its designated task, but no more.
An identifier declared outside any function has file scope. Such an identifier is
“known” (i.e., accessible) in all functions from the point at which the identifier is declared
until the end of the file. Global variables, function definitions, and function prototypes
placed outside a function all have file scope.
Identifiers defined inside a block have block scope. Block scope ends at the terminating right brace (}) of the block. Local variables defined at the beginning of a function
5.13 Scope Rules
185
have block scope, as do function parameters, which are considered local variables by the
function. Any block may contain variable definitions. When blocks are nested, and an identifier in an outer block has the same name as an identifier in an inner block, the identifier
in the outer block is hidden until the inner block terminates. This means that while executing in the inner block, the inner block sees the value of its own local identifier and not
the value of the identically named identifier in the enclosing block. Local variables
declared static still have block scope, even though they exist from before program
startup. Thus, storage duration does not affect the scope of an identifier.
The only identifiers with function-prototype scope are those used in the parameter
list of a function prototype. As mentioned previously, function prototypes do not require
names in the parameter list—only types are required. If a name is used in the parameter list
of a function prototype, the compiler ignores the name. Identifiers used in a function prototype can be reused elsewhere in the program without ambiguity.
Common Programming Error 5.9
Accidentally using the same name for an identifier in an inner block as is used for an identifier in an outer block, when in fact you want the identifier in the outer block to be active
for the duration of the inner block.
Error-Prevention Tip 5.4
Avoid variable names that hide names in outer scopes.
Figure 5.16 demonstrates scoping issues with global variables, automatic local variables and static local variables. A global variable x is defined and initialized to 1 (line 9).
This global variable is hidden in any block (or function) in which a variable named x is
defined. In main, a local variable x is defined and initialized to 5 (line 14). This variable is
then printed to show that the global x is hidden in main. Next, a new block is defined in
main with another local variable x initialized to 7 (line 19). This variable is printed to show
that it hides x in the outer block of main. The variable x with value 7 is automatically
destroyed when the block is exited, and the local variable x in the outer block of main is
printed again to show that it’s no longer hidden.
1
2
3
4
5
6
7
8
9
10
11
12
13
// Fig. 5.16: fig05_16.c
// Scoping.
#include <stdio.h>
void useLocal( void ); // function prototype
void useStaticLocal( void ); // function prototype
void useGlobal( void ); // function prototype
int x = 1; // global variable
// function main begins program execution
int main( void )
{
Fig. 5.16 | Scoping. (Part 1 of 3.)
186
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
Chapter 5 C Functions
int x = 5; // local variable to main
printf("local x in outer scope of main is %d\n", x );
{ // start new scope
int x = 7; // local variable to new scope
printf( "local x in inner scope of main is %d\n", x );
} // end new scope
printf( "local x in outer scope of main is %d\n", x );
useLocal(); // useLocal has automatic local x
useStaticLocal(); // useStaticLocal has static local x
useGlobal(); // useGlobal uses global x
useLocal(); // useLocal reinitializes automatic local x
useStaticLocal(); // static local x retains its prior value
useGlobal(); // global x also retains its value
printf( "\nlocal x in main is %d\n", x );
} // end main
// useLocal reinitializes local variable x during each call
void useLocal( void )
{
int x = 25; // initialized each time useLocal is called
printf( "\nlocal x in useLocal is %d after entering useLocal\n", x );
++x;
printf( "local x in useLocal is %d before exiting useLocal\n", x );
} // end function useLocal
// useStaticLocal initializes static local variable x only the first time
// the function is called; value of x is saved between calls to this
// function
void useStaticLocal( void )
{
// initialized once before program startup
static int x = 50;
printf( "\nlocal static x is %d on entering useStaticLocal\n", x );
++x;
printf( "local static x is %d on exiting useStaticLocal\n", x );
} // end function useStaticLocal
// function useGlobal modifies global variable x during each call
void useGlobal( void )
{
printf( "\nglobal x is %d on entering useGlobal\n", x );
x *= 10;
printf( "global x is %d on exiting useGlobal\n", x );
} // end function useGlobal
Fig. 5.16 | Scoping. (Part 2 of 3.)
5.14 Recursion
187
local x in outer scope of main is 5
local x in inner scope of main is 7
local x in outer scope of main is 5
local x in useLocal is 25 after entering useLocal
local x in useLocal is 26 before exiting useLocal
local static x is 50 on entering useStaticLocal
local static x is 51 on exiting useStaticLocal
global x is 1 on entering useGlobal
global x is 10 on exiting useGlobal
local x in useLocal is 25 after entering useLocal
local x in useLocal is 26 before exiting useLocal
local static x is 51 on entering useStaticLocal
local static x is 52 on exiting useStaticLocal
global x is 10 on entering useGlobal
global x is 100 on exiting useGlobal
local x in main is 5
Fig. 5.16 | Scoping. (Part 3 of 3.)
The program defines three functions that each take no arguments and return nothing.
Function useLocal defines an automatic variable x and initializes it to 25 (line 39). When
useLocal is called, the variable is printed, incremented, and printed again before exiting
the function. Each time this function is called, automatic variable x is reinitialized to 25.
Function useStaticLocal defines a static variable x and initializes it to 50 in line 52
(recall that the storage for static variables is allocated and initialized only once, before the
program begins execution). Local variables declared as static retain their values even when
they’re out of scope. When useStaticLocal is called, x is printed, incremented, and
printed again before exiting the function. In the next call to this function, static local
variable x will contain the value 51. Function useGlobal does not define any variables.
Therefore, when it refers to variable x, the global x (line 9) is used. When useGlobal is
called, the global variable is printed, multiplied by 10, and printed again before exiting the
function. The next time function useGlobal is called, the global variable still has its modified value, 10. Finally, the program prints the local variable x in main again (line 33) to
show that none of the function calls modified the value of x because the functions all
referred to variables in other scopes.
5.14 Recursion
The programs we’ve discussed are generally structured as functions that call one another
in a disciplined, hierarchical manner. For some types of problems, it’s useful to have functions call themselves. A recursive function is a function that calls itself either directly or
indirectly through another function. Recursion is a complex topic discussed at length in
upper-level computer science courses. In this section and the next, simple examples of re-
188
Chapter 5 C Functions
cursion are presented. This book contains an extensive treatment of recursion, which is
spread throughout Chapters 5–8 and 12 and Appendix F. Figure 5.21, in Section 5.16,
summarizes the recursion examples and exercises in the book.
We consider recursion conceptually first, then examine several programs containing
recursive functions. Recursive problem-solving approaches have a number of elements in
common. A recursive function is called to solve a problem. The function actually knows
how to solve only the simplest case(s), or so-called base case(s). If the function is called with
a base case, the function simply returns a result. If the function is called with a more complex problem, the function divides the problem into two conceptual pieces: a piece that
the function knows how to do and a piece that it does not know how to do. To make
recursion feasible, the latter piece must resemble the original problem, but be a slightly
simpler or smaller version. Because this new problem looks like the original problem, the
function launches (calls) a fresh copy of itself to go to work on the smaller problem—this
is referred to as a recursive call or the recursion step. The recursion step also includes the
keyword return, because its result will be combined with the portion of the problem the
function knew how to solve to form a result that will be passed back to the original caller.
The recursion step executes while the original call to the function has not yet finished
executing. The recursion step can result in many more such recursive calls, as the function
keeps dividing each problem it’s called with into two conceptual pieces. For the recursion
to terminate, each time the function calls itself with a slightly simpler version of the original problem, this sequence of smaller problems must eventually converge on the base case.
When the function recognizes the base case, it returns a result to the previous copy of the
function, and a sequence of returns ensues all the way up the line until the original call of
the function eventually returns the final result to main. All of this sounds quite exotic compared to the kind of problem solving we’ve been using with conventional function calls to
this point. It can take a great deal of practice writing recursive programs before the process
will appear natural. As an example of these concepts at work, let’s write a recursive program to perform a popular mathematical calculation.
Recursively Calculating Factorials
The factorial of a nonnegative integer n, written n! (pronounced “n factorial”), is the product
n · (n – 1) · (n – 2) · … · 1
with 1! equal to 1, and 0! defined to be 1. For example, 5! is the product 5 * 4 * 3 * 2 * 1,
which is equal to 120.
The factorial of an integer, number, greater than or equal to 0 can be calculated
iteratively (nonrecursively) using a for statement as follows:
factorial = 1;
for ( counter = number; counter >= 1; --counter )
factorial *= counter;
A recursive definition of the factorial function is arrived at by observing the following
relationship:
n! = n · (n – 1)!
5.14 Recursion
189
For example, 5! is clearly equal to 5 * 4! as is shown by the following:
5! = 5 · 4 · 3 · 2 · 1
5! = 5 · (4 · 3 · 2 · 1)
5! = 5 · (4!)
The evaluation of 5! would proceed as shown in Fig. 5.17. Figure 5.17(a) shows how
the succession of recursive calls proceeds until 1! is evaluated to be 1 (i.e., the base case),
which terminates the recursion. Figure 5.17(b) shows the values returned from each recursive call to its caller until the final value is calculated and returned.
Final value = 120
5!
5!
5! = 5 * 24 = 120 is returned
5 * 4!
5 * 4!
4! = 4 * 6 = 24 is returned
4 * 3!
4 * 3!
3! = 3 * 2 = 6 is returned
3 * 2!
3 * 2!
2! = 2 * 1 = 2 is returned
2 * 1!
2 * 1!
1 is returned
1
(a) Sequence of recursive calls
1
(b) Values returned from each recursive call
Fig. 5.17 | Recursive evaluation of 5!.
Figure 5.18 uses recursion to calculate and print the factorials of the integers 0–10
(the choice of the type unsigned long long int will be explained momentarily).
1
2
3
4
5
6
7
8
9
10
11
// Fig. 5.18: fig05_18.c
// Recursive factorial function.
#include <stdio.h>
unsigned long long int factorial( unsigned int number );
// function main begins program execution
int main( void )
{
unsigned int i; // counter
Fig. 5.18 | Recursive factorial function. (Part 1 of 2.)
190
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
Chapter 5 C Functions
// during each iteration, calculate
// factorial( i ) and display result
for ( i = 0; i <= 21; ++i ) {
printf( "%u! = %llu\n", i, factorial( i ) );
} // end for
} // end main
// recursive definition of function factorial
unsigned long long int factorial( unsigned int number )
{
// base case
if ( number <= 1 ) {
return 1;
} // end if
else { // recursive step
return ( number * factorial( number - 1 ) );
} // end else
} // end function factorial
0! = 1
1! = 1
2! = 2
3! = 6
4! = 24
5! = 120
6! = 720
7! = 5040
8! = 40320
9! = 362880
10! = 3628800
11! = 39916800
12! = 479001600
13! = 6227020800
14! = 87178291200
15! = 1307674368000
16! = 20922789888000
17! = 355687428096000
18! = 6402373705728000
19! = 121645100408832000
20! = 2432902008176640000
21! = 14197454024290336768
Fig. 5.18 | Recursive factorial function. (Part 2 of 2.)
The recursive factorial function first tests whether a terminating condition is true,
i.e., whether number is less than or equal to 1. If number is indeed less than or equal to 1,
factorial returns 1, no further recursion is necessary, and the program terminates. If
number is greater than 1, the statement
return number * factorial( number - 1 );
expresses the problem as the product of number and a recursive call to factorial evaluating the factorial of number - 1. The call factorial( number - 1 ) is a slightly simpler problem than the original calculation factorial( number ).
5.15 Example Using Recursion: Fibonacci Series
191
Function factorial (lines 20–29) receives an unsigned int and returns a result of
type unsigned long long int. The C standard specifies that a variable of type unsigned
long long int can hold a value at least as large as 18,446,744,073,709,551,615. As can
be seen in Fig. 5.18, factorial values become large quickly. We’ve chosen the data type
unsigned long long int so the program can calculate larger factorial values. The conversion specifier %llu is used to print unsigned long long int values. Unfortunately, the
factorial function produces large values so quickly that even unsigned long long int
does not help us print very many factorial values before the maximum value of a unsigned
long long int variable is exceeded.
Even when we use unsigned long long int, we still can’t calculate factorials beyond
21! This points to a weakness in C (and most other procedural programming languages)—
namely that the language is not easily extended to handle the unique requirements of various applications. As we’ll see later in the book, C++ is an extensible language that, through
“classes,” allows us to create new data types, including ones that could hold arbitrarily large
integers if we wish.
Common Programming Error 5.10
Forgetting to return a value from a recursive function when one is needed.
Common Programming Error 5.11
Either omitting the base case, or writing the recursion step incorrectly so that it does not
converge on the base case, will cause infinite recursion, eventually exhausting memory.
This is analogous to the problem of an infinite loop in an iterative (nonrecursive) solution.
Infinite recursion can also be caused by providing an unexpected input.
5.15 Example Using Recursion: Fibonacci Series
The Fibonacci series
0, 1, 1, 2, 3, 5, 8, 13, 21, …
begins with 0 and 1 and has the property that each subsequent Fibonacci number is the
sum of the previous two Fibonacci numbers.
The series occurs in nature and, in particular, describes a form of spiral. The ratio of
successive Fibonacci numbers converges to a constant value of 1.618…. This number, too,
repeatedly occurs in nature and has been called the golden ratio or the golden mean.
Humans tend to find the golden mean aesthetically pleasing. Architects often design windows, rooms, and buildings whose length and width are in the ratio of the golden mean.
Postcards are often designed with a golden mean length/width ratio.
The Fibonacci series may be defined recursively as follows:
fibonacci(0) = 0
fibonacci(1) = 1
fibonacci(n) = fibonacci(n – 1) + fibonacci(n – 2)
Figure 5.19 calculates the nth Fibonacci number recursively using function fibonacci.
Notice that Fibonacci numbers tend to become large quickly. Therefore, we’ve chosen the
data type unsigned int for the parameter type and the data type unsigned long long int
192
Chapter 5 C Functions
for the return type in function fibonacci. In Fig. 5.19, each pair of output lines shows a
separate run of the program.
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
// Fig. 5.19: fig05_19.c
// Recursive fibonacci function
#include <stdio.h>
unsigned long long int fibonacci( unsigned int n ); // function prototype
// function main begins program execution
int main( void )
{
unsigned long long int result; // fibonacci value
unsigned int number; // number input by user
// obtain integer from user
printf( "%s", "Enter an integer: " );
scanf( "%u", &number );
// calculate fibonacci value for number input by user
result = fibonacci( number );
// display result
printf( "Fibonacci( %u ) = %llu\n", number, result );
} // end main
// Recursive definition of function fibonacci
unsigned long long int fibonacci( unsigned int n )
{
// base case
if ( 0 == n || 1 == n ) {
return n;
} // end if
else { // recursive step
return fibonacci( n - 1 ) + fibonacci( n - 2 );
} // end else
} // end function fibonacci
Enter an integer: 0
Fibonacci( 0 ) = 0
Enter an integer: 1
Fibonacci( 1 ) = 1
Enter an integer: 2
Fibonacci( 2 ) = 1
Enter an integer: 3
Fibonacci( 3 ) = 2
Fig. 5.19 | Recursive fibonacci function. (Part 1 of 2.)
5.15 Example Using Recursion: Fibonacci Series
193
Enter an integer: 10
Fibonacci( 10 ) = 55
Enter an integer: 20
Fibonacci( 20 ) = 6765
Enter an integer: 30
Fibonacci( 30 ) = 832040
Enter an integer: 40
Fibonacci( 40 ) = 102334155
Fig. 5.19 | Recursive fibonacci function. (Part 2 of 2.)
The call to fibonacci from main is not a recursive call (line 18), but all subsequent
calls to fibonacci are recursive (line 32). Each time fibonacci is invoked, it immediately
tests for the base case—n is equal to 0 or 1. If this is true, n is returned. Interestingly, if n
is greater than 1, the recursion step generates two recursive calls, each a slightly simpler
problem than the original call to fibonacci. Figure 5.20 shows how function fibonacci
would evaluate fibonacci(3).
fibonacci( 3 )
return
return
fibonacci( 1 )
return 1
fibonacci( 2 )
+
fibonacci( 0 )
+
fibonacci( 1 )
return 1
return 0
Fig. 5.20 | Set of recursive calls for fibonacci( 3 ).
Order of Evaluation of Operands
This figure raises some interesting issues about the order in which C compilers will evaluate
the operands of operators. This is a different issue from the order in which operators are
applied to their operands, namely the order dictated by the rules of operator precedence.
Figure 5.20 shows that while evaluating fibonacci(3), two recursive calls will be made,
namely fibonacci(2) and fibonacci(1). But in what order will these calls be made? You
might simply assume the operands will be evaluated left to right. For optimization reasons,
194
Chapter 5 C Functions
C does not specify the order in which the operands of most operators (including +) are to
be evaluated. Therefore, you should make no assumption about the order in which these
calls will execute. The calls could in fact execute fibonacci(2) first and then fibonacci(1), or the calls could execute in the reverse order, fibonacci(1) then fibonacci(2).
In this and most other programs, the final result would be the same. But in some programs
the evaluation of an operand may have side effects that could affect the final result of the
expression. C specifies the order of evaluation of the operands of only four operators—
namely &&, ||, the comma (,) operator and ?:. The first three of these are binary operators
whose operands are guaranteed to be evaluated left to right. [Note: The commas used to
separate the arguments in a function call are not comma operators.] The last operator is
C’s only ternary operator. Its leftmost operand is always evaluated first; if the leftmost operand evaluates to nonzero, the middle operand is evaluated next and the last operand is
ignored; if the leftmost operand evaluates to zero, the third operand is evaluated next and
the middle operand is ignored.
Common Programming Error 5.12
Writing programs that depend on the order of evaluation of the operands of operators other than &&, ||, ?:, and the comma (,) operator can lead to errors because compilers may
not necessarily evaluate the operands in the order you expect.
Portability Tip 5.2
Programs that depend on the order of evaluation of the operands of operators other than
&&, ||, ?:, and the comma (,) operator can function differently on different compilers.
Exponential Complexity
A word of caution is in order about recursive programs like the one we use here to generate
Fibonacci numbers. Each level of recursion in the fibonacci function has a doubling effect
on the number of calls—the number of recursive calls that will be executed to calculate the
nth Fibonacci number is on the order of 2n. This rapidly gets out of hand. Calculating only
the 20th Fibonacci number would require on the order of 220 or about a million calls, calculating the 30th Fibonacci number would require on the order of 230 or about a billion
calls, and so on. Computer scientists refer to this as exponential complexity. Problems of this
nature humble even the world’s most powerful computers! Complexity issues in general,
and exponential complexity in particular, are discussed in detail in the upper-level computer science course generally called “Algorithms.”
The example we showed in this section used an intuitively appealing solution to calculate Fibonacci numbers, but there are better approaches. Exercise 5.48 asks you to investigate recursion in more depth and propose alternate approaches to implementing the
recursive Fibonacci algorithm.
5.16 Recursion vs. Iteration
In the previous sections, we studied two functions that can easily be implemented either
recursively or iteratively. In this section, we compare the two approaches and discuss why
you might choose one approach over the other in a particular situation.
•
Both iteration and recursion are based on a control structure: Iteration uses a
repetition structure; recursion uses a selection structure.
5.16 Recursion vs. Iteration
195
•
Both iteration and recursion involve repetition: Iteration explicitly uses a repetition statement; recursion achieves repetition through repeated function calls.
•
Iteration and recursion each involve a termination test: Iteration terminates when
the loop-continuation condition fails; recursion when a base case is recognized.
•
Iteration with counter-controlled repetition and recursion each gradually approach termination: Iteration keeps modifying a counter until the counter assumes a value that makes the loop-continuation condition fail; recursion keeps
producing simpler versions of the original problem until the base case is reached.
•
Both iteration and recursion can occur infinitely: An infinite loop occurs with iteration if the loop-continuation test never becomes false; infinite recursion occurs
if the recursion step does not reduce the problem each time in a manner that converges on the base case.
Recursion has many negatives. It repeatedly invokes the mechanism, and consequently
the overhead, of function calls. This can be expensive in both processor time and memory
space. Each recursive call causes another copy of the function (actually only the function’s
variables) to be created; this can consume considerable memory. Iteration normally occurs
within a function, so the overhead of repeated function calls and extra memory assignment
is omitted. So why choose recursion?
Software Engineering Observation 5.12
Any problem that can be solved recursively can also be solved iteratively (nonrecursively).
A recursive approach is normally chosen in preference to an iterative approach when the
recursive approach more naturally mirrors the problem and results in a program that’s
easier to understand and debug. Another reason to choose a recursive solution is that an
iterative solution may not be apparent.
Most programming textbooks introduce recursion much later than we’ve done here.
We feel that recursion is a sufficiently rich and complex topic that it’s better to introduce
it earlier and spread the examples over the remainder of the text. Figure 5.21 summarizes
by chapter the 31 recursion examples and exercises in the text.
Chapter
Recursion examples and exercises
Chapter 5
Factorial function
Fibonacci function
Greatest common divisor
Sum of two integers
Multiply two integers
Raising an integer to an integer power
Towers of Hanoi
Recursive main
Printing keyboard inputs in reverse
Visualizing recursion
Fig. 5.21 | Recursion examples and exercises in the text. (Part 1 of 2.)
196
Chapter 5 C Functions
Chapter
Recursion examples and exercises
Chapter 6
Sum the elements of an array
Print an array
Print an array backward
Print a string backward
Check whether a string is a palindrome
Minimum value in an array
Linear search
Binary search
Eight Queens
Maze traversal
Printing a string input at the keyboard backward
Linked list insert
Linked list delete
Search a linked list
Print a linked list backward
Binary tree insert
Preorder traversal of a binary tree
Inorder traversal of a binary tree
Postorder traversal of a binary tree
Selection sort
Quicksort
Chapter 7
Chapter 8
Chapter 12
Appendix E
Fig. 5.21 | Recursion examples and exercises in the text. (Part 2 of 2.)
Let’s close this chapter with some observations that we make repeatedly throughout the
book. Good software engineering is important. High performance is important. Unfortunately, these goals are often at odds with one another. Good software engineering is key to
making more manageable the task of developing the larger and more complex software systems we need. High performance is key to realizing the systems of the future that will place
ever greater computing demands on hardware. Where do functions fit in here?
Performance Tip 5.2
Functionalizing programs promotes good software engineering. But it has a price. A heavily functionalized program—as compared to a monolithic (i.e., one-piece) program without functions—makes potentially large numbers of function calls, and these consume
execution time on a computer’s processor(s). Although monolithic programs may perform
better, they’re more difficult to program, test, debug, maintain, and evolve.
Performance Tip 5.3
Today’s hardware architectures are tuned to make function calls efficient, and today’s
hardware processors are incredibly fast. For the vast majority of applications and software
systems you’ll build, concentrating on good software engineering will be more important
than programming for high performance. Nevertheless, in many C applications and systems, such as game programming, real-time systems, operating systems and embedded systems, performance is crucial, so we include performance tips throughout the book.
5.17 Secure C Programming
197
5.17 Secure C Programming
Secure Random Numbers
In Section 5.10, we introduced the rand function for generating pseudorandom numbers.
The C standard library does not provide a secure random-number generator. According to
the C standard document’s description of function rand, “There are no guarantees as to the
quality of the random sequence produced and some implementations are known to produce
sequences with distressingly non-random low-order bits.” The CERT guideline MSC30-C
indicates that implementation-specific random-number generation functions must be used
to ensure that the random numbers produced are not predictable—this is extremely important, for example, in cryptography and other security applications. The guideline presents
several platform-specific random-number generators that are considered to be secure. For example, Microsoft Windows provides the CryptGenRandom function, and POSIX based systems (such as Linux) provide a random function that produces more secure results. For more
information, see guideline MSC30-C at https://www.securecoding.cert.org. If you’re
building industrial-strength applications that require random numbers, you should investigate for your platform the recommended function(s) to use.
Summary
Section 5.1 Introduction
• The best way to develop and maintain a large program is to divide it into several smaller program
modules, each more manageable than the original program. Modules are written as functions in
C.
Section 5.2 Program Modules in C
• A function is invoked by a function call. The function call mentions the function by name and
provides information (as arguments) that the called function needs to perform its task.
• The purpose of information hiding is to give functions access only to the information they need
to complete their tasks. This is a means of implementing the principle of least privilege, one of
the most important principles of good software engineering.
Section 5.3 Math Library Functions
• A function os normally invoked in a program by writing the function’s name followed by a left
parenthesis followed by the argument (or a comma-separated list of arguments) of the function
followed by a right parenthesis.
• Data type double is a floating-point type like float. A variable of type double can store a value
of much greater magnitude and precision than float can store.
• Each argument of a function may be a constant, a variable, or an expression.
Section 5.4 Functions
• A local variable is known only in a function definition. Other functions are not allowed to know
the names of a function’s local variables, nor is any function allowed to know the implementation
details of any other function.
198
Chapter 5 C Functions
Section 5.5 Function Definitions
• The general format for a function definition is
return-value-type function-name( parameter-list
)
{
definitions
statements
}
The return-value-type states the type of the value returned to the calling function. If a function does
not return a value, the return-value-type is declared as void. The function-name is any valid identifier. The parameter-list is a comma-separated list containing the definitions of the variables that will
be passed to the function. If a function does not receive any values, parameter-list is declared as void.
The function body is the set of definitions and statements that constitute the function.
• The arguments passed to a function should match in number, type and order with the parameters
in the function definition.
• When a program encounters a function call, control is transferred from the point of invocation
to the called function, the statements of the called function are executed and control returns to
the caller.
• A called function can return control to the caller in one of three ways. If the function does not
return a value, control is returned when the function-ending right brace is reached, or by executing the statement
return;
If the function does return a value, the statement
return
expression;
returns the value of expression.
Section 5.6 Function Prototypes: A Deeper Look
• A function prototype declares the function’s return type and declares the number, types, and order of the parameters the function expects to receive.
• Function prototypes enable the compiler to verify that functions are called correctly.
• The compiler ignores variable names mentioned in the function prototype.
• Arguments in a mixed-type expression are converted to the same type via the C standard’s usual
arithmetic conversion rules.
Section 5.7 Function Call Stack and Stack Frames
• Stacks are known as last-in, first-out (LIFO) data structures—the last item pushed (inserted) on
the stack is the first item popped (removed) from the stack.
• A called function must know how to return to its caller, so the return address of the calling function is pushed onto the program execution stack when the function is called. If a series of function calls occurs, the successive return addresses are pushed onto the stack in last-in, first-out
order so that the last function to execute will be the first to return to its caller.
• The program execution stack contains the memory for the local variables used in each invocation
of a function during a program’s execution. This data is known as the stack frame of the function
call. When a function call is made, the stack frame for that function call is pushed onto the program execution stack. When the function returns to its caller, the stack frame for this function
call is popped off the stack and those local variables are no longer known to the program.
• The amount of memory in a computer is finite, so only a certain amount of memory can be used
to store stack frames on the program execution stack. If there are more function calls than can
Summary
199
have their stack frames stored on the program execution stack, an error known as a stack overflow
occurs. The application will compile correctly, but its execution causes a stack overflow.
Section 5.8 Headers
• Each standard library has a corresponding header containing the function prototypes for all the
functions in that library, as well as definitions of various symbolic constants needed by those
functions.
• You can create and include your own headers.
Section 5.9 Passing Arguments By Value and By Reference
• When an argument is passed by value, a copy of the variable’s value is made and the copy is
passed to the called function. Changes to the copy in the called function do not affect the original
variable’s value.
• All calls in C are call-by-value.
• It’s possible to simulate call-by-reference by using address operators and indirection operators.
Section 5.10 Random Number Generation
• Function rand generates an integer between 0 and RAND_MAX which is defined by the C standard
to be at least 32767.
• The function prototypes for rand and srand are contained in <stdlib.h>.
• Values produced by rand can be scaled and shifted to produce values in a specific range.
• To randomize a program, use the C standard library function srand.
• The srand function call is ordinarily inserted in a program only after it has been thoroughly debugged. While debugging, it’s better to omit srand. This ensures repeatability, which is essential
to proving that corrections to a random number generation program work properly.
• To randomize without the need for entering a seed each time, we use srand(time(NULL)).
• The general equation for scaling and shifting a random number is
n = a + rand() % b;
where a is the shifting value (i.e., the first number in the desired range of consecutive integers)
and b is the scaling factor (i.e,. the width of the desired range of consecutive integers).
Section 5.11 Example: A Game of Chance
• An enumeration, introduced by the keyword enum, is a set of integer constants represented by
identifiers. Values in an enum start with 0 and are incremented by 1. It’s also possible to assign an
integer value to each identifier in an enum. The identifiers in an enumeration must be unique,
but the values may be duplicated.
Section 5.12 Storage Classes
• Each identifier in a program has the attributes storage class, storage duration, scope and linkage.
• C provides four storage classes indicated by the storage class specifiers: auto, register, extern
and static.
• An identifier’s storage duration is when that identifier exists in memory.
Section 5.13 Scope Rules
• An identifier’s scope is where the identifier can be referenced in a program.
• An identifier’s linkage determines for a multiple-source-file program whether an identifier is
known only in the current source file or in any source file with proper declarations.
200
Chapter 5 C Functions
• Variables with automatic storage duration are created when the block in which they’re defined is
entered, exist while the block is active and are destroyed when the block is exited. A function’s
local variables normally have automatic storage duration.
• Keywords extern and static are used to declare identifiers for variables and functions of static
storage duration.
• Static storage duration variables are allocated and initialized once, before the program begins execution.
• There are two types of identifiers with static storage duration: external identifiers (such as global
variables and function names) and local variables declared with the storage-class specifier static.
• Global variables are created by placing variable definitions outside any function definition. Global variables retain their values throughout the execution of the program.
• Local static variables retain their value between calls to the function in which they’re defined.
• All numeric variables of static storage duration are initialized to zero if you do not explicitly initialize them.
• An identifier can have function scope, file scope, block scope or function-prototype scope.
• Labels are the only identifiers with function scope. Labels can be used anywhere in the function
in which they appear but cannot be referenced outside the function body.
• An identifier declared outside any function has file scope. Such an identifier is “known” in all
functions from the point at which it’s declared until the end of the file.
• Identifiers defined inside a block have block scope. Block scope ends at the terminating right
brace (}) of the block.
• Local variables defined at the beginning of a function have block scope, as do function parameters, which are considered local variables by the function.
• Any block may contain variable definitions. When blocks are nested, and an identifier in an outer
block has the same name as an identifier in an inner block, the identifier in the outer block is
“hidden” until the inner block terminates.
• The only identifiers with function-prototype scope are those used in the parameter list of a function prototype. Identifiers used in a function prototype can be reused elsewhere in the program
without ambiguity.
Section 5.14 Recursion
• A recursive function is a function that calls itself either directly or indirectly.
• If a recursive function is called with a base case, the function simply returns a result. If it’s called
with a more complex problem, the function divides the problem into two conceptual pieces: a
piece that the function knows how to do and a slightly smaller version of the original problem.
Because this new problem looks like the original problem, the function launches a recursive call
to work on the smaller problem.
• For recursion to terminate, each time the recursive function calls itself with a slightly simpler version of the original problem, the sequence of smaller and smaller problems must converge on the
base case. When the function recognizes the base case, the result is returned to the previous function call, and a sequence of returns ensues all the way up the line until the original call of the
function eventually returns the final result.
• Standard C does not specify the order in which the operands of most operators (including +) are
to be evaluated. Of C’s many operators, the standard specifies the order of evaluation of the operands of only the operators &&, ||, the comma (,) operator and ?:. The first three of these are
binary operators whose two operands are evaluated left to right. The last operator is C’s only ter-
Terminology
201
nary operator. Its leftmost operand is evaluated first; if it evaluates to nonzero, the middle operand is evaluated next and the last operand is ignored; if the leftmost operand evaluates to zero,
the third operand is evaluated next and the middle operand is ignored.
Section 5.16 Recursion vs. Iteration
• Both iteration and recursion are based on a control structure: Iteration uses a repetition structure;
recursion uses a selection structure.
• Both iteration and recursion involve repetition: Iteration explicitly uses a repetition structure; recursion achieves repetition through repeated function calls.
• Iteration and recursion each involve a termination test: Iteration terminates when the loop-continuation condition fails; recursion terminates when a base case is recognized.
• Iteration and recursion can occur infinitely: An infinite loop occurs with iteration if the loopcontinuation test never becomes false; infinite recursion occurs if the recursion step does not reduce the problem in a manner that converges on the base case.
• Recursion repeatedly invokes the mechanism, and consequently the overhead, of function calls.
This can be expensive in both processor time and memory space.
Terminology
abstraction 162
argument (of a function) 160
auto 183
automatic storage duration 183
automatic variable 183
base case 188
block 164
block scope 184
C standard library 159
call a function 160
called 163
called function 160
caller 160
calling function 160
coercion of arguments 167
divide and conquer 159
enum 182
enumeration 182
enumeration constant 182
file scope 184
function 159
function body 164
function call 160
function call stack 169
function prototype 163
function-prototype scope 185
function scope 184
header 164
information hiding 184
invoke a function 160
last-in, first-out (LIFO) 169
linkage of an identifier 183
local variable 162
mixed-type expression 167
module 159
parameter 162
parameter list 164
pass-by-reference 173
pass-by-value 173
pop off a stack 169
principle of least privilege 184
program execution stack 169
programmer-defined function 160
pseudorandom numbers 177
push onto a stack 169
randomizing 177
recursion step 188
recursive call 188
recursive function 187
return from a function 160
return statement 163
return value type 164
scaling 174
scaling factor 174
scope 183
scope of an identifier 184
seed the rand function 177
shift 174
shifting value 179
side effect 173
simulate 174
software reusability 162
202
Chapter 5 C Functions
stack 169
stack frame 169
stack overflow 170
standard library header 172
static 183
static storage duration 183
storage class of an identifier 183
storage class specifier 183
storage duration 183
usual arithmetic conversion rules 167
Self-Review Exercises
5.1
Answer each of the following:
.
a) A program module in C is called a(n)
b) A function is invoked with a(n)
.
c) A variable that’s known only within the function in which it’s defined is called a(n)
.
d) The
statement in a called function is used to pass the value of an expression
back to the calling function.
is used in a function header to indicate that a function does not ree) Keyword
turn a value or to indicate that a function contains no parameters.
of an identifier is the portion of the program in which the identifier can
f) The
be used.
g) The three ways to return control from a called function to a caller are
,
and
.
h) A(n)
allows the compiler to check the number, types, and order of the arguments passed to a function.
function is used to produce random numbers.
i) The
j) The
function is used to set the random number seed to randomize a program.
k) The storage-class specifiers are
,
,
and
.
l) Variables declared in a block or in the parameter list of a function are assumed to be of
unless specified otherwise.
storage class
m) A non-static variable defined outside any block or function is a(n)
variable.
n) For a local variable in a function to retain its value between calls to the function, it must
storage-class specifier.
be declared with the
o) The four possible scopes of an identifier are
,
,
and
.
p) A function that calls itself either directly or indirectly is a(n)
function.
q) A recursive function typically has two components: one that provides a means for the
case, and one that expresses the
recursion to terminate by testing for a(n)
problem as a recursive call for a slightly simpler problem than the original call.
5.2
For the following program, state the scope (either function scope, file scope, block scope or
function-prototype scope) of each of the following elements.
a) The variable x in main.
b) The variable y in cube.
c) The function cube.
d) The function main.
e) The function prototype for cube.
f) The identifier y in the function prototype for cube.
1
2
3
#include <stdio.h>
int cube( int y );
Self-Review Exercises
4
5
6
7
8
9
10
11
12
13
14
15
203
int main( void )
{
int x;
for ( x = 1; x <= 10; ++x )
printf( "%d\n", cube( x ) );
}
int cube( int y )
{
return y * y * y;
}
5.3
Write a program that tests whether the examples of the math library function calls shown
in Fig. 5.2 actually produce the indicated results.
5.4
Give the function header for each of the following functions.
a) Function hypotenuse that takes two double-precision floating-point arguments, side1
and side2, and returns a double-precision floating-point result.
b) Function smallest that takes three integers, x, y, z, and returns an integer.
c) Function instructions that does not receive any arguments and does not return a value. [Note: Such functions are commonly used to display instructions to a user.]
d) Function intToFloat that takes an integer argument, number, and returns a floatingpoint result.
5.5
Give the function prototype for each of the following:
a) The function described in Exercise 5.4(a).
b) The function described in Exercise 5.4(b).
c) The function described in Exercise 5.4(c).
d) The function described in Exercise 5.4(d).
5.6
Write a declaration for floating-point variable lastVal that’s to retain its value between calls
to the function in which it’s defined.
5.7
Find the error in each of the following program segme nts and explain how the error can be
corrected (see also Exercise 5.46):
a) int g( void )
{
printf( "%s", Inside function g\n" );
int h( void )
{
printf( "%s", Inside function h\n" );
}
}
b)
int sum( int x, int y )
{
int result;
result = x + y;
}
c)
void f( float a );
{
float a;
printf( "%f", a );
}
204
Chapter 5 C Functions
d)
int sum( int n )
{
if ( 0 == n ) {
return 0; //
}
else {
n + sum( n - 1 );
}
}
e)
void product( void )
{
int a, b, c, result;
printf( "%s", "Enter three integers: " )
scanf( "%d%d%d", &a, &b, &c );
result = a * b * c;
printf( "Result is %d", result );
return result;
}
Answers to Self-Review Exercises
a) function. b) function call. c) local variable. d) return. e) void. f) Scope. g) return; or
or encountering the closing right brace of a function. h) function prototype.
i) rand. j) srand. k) auto, register, extern, static. l) auto. m) external, global. n) static.
o) function scope, file scope, block scope, function-prototype scope. p) recursive. q) base.
5.1
return expression;
5.2
a) Block scope. b) Block Scope. c) File scope. d) File scope. e) File scope. f) Function-prototype scope.
5.3
See below. [Note: On most Linux systems, you must use the -lm option when compiling
this program.]
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
// ex05_03.c
// Testing the math library functions
#include <stdio.h>
#include <math.h>
// function main begins program execution
int main( void )
{
// calculates and outputs the square root
printf( "sqrt(%.1f) = %.1f\n", 900.0, sqrt( 900.0 ) );
printf( "sqrt(%.1f) = %.1f\n", 9.0, sqrt( 9.0 ) );
// calculates and outputs the exponential function e to the x
printf( "exp(%.1f) = %f\n", 1.0, exp( 1.0 ) );
printf( "exp(%.1f) = %f\n", 2.0, exp( 2.0 ) );
// calculates and outputs the logarithm (base e)
printf( "log(%f) = %.1f\n", 2.718282, log( 2.718282 ) );
printf( "log(%f) = %.1f\n", 7.389056, log( 7.389056 ) );
// calculates and outputs the logarithm (base 10)
printf( "log10(%.1f) = %.1f\n", 1.0, log10( 1.0 ) );
printf( "log10(%.1f) = %.1f\n", 10.0, log10( 10.0 ) );
printf( "log10(%.1f) = %.1f\n", 100.0, log10( 100.0 ) );
Answers to Self-Review Exercises
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
// calculates and outputs the absolute value
printf( "fabs(%.1f) = %.1f\n", 13.5, fabs( 13.5 ) );
printf( "fabs(%.1f) = %.1f\n", 0.0, fabs( 0.0 ) );
printf( "fabs(%.1f) = %.1f\n", -13.5, fabs( -13.5 ) );
// calculates and outputs ceil( x )
printf( "ceil(%.1f) = %.1f\n", 9.2, ceil( 9.2 ) );
printf( "ceil(%.1f) = %.1f\n", -9.8, ceil( -9.8 ) );
// calculates and outputs floor( x )
printf( "floor(%.1f) = %.1f\n", 9.2, floor( 9.2 ) );
printf( "floor(%.1f) = %.1f\n", -9.8, floor( -9.8 ) );
// calculates and outputs pow( x, y )
printf( "pow(%.1f, %.1f) = %.1f\n", 2.0, 7.0, pow( 2.0, 7.0 ) );
printf( "pow(%.1f, %.1f) = %.1f\n", 9.0, 0.5, pow( 9.0, 0.5 ) );
// calculates and outputs fmod( x, y )
printf( "fmod(%.3f/%.3f) = %.3f\n", 13.657, 2.333,
fmod( 13.657, 2.333 ) );
// calculates and outputs sin( x )
printf( "sin(%.1f) = %.1f\n", 0.0, sin( 0.0 ) );
// calculates and outputs cos( x )
printf( "cos(%.1f) = %.1f\n", 0.0, cos( 0.0 ) );
// calculates and outputs tan( x )
printf( "tan(%.1f) = %.1f\n", 0.0, tan( 0.0 ) );
} // end main
sqrt(900.0) = 30.0
sqrt(9.0) = 3.0
exp(1.0) = 2.718282
exp(2.0) = 7.389056
log(2.718282) = 1.0
log(7.389056) = 2.0
log10(1.0) = 0.0
log10(10.0) = 1.0
log10(100.0) = 2.0
fabs(13.5) = 13.5
fabs(0.0) = 0.0
fabs(-13.5) = 13.5
ceil(9.2) = 10.0
ceil(-9.8) = -9.0
floor(9.2) = 9.0
floor(-9.8) = -10.0
pow(2.0, 7.0) = 128.0
pow(9.0, 0.5) = 3.0
fmod(13.657/2.333) = 2.010
sin(0.0) = 0.0
cos(0.0) = 1.0
tan(0.0) = 0.0
5.4
5.5
a)
b)
c)
d)
double hypotenuse( double side1, double side2 )
a)
b)
double hypotenuse( double side1, double side2 );
int smallest( int x, int y, int z )
void instructions( void )
float intToFloat( int number )
int smallest( int x, int y, int z );
205
206
Chapter 5 C Functions
c)
d)
void instructions( void );
float intToFloat( int number );
5.6
static float lastVal;
5.7
a) Error: Function h is defined in function g.
Correction: Move the definition of h out of the definition of g.
b) Error: The body of the function is supposed to return an integer, but does not.
Correction: Delete variable result and place the following statement in the function:
return x + y;
c) Error: Semicolon after the right parenthesis that encloses the parameter list, and redefining the parameter a in the function definition.
Correction: Delete the semicolon after the right parenthesis of the parameter list, and
delete the declaration float a; in the function body.
d) Error: The result of n + sum( n - 1 ) is not returned; sum returns an improper result.
Correction: Rewrite the statement in the else clause as
return n + sum( n - 1 );
e) Error: The function returns a value when it’s not supposed to.
Correction: Eliminate the return statement.
Exercises
5.8
Show the value of x after each of the following statements is performed:
a) x = fabs( 7.5 );
b) x = floor( 7.5 );
c) x = fabs( 0.0 );
d) x = ceil( 0.0 );
e) x = fabs( -6.4 );
f) x = ceil( -6.4 );
g) x = ceil( -fabs( -8 + floor( -5.5 ) ) );
5.9
(Parking Charges) A parking garage charges a $2.00 minimum fee to park for up to three
hours and an additional $0.50 per hour for each hour or part thereof over three hours. The maximum
charge for any given 24-hour period is $10.00. Assume that no car parks for longer than 24 hours
at a time. Write a program that will calculate and print the parking charges for each of three customers who parked their cars in this garage yesterday. You should enter the hours parked for each
customer. Your program should print the results in a tabular format, and should calculate and print
the total of yesterday's receipts. The program should use the function calculateCharges to determine the charge for each customer. Your outputs should appear in the following format:
Car
1
2
3
TOTAL
Hours
1.5
4.0
24.0
29.5
Charge
2.00
2.50
10.00
14.50
5.10 (Rounding Numbers) An application of function floor is rounding a value to the nearest
integer. The statement
y = floor( x + .5 );
will round the number x to the nearest integer and assign the result to y. Write a program that reads
several numbers and uses the preceding statement to round each of these numbers to the nearest
integer. For each number processed, print both the original number and the rounded number.
Exercises
207
5.11 (Rounding Numbers) Function floor may be used to round a number to a specific decimal
place. The statement
y = floor( x * 10 + .5 ) / 10;
rounds x to the tenths position (the first position to the right of the decimal point). The statement
y = floor( x * 100 + .5 ) / 100;
rounds x to the hundredths position (the second position to the right of the decimal point). Write
a program that defines four functions to round a number x in various ways
a) roundToInteger( number )
b) roundToTenths( number )
c) roundToHundreths( number )
d) roundToThousandths( number )
For each value read, your program should print the original value, the number rounded to the
nearest integer, the number rounded to the nearest tenth, the number rounded to the nearest hundredth, and the number rounded to the nearest thousandth.
5.12
Answer each of the following questions.
a) What does it mean to choose numbers “at random”?
b) Why is the rand function useful for simulating games of chance?
c) Why would you randomize a program by using srand? Under what circumstances is it
desirable not to randomize?
d) Why is it often necessary to scale and/or shift the values produced by rand?
5.13
Write statements that assign random integers to the variable n in the following ranges:
a) 1 ≤ n ≤ 2
b) 1 ≤ n ≤ 100
c) 0 ≤ n ≤ 9
d) 1000 ≤ n ≤ 1112
e) –1 ≤ n ≤ 1
f) –3 ≤ n ≤ 11
5.14 For each of the following sets of integers, write a single statement that will print a number
at random from the set.
a) 2, 4, 6, 8, 10.
b) 3, 5, 7, 9, 11.
c) 6, 10, 14, 18, 22.
5.15 (Hypotenuse Calculations) Define a function called hypotenuse that calculates the length
of the hypotenuse of a right triangle when the other two sides are given. The function should take
two arguments of type double and return the hypotenuse as a double. Test your program with the
side values specified in Fig. 5.22.
Triangle
Side 1
Side 2
1
2
3
3.0
5.0
8.0
4.0
12.0
15.0
Fig. 5.22 | Sample triangle side values for Exercise 5.15.
208
Chapter 5 C Functions
(Exponentiation) Write a function integerPower(base,
5.16
exponent)
that returns the value of
baseexponent
For example, integerPower( 3, 4 ) = 3 * 3 * 3 * 3. Assume that exponent is a positive, nonzero
integer, and base is an integer. Function integerPower should use for to control the calculation.
Do not use any math library functions.
5.17 (Multiples) Write a function multiple that determines for a pair of integers whether the second integer is a multiple of the first. The function should take two integer arguments and return 1
(true) if the second is a multiple of the first, and 0 (false) otherwise. Use this function in a program
that inputs a series of pairs of integers.
5.18 (Even or Odd) Write a program that inputs a series of integers and passes them one at a time
to function even, which uses the remainder operator to determine whether an integer is even. The
function should take an integer argument and return 1 if the integer is even and 0 otherwise.
5.19 (Square of Asterisks) Write a function that displays a solid square of asterisks whose side is
specified in integer parameter side. For example, if side is 4, the function displays:
****
****
****
****
5.20 (Displaying a Square of Any Character) Modify the function created in Exercise 5.19 to
form the square out of whatever character is contained in character parameter fillCharacter. Thus
if side is 5 and fillCharacter is “#”, then this function should print:
#####
#####
#####
#####
#####
5.21 (Project: Drawing Shapes with Characters) Use techniques similar to those developed in
Exercises 5.19–5.20 to produce a program that graphs a wide range of shapes.
(Separating Digits) Write program segments that accomplish each of the following:
a) Calculate the integer part of the quotient when integer a is divided by integer b.
b) Calculate the integer remainder when integer a is divided by integer b.
c) Use the program pieces developed in a) and b) to write a function that inputs an integer
between 1 and 32767 and prints it as a series of digits,with two spaces between each digit.
For example, the integer 4562 should be printed as:
5.22
4
5
6
2
5.23 (Time in Seconds) Write a function that takes the time as three integer arguments (for
hours, minutes, and seconds) and returns the number of seconds since the last time the clock “struck
12.” Use this function to calculate the amount of time in seconds between two times, both of which
are within one 12-hour cycle of the clock.
5.24
(Temperature Conversions) Implement the following integer functions:
a) Function celsius returns the Celsius equivalent of a Fahrenheit temperature.
b) Function fahrenheit returns the Fahrenheit equivalent of a Celsius temperature.
c) Use these functions to write a program that prints charts showing the Fahrenheit equivalents of all Celsius temperatures from 0 to 100 degrees, and the Celsius equivalents of
Exercises
209
all Fahrenheit temperatures from 32 to 212 degrees. Print the outputs in a tabular format that minimizes the number of lines of output while remaining readable.
5.25 (Find the Minimum) Write a function that returns the smallest of three floating-point
numbers.
5.26 (Perfect Numbers) An integer number is said to be a perfect number if its factors, including
1 (but not the number itself), sum to the number. For example, 6 is a perfect number because 6 =
1 + 2 + 3. Write a function perfect that determines whether parameter number is a perfect number.
Use this function in a program that determines and prints all the perfect numbers between 1 and
1000. Print the factors of each perfect number to confirm that the number is indeed perfect. Challenge the power of your computer by testing numbers much larger than 1000.
5.27 (Prime Numbers) An integer is said to be prime if it’s divisible by only 1 and itself. For example, 2, 3, 5 and 7 are prime, but 4, 6, 8 and 9 are not.
a) Write a function that determines whether a number is prime.
b) Use this function in a program that determines and prints all the prime numbers between 1 and 10,000. How many of these 10,000 numbers do you really have to test before being sure that you have found all the primes?
c) Initially you might think that n/2 is the upper limit for which you must test to see
whether a number is prime, but you need go only as high as the square root of n. Rewrite the program, and run it both ways. Estimate the performance improvement.
5.28 (Reversing Digits) Write a function that takes an integer value and returns the number with
its digits reversed. For example, given the number 7631, the function should return 1367.
5.29 (Greatest Common Divisor) The greatest common divisor (GCD) of two integers is the largest
integer that evenly divides each of the two numbers. Write function gcd that returns the greatest
common divisor of two integers.
5.30 (Quality Points for Student’s Grades) Write a function qualityPoints that inputs a student’s average and returns 4 it’s 90–100, 3 if it’s 80–89, 2 if it’s 70–79, 1 if it’s 60–69, and 0 if the
average is lower than 60.
5.31 (Coin Tossing) Write a program that simulates coin tossing. For each toss of the coin the
program should print Heads or Tails. Let the program toss the coin 100 times, and count the number of times each side of the coin appears. Print the results. The program should call a separate function flip that takes no arguments and returns 0 for tails and 1 for heads. [Note: If the program
realistically simulates the coin tossing, then each side of the coin should appear approximately half
the time for a total of approximately 50 heads and 50 tails.]
5.32 (Guess the Number) Write a C program that plays the game of “guess the number” as follows: Your program chooses the number to be guessed by selecting an integer at random in the range
1 to 1000. The program then types:
I have a number between 1 and 1000.
Can you guess my number?
Please type your first guess.
The player then types a first guess. The program responds with one of the following:
1. Excellent! You guessed the number!
Would you like to play again (y or n)?
2. Too low. Try again.
3. Too high. Try again.
210
Chapter 5 C Functions
If the player’s guess is incorrect, your program should loop until the player finally gets the number
right. Your program should keep telling the player Too high or Too low to help the player “zero in”
on the correct answer. [Note: The searching technique employed in this problem is called binary
search. We’ll say more about this in the next problem.]
5.33 (Guess the Number Modification) Modify the program of Exercise 5.32 to count the number of guesses the player makes. If the number is 10 or fewer, print Either you know the secret or
you got lucky! If the player guesses the number in 10 tries, then print Ahah! You know the secret!
If the player makes more than 10 guesses, then print You should be able to do better! Why should
it take no more than 10 guesses? Well, with each “good guess” the player should be able to eliminate
half of the numbers. Now show why any number 1 to 1000 can be guessed in 10 or fewer tries.
5.34 (Recursive Exponentiation) Write a recursive function power( base,
invoked returns
exponent )
that when
baseexponent
For example, power( 3, 4 ) = 3 * 3 * 3 * 3. Assume that exponent is an integer greater than or equal
to 1. Hint: The recursion step would use the relationship
baseexponent = base * baseexponent–1
and the terminating condition occurs when exponent is equal to 1 because
base1 = base
5.35
(Fibonacci) The Fibonacci series
0, 1, 1, 2, 3, 5, 8, 13, 21, …
begins with the terms 0 and 1 and has the property that each succeeding term is the sum of the two
preceding terms. a) Write a nonrecursive function fibonacci(n) that calculates the nth Fibonacci
number. Use unsigned int for the function’s parameter and unsigned long long int for its return
type. b) Determine the largest Fibonacci number that can be printed on your system.
5.36 (Towers of Hanoi) Every budding computer scientist must grapple with certain classic
problems, and the Towers of Hanoi (see Fig. 5.23) is one of the most famous of these. Legend has
it that in a temple in the Far East, priests are attempting to move a stack of disks from one peg to
another. The initial stack had 64 disks threaded onto one peg and arranged from bottom to top by
decreasing size. The priests are attempting to move the stack from this peg to a second peg under
the constraints that exactly one disk is moved at a time, and at no time may a larger disk be placed
Fig. 5.23 | Towers of Hanoi for the case with four disks.
Exercises
211
above a smaller disk. A third peg is available for temporarily holding the disks. Supposedly the world
will end when the priests complete their task, so there’s little incentive for us to facilitate their efforts.
Let’s assume that the priests are attempting to move the disks from peg 1 to peg 3. We wish to
develop an algorithm that will print the precise sequence of disk-to-disk peg transfers.
If we were to approach this problem with conventional methods, we’d rapidly find ourselves
hopelessly knotted up in managing the disks. Instead, if we attack the problem with recursion in
mind, it immediately becomes tractable. Moving n disks can be viewed in terms of moving only
n – 1 disks (and hence the recursion) as follows:
a) Move n – 1 disks from peg 1 to peg 2, using peg 3 as a temporary holding area.
b) Move the last disk (the largest) from peg 1 to peg 3.
c) Move the n – 1 disks from peg 2 to peg 3, using peg 1 as a temporary holding area.
The process ends when the last task involves moving n = 1 disk, i.e., the base case. This is
accomplished by trivially moving the disk without the need for a temporary holding area.
Write a program to solve the Towers of Hanoi problem. Use a recursive function with four
parameters:
a) The number of disks to be moved
b) The peg on which these disks are initially threaded
c) The peg to which this stack of disks is to be moved
d) The peg to be used as a temporary holding area
Your program should print the precise instructions it will take to move the disks from the
starting peg to the destination peg. For example, to move a stack of three disks from peg 1 to peg 3,
your program should print the following series of moves:
1 → 3 (This means move one disk from peg 1 to peg 3.)
1→ 2
3→ 2
1→ 3
2→ 1
2→ 3
1→ 3
5.37 (Towers of Hanoi: Iterative Solution) Any program that can be implemented recursively
can be implemented iteratively, although sometimes with considerably more difficulty and considerably less clarity. Try writing an iterative version of the Towers of Hanoi. If you succeed, compare
your iterative version with the recursive version you developed in Exercise 5.36. Investigate issues
of performance, clarity, and your ability to demonstrate the correctness of the programs.
5.38 (Visualizing Recursion) It’s interesting to watch recursion “in action.” Modify the factorial
function of Fig. 5.18 to print its local variable and recursive call parameter. For each recursive call,
display the outputs on a separate line and add a level of indentation. Do your utmost to make the
outputs clear, interesting and meaningful. Your goal here is to design and implement an output format that helps a person understand recursion better. You may want to add such display capabilities
to the many other recursion examples and exercises throughout the text.
5.39 (Recursive Greatest Common Divisor) The greatest common divisor of integers x and y is
the largest integer that evenly divides both x and y. Write a recursive function gcd that returns the
greatest common divisor of x and y. The gcd of x and y is defined recursively as follows: If y is equal
to 0, then gcd(x, y) is x; otherwise gcd(x, y) is gcd(y, x % y), where % is the remainder operator.
5.40
(Recursive
main)
Can
main
be called recursively? Write a program containing a function
main. Include static local variable count initialized to 1. Postincrement and print the value of count
each time main is called. Run your program. What happens?
212
Chapter 5 C Functions
5.41 (Distance Between Points) Write a function distance that calculates the distance between
two points (x1, y1) and (x2, y2). All numbers and return values should be of type double.
5.42
1
2
3
4
5
6
7
8
9
10
11
12
#include <stdio.h>
// function main begins program execution
int main( void )
{
int c; // variable to hold character input by user
if ( ( c = getchar() ) != EOF ) {
main();
printf( "%c", c );
} // end if
} // end main
5.43
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
What does the following program do?
What does the following program do?
#include <stdio.h>
unsigned int mystery( unsigned int a, unsigned int b ); // function prototype
// function main begins program execution
int main( void )
{
unsigned int x; // first integer
unsigned int y; // second integer
printf( "%s", "Enter two positive integers: " );
scanf( "%u%u", &x, &y );
printf( "The result is %u\n", mystery( x, y ) );
} // end main
// Parameter b must be a positive integer
// to prevent infinite recursion
unsigned int mystery( unsigned int a, unsigned int b )
{
// base case
if ( 1 == b ) {
return a;
} // end if
else { // recursive step
return a + mystery( a, b - 1 );
} // end else
} // end function mystery
5.44 After you determine what the program of Exercise 5.43 does, modify the program to function properly after removing the restriction of the second argument’s being nonnegative.
5.45 (Testing Math Library Functions) Write a program that tests the math library functions in
Fig. 5.2. Exercise each of these functions by having your program print out tables of return values
for a diversity of argument values.
Making a Difference
5.46
213
Find the error in each of the following program segments and explain how to correct it:
a) double cube( float ); // function prototype
cube( float number ) // function definition
{
return number * number * number;
}
b)
c)
int randomNumber = srand();
double y = 123.45678;
int x;
x = y;
printf( "%f\n", (double) x );
d)
double square( double number )
{
double number;
return number * number;
}
e)
int sum( int n )
{
if ( 0 == n ) {
return 0;
}
else {
return n + sum( n );
}
}
5.47 (Craps Game Modification) Modify the craps program of Fig. 5.14 to allow wagering. Package as a function the portion of the program that runs one game of craps. Initialize variable bankBalance to 1000 dollars. Prompt the player to enter a wager. Use a while loop to check that wager
is less than or equal to bankBalance, and if not, prompt the user to reenter wager until a valid wager
is entered. After a correct wager is entered, run one game of craps. If the player wins, increase
bankBalance by wager and print the new bankBalance. If the player loses, decrease bankBalance by
wager, print the new bankBalance, check whether bankBalance has become zero, and if so print the
message, "Sorry. You busted!" As the game progresses, print various messages to create some “chatter” such as, "Oh, you're going for broke, huh?" or "Aw cmon, take a chance!" or "You're up big.
Now's the time to cash in your chips!"
5.48 (Research Project: Improving the Recursive Fibonacci Implementation) In Section 5.15, the
recursive algorithm we used to calculate Fibonacci numbers was intuitively appealing. However, recall that the algorithm resulted in the exponential explosion of recursive function calls. Research the
recursive Fibonacci implementation online. Study the various approaches, including the iterative
version in Exercise 5.35 and versions that use only so-called “tail recursion.” Discuss the relative
merits of each.
Making a Difference
5.49 (Global Warming Facts Quiz) The controversial issue of global warming has been widely
publicized by the film An Inconvenient Truth, featuring former Vice President Al Gore. Mr. Gore
and a U.N. network of scientists, the Intergovernmental Panel on Climate Change, shared the 2007
Nobel Peace Prize in recognition of “their efforts to build up and disseminate greater knowledge
about man-made climate change.” Research both sides of the global warming issue online (you
might want to search for phrases like “global warming skeptics”). Create a five-question multiple-
214
Chapter 5 C Functions
choice quiz on global warming, each question having four possible answers (numbered 1–4). Be objective and try to fairly represent both sides of the issue. Next, write an application that administers
the quiz, calculates the number of correct answers (zero through five) and returns a message to the
user. If the user correctly answers five questions, print “Excellent”; if four, print “Very good”; if
three or fewer, print “Time to brush up on your knowledge of global warming,” and include a list
of some of the websites where you found your facts.
Computer-Assisted Instruction
As computer costs decline, it becomes feasible for every student, regardless of economic circumstance, to have a computer and use it in school. This creates exciting possibilities for improving the
educational experience of all students worldwide as suggested by the next five exercises. [Note:
Check out initiatives such as the One Laptop Per Child Project (www.laptop.org). Also, research
“green” laptops—what are some key “going green” characteristics of these devices? Look into the
Electronic Product Environmental Assessment Tool (www.epeat.net) which can help you assess the
“greenness” of desktops, notebooks and monitors to help you decide which products to purchase.]
5.50 (Computer-Assisted Instruction) The use of computers in education is referred to as computer-assisted instruction (CAI). Write a program that will help an elementary school student learn
multiplication. Use the rand function to produce two positive one-digit integers. The program
should then prompt the user with a question, such as
How much is 6 times 7?
The student then inputs the answer. Next, the program checks the student’s answer. If it’s correct,
display the message "Very good!" and ask another multiplication question. If the answer is wrong,
display the message "No. Please try again." and let the student try the same question repeatedly
until the student finally gets it right. A separate function should be used to generate each new question. This function should be called once when the application begins execution and each time the
user answers the question correctly.
5.51 (Computer-Assisted Instruction: Reducing Student Fatigue) One problem in CAI environments is student fatigue. This can be reduced by varying the computer’s responses to hold the student’s attention. Modify the program of Exercise 5.50 so that various comments are displayed for
each answer as follows:
Possible responses to a correct answer:
Very good!
Excellent!
Nice work!
Keep up the good work!
Possible responses to an incorrect answer:
No. Please try again.
Wrong. Try once more.
Don't give up!
No. Keep trying.
Use random-number generation to choose a number from 1 to 4 that will be used to select
one of the four appropriate responses to each correct or incorrect answer. Use a switch statement to
issue the responses.
5.52 (Computer-Assisted Instruction: Monitoring Student Performance) More sophisticated
computer-assisted instruction systems monitor the student’s performance over a period of time. The
decision to begin a new topic is often based on the student’s success with previous topics. Modify
the program of Exercise 5.51 to count the number of correct and incorrect responses typed by the
student. After the student types 10 answers, your program should calculate the percentage that are
Making a Difference
215
correct. If the percentage is lower than 75%, display "Please ask your teacher for extra help.",
then reset the program so another student can try it. If the percentage is 75% or higher, display
"Congratulations, you are ready to go to the next level!", then reset the program so another
student can try it.
5.53 (Computer-Assisted Instruction: Difficulty Levels) Exercises 5.50– through Exercise 5.52
developed a computer-assisted instruction program to help teach an elementary-school student multiplication. Modify the program to allow the user to enter a difficulty level. At a difficulty level of
1, the program should use only single-digit numbers in the problems; at a difficulty level of 2, numbers as large as two digits, and so on.
5.54 (Computer-Assisted Instruction: Varying the Types of Problems) Modify the program of
Exercise 5.53 to allow the user to pick a type of arithmetic problem to study. An option of 1 means
addition problems only, 2 means subtraction problems only, 3 means multiplication problems only
and 4 means a random mixture of all these types.
6
Now go, write it
before them in a table,
and note it in a book.
—Isaiah 30:8
To go beyond is as
wrong as to fall short.
—Confucius
Begin at the beginning, … and
go on till you come to the end:
then stop.
—Lewis Carroll
Objectives
In this chapter, you’ll learn:
■
To use the array data
structure to represent lists
and tables of values.
■
To define an array, initialize
an array and refer to
individual elements of an
array.
■
To define symbolic
constants.
■
To pass arrays to functions.
■
To use arrays to store, sort
and search lists and tables of
values.
■
To define and manipulate
multidimensional arrays.
C Arrays
6.1 Introduction
6.1
6.2
6.3
6.4
6.5
6.6
Introduction
Arrays
Defining Arrays
Array Examples
Passing Arrays to Functions
Sorting Arrays
217
6.7 Case Study: Computing Mean,
Median and Mode Using Arrays
6.8 Searching Arrays
6.9 Multidimensional Arrays
6.10 Variable-Length Arrays
6.11 Secure C Programming
Summary | Terminology | Self-Review Exercises | Answers to Self-Review Exercises | Exercises
Recursion Exercises
6.1 Introduction
This chapter serves as an introduction to data structures. Arrays are data structures consisting of related data items of the same type. In Chapter 10, we discuss C’s notion of
struct (structure)—a data structure consisting of related data items of possibly different
types. Arrays and structures are “static” entities in that they remain the same size throughout program execution (they may, of course, be of automatic storage class and hence created and destroyed each time the blocks in which they’re defined are entered and exited).
6.2 Arrays
An array is a group of contiguous memory locations that all have the same type. To refer to
a particular location or element in the array, we specify the array’s name and the position
number of the particular element in the array.
Figure 6.1 shows an integer array called c, containing 12 elements. Any one of these
elements may be referred to by giving the array’s name followed by the position number of
the particular element in square brackets ([]). The first element in every array is the zeroth
element. An array name, like other variable names, can contain only letters, digits and
underscores and cannot begin with a digit.
All elements of this array
share the array name, c
Position number of the
element within array c
Fig. 6.1 | 12-element array.
c[ 0 ]
-45
c[ 1 ]
6
c[ 2 ]
0
c[ 3 ]
72
c[ 4 ]
1543
c[ 5 ]
-89
c[ 6 ]
0
c[ 7 ]
62
c[ 8 ]
-3
c[ 9 ]
1
c[ 10 ]
6453
c[ 11 ]
78
218
Chapter 6 C Arrays
The position number within square brackets is called a subscript. A subscript must be
an integer or an integer expression. For example, if a = 5 and b = 6, then the statement
c[ a + b ] += 2;
adds 2 to array element c[11]. A subscripted array name is an lvalue—it can be used on
the left side of an assignment.
Let’s examine array c (Fig. 6.1) more closely. The array’s name is c. Its 12 elements
are referred to as c[0], c[1], c[2], …, c[10] and c[11]. The value stored in c[0] is –45,
the value of c[1] is 6, c[2] is 0, c[7] is 62 and c[11] is 78. To print the sum of the values
contained in the first three elements of array c, we’d write
printf( "%d", c[ 0 ] + c[ 1 ] + c[ 2 ] );
To divide the value of element 6 of array c by 2 and assign the result to the variable x, write
x = c[ 6 ] / 2;
The brackets used to enclose the subscript of an array are actually considered to be an
operator in C. They have the same level of precedence as the function call operator (i.e., the
parentheses that are placed after a function name to call that function). Figure 6.2 shows
the precedence and associativity of the operators introduced to this point in the text.
Operators
[]
()
++
+
-
!
*
/
%
+
-
<
<=
==
!=
(postfix) -- (postfix)
++ (prefix) -- (prefix)
>
>=
-=
*=
&&
||
?:
=
+=
/=
%=
,
(type)
Associativity
Type
left to right
right to left
left to right
left to right
left to right
left to right
left to right
left to right
right to left
right to left
left to right
highest
unary
multiplicative
additive
relational
equality
logical AND
logical OR
conditional
assignment
comma
Fig. 6.2 | Operator precedence and associativity.
6.3 Defining Arrays
Arrays occupy space in memory. You specify the type of each element and the number of
elements each array requires so that the computer may reserve the appropriate amount of
memory. The following definition reserves 12 elements for integer array c, which has subscripts in the range 0–11.
int c[ 12 ];
6.4 Array Examples
219
The definition
int b[ 100 ], x[ 27 ];
reserves 100 elements for integer array b and 27 elements for integer array x. These arrays
have subscripts in the ranges 0–99 and 0–26, respectively.
Arrays may contain other data types. For example, an array of type char can store a
character string. Character strings and their similarity to arrays are discussed in Chapter 8.
The relationship between pointers and arrays is discussed in Chapter 7.
6.4 Array Examples
This section presents several examples that demonstrate how to define and initialize arrays,
and how to perform many common array manipulations.
Defining an Array and Using a Loop to Initialize the Array’s Elements
Like any other variables, uninitialized array elements contain garbage values. Figure 6.3
uses for statements to initialize the elements of a 10-element integer array n to zeros and
print the array in tabular format. The first printf statement (line 16) displays the column
heads for the two columns printed in the subsequent for statement.
Notice that the variable i is declared to be of type size_t (line 9), which according
to the C standard represents an unsigned integral type. This type is recommended for any
variable that represents an array’s size or an array’s subscripts. Type size_t is defined in
header <stddef.h>, which is often included by other headers (such as <stdio.h>). [Note:
If you attempt to compile Fig. 6.3 and receive errors, simply include <stddef.h> in your
program.]
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
// Fig. 6.3: fig06_03.c
// Initializing the elements of an array to zeros.
#include <stdio.h>
// function main begins program execution
int main( void )
{
int n[ 10 ]; // n is an array of 10 integers
size_t i; // counter
// initialize elements of array n to 0
for ( i = 0; i < 10; ++i ) {
n[ i ] = 0; // set element at location i to 0
} // end for
printf( "%s%13s\n", "Element", "Value" );
// output contents of array n in tabular format
for ( i = 0; i < 10; ++i ) {
printf( "%7u%13d\n", i, n[ i ] );
} // end for
} // end main
Fig. 6.3 | Initializing the elements of an array to zeros. (Part 1 of 2.)
220
Chapter 6 C Arrays
Element
0
1
2
3
4
5
6
7
8
9
Value
0
0
0
0
0
0
0
0
0
0
Fig. 6.3 | Initializing the elements of an array to zeros. (Part 2 of 2.)
Initializing an Array in a Definition with an Initializer List
The elements of an array can also be initialized when the array is defined by following the
definition with an equals sign and braces, {}, containing a comma-separated list of array
initializers. Figure 6.4 initializes an integer array with 10 values (line 9) and prints the array in tabular format.
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
// Fig. 6.4: fig06_04.c
// Initializing the elements of an array with an initializer list.
#include <stdio.h>
// function main begins program execution
int main( void )
{
// use initializer list to initialize array n
int n[ 10 ] = { 32, 27, 64, 18, 95, 14, 90, 70, 60, 37 };
size_t i; // counter
printf( "%s%13s\n", "Element", "Value" );
// output contents of array in tabular format
for ( i = 0; i < 10; ++i ) {
printf( "%7u%13d\n", i, n[ i ] );
} // end for
} // end main
Element
0
1
2
3
4
5
6
7
8
9
Value
32
27
64
18
95
14
90
70
60
37
Fig. 6.4 | Initializing the elements of an array with an initializer list.
6.4 Array Examples
221
If there are fewer initializers than elements in the array, the remaining elements are
initialized to zero. For example, the elements of the array n in Fig. 6.3 could have been
initialized to zero as follows:
int n[ 10 ] = { 0 }; // initializes entire array to zeros
This explicitly initializes the first element to zero and initializes the remaining nine elements to zero because there are fewer initializers than there are elements in the array. It’s
important to remember that arrays are not automatically initialized to zero. You must at
least initialize the first element to zero for the remaining elements to be automatically zeroed. Array elements are initialized before program startup for static arrays and at runtime for automatic arrays.
Common Programming Error 6.1
Forgetting to initialize the elements of an array.
The array definition
int n[ 5 ] = { 32, 27, 64, 18, 95, 14 };
causes a syntax error because there are six initializers and only five array elements.
Common Programming Error 6.2
Providing more initializers in an array initializer list than there are elements in the array
is a syntax error.
If the array size is omitted from a definition with an initializer list, the number of elements in the array will be the number of elements in the initializer list. For example,
int n[] = { 1, 2, 3, 4, 5 };
would create a five-element array initialized with the indicated values.
Specifying an Array’s Size with a Symbolic Constant and Initializing Array Elements
with Calculations
Figure 6.5 initializes the elements of a 10-element array s to the values 2, 4, 6, …, 20 and
prints the array in tabular format. The values are generated by multiplying the loop counter by 2 and adding 2.
1
2
3
4
5
6
7
8
9
10
11
// Fig. 6.5: fig06_05.c
// Initializing the elements of array s to the even integers from 2 to 20.
#include <stdio.h>
#define SIZE 10 // maximum size of array
// function main begins program execution
int main( void )
{
// symbolic constant SIZE can be used to specify array size
int s[ SIZE ]; // array s has SIZE elements
size_t j; // counter
Fig. 6.5 | Initialize the elements of array s to the even integers from 2 to 20. (Part 1 of 2.)
222
12
13
14
15
16
17
18
19
20
21
22
23
Chapter 6 C Arrays
for ( j = 0; j < SIZE; ++j ) { // set the values
s[ j ] = 2 + 2 * j;
} // end for
printf( "%s%13s\n", "Element", "Value" );
// output contents of array s in tabular format
for ( j = 0; j < SIZE; ++j ) {
printf( "%7u%13d\n", j, s[ j ] );
} // end for
} // end main
Element
0
1
2
3
4
5
6
7
8
9
Value
2
4
6
8
10
12
14
16
18
20
Fig. 6.5 | Initialize the elements of array s to the even integers from 2 to 20. (Part 2 of 2.)
The #define preprocessor directive is introduced in this program. Line 4
#define SIZE 10
defines a symbolic constant SIZE whose value is 10. A symbolic constant is an identifier
that’s replaced with replacement text by the C preprocessor before the program is compiled. When the program is preprocessed, all occurrences of the symbolic constant SIZE
are replaced with the replacement text 10. Using symbolic constants to specify array sizes
makes programs more scalable. In Fig. 6.5, we could have the first for loop (line 13) fill
a 1000-element array by simply changing the value of SIZE in the #define directive from
10 to 1000. If the symbolic constant SIZE had not been used, we’d have to change the program in three separate places. As programs get larger, this technique becomes more useful
for writing clear, maintainable programs.
Common Programming Error 6.3
Ending a #define or #include preprocessor directive with a semicolon. Remember that
preprocessor directives are not C statements.
If the #define preprocessor directive in line 4 is terminated with a semicolon, the preprocessor replaces all occurrences of the symbolic constant SIZE in the program with the
text 10;. This may lead to syntax errors at compile time, or logic errors at execution time.
Remember that the preprocessor is not the C compiler.
Software Engineering Observation 6.1
Defining the size of each array as a symbolic constant makes programs more scalable.
6.4 Array Examples
223
Common Programming Error 6.4
Assigning a value to a symbolic constant in an executable statement is a syntax error. A
symbolic constant is not a variable. The compiler does not reserve space for symbolic constants as it does for variables that hold values at execution time.
Good Programming Practice 6.1
Use only uppercase letters for symbolic constant names. This makes these constants stand
out in a program and reminds you that symbolic constants are not variables.
Good Programming Practice 6.2
In multiword symbolic constant names, separate the words with underscores for readability.
Summing the Elements of an Array
Figure 6.6 sums the values contained in the 12-element integer array a. The
ment’s body (line 16) does the totaling.
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
for
state-
// Fig. 6.6: fig06_06.c
// Computing the sum of the elements of an array.
#include <stdio.h>
#define SIZE 12
// function main begins program execution
int main( void )
{
// use an initializer list to initialize the array
int a[ SIZE ] = { 1, 3, 5, 4, 7, 2, 99, 16, 45, 67, 89, 45 };
size_t i; // counter
int total = 0; // sum of array
// sum contents of array a
for ( i = 0; i < SIZE; ++i ) {
total += a[ i ];
} // end for
printf( "Total of array element values is %d\n", total );
} // end main
Total of array element values is 383
Fig. 6.6 | Computing the sum of the elements of an array.
Using Arrays to Summarize Survey Results
Our next example uses arrays to summarize the results of data collected in a survey. Consider the problem statement.
Forty students were asked to rate the quality of the food in the student cafeteria on a
scale of 1 to 10 (1 means awful and 10 means excellent). Place the 40 responses in an
integer array and summarize the results of the poll.
224
Chapter 6 C Arrays
This is a typical array application (see Fig. 6.7). We wish to summarize the number
of responses of each type (i.e., 1 through 10). The array responses (line 17) is a 40-element array of the students’ responses. We use an 11-element array frequency (line 14) to
count the number of occurrences of each response. We ignore frequency[0] because it’s
logical to have response 1 increment frequency[1] rather than frequency[0]. This allows
us to use each response directly as the subscript in the frequency array.
Good Programming Practice 6.3
Strive for program clarity. Sometimes it may be worthwhile to trade off the most efficient
use of memory or processor time in favor of writing clearer programs.
Performance Tip 6.1
Sometimes performance considerations far outweigh clarity considerations.
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
// Fig. 6.7: fig06_07.c
// Analyzing a student poll.
#include <stdio.h>
#define RESPONSES_SIZE 40 // define array sizes
#define FREQUENCY_SIZE 11
// function main begins program execution
int main( void )
{
size_t answer; // counter to loop through 40 responses
size_t rating; // counter to loop through frequencies 1-10
// initialize frequency counters to 0
int frequency[ FREQUENCY_SIZE ] = { 0 };
// place the survey responses in the responses array
int responses[ RESPONSES_SIZE ] = { 1, 2, 6, 4, 8, 5, 9, 7, 8, 10,
1, 6, 3, 8, 6, 10, 3, 8, 2, 7, 6, 5, 7, 6, 8, 6, 7, 5, 6, 6,
5, 6, 7, 5, 6, 4, 8, 6, 8, 10 };
// for each answer, select value of an element of array responses
// and use that value as subscript in array frequency to
// determine element to increment
for ( answer = 0; answer < RESPONSES_SIZE; ++answer ) {
++frequency[ responses [ answer ] ];
} // end for
// display results
printf( "%s%17s\n", "Rating", "Frequency" );
// output the frequencies in a tabular format
for ( rating = 1; rating < FREQUENCY_SIZE; ++rating ) {
printf( "%6d%17d\n", rating, frequency[ rating ] );
} // end for
} // end main
Fig. 6.7 | Analyzing a student poll. (Part 1 of 2.)
6.4 Array Examples
Rating
1
2
3
4
5
6
7
8
9
10
225
Frequency
2
2
2
2
5
11
5
7
1
3
Fig. 6.7 | Analyzing a student poll. (Part 2 of 2.)
The for loop (line 24) takes the responses one at a time from the array responses and
increments one of the 10 counters (frequency[1] to frequency[10]) in the frequency
array. The key statement in the loop is line 25
++frequency[ responses[ answer ] ];
which increments the appropriate frequency counter depending on the value of responses[answer]. When the counter variable answer is 0, responses[answer] is 1, so
++frequency[ responses[answer]]; is interpreted as
++frequency[ 1 ];
which increments array element one. When answer is 1,
++frequency[responses[answer]]; is interpreted as
responses[answer]
is 2, so
responses[answer]
is 6, so
++frequency[ 2 ];
which increments array element two. When answer is 2,
is interpreted as
++frequency[responses[answer]];
++frequency[ 6 ];
which increments array element six, and so on. Regardless of the number of responses processed in the survey, only an 11-element array is required (ignoring element zero) to summarize the results. If the data contained invalid values such as 13, the program would
attempt to add 1 to frequency[13]. This would be outside the bounds of the array. C has
no array bounds checking to prevent the program from referring to an element that does not exist. Thus, an executing program can “walk off” either end of an array without warning—
a security problem that we discuss in Section 6.11. You should ensure that all array references remain within the bounds of the array.
Common Programming Error 6.5
Referring to an element outside the array bounds.
Error-Prevention Tip 6.1
When looping through an array, the array subscript should never go below 0 and should
always be less than the total number of elements in the array (size – 1). Make sure the
loop-terminating condition prevents accessing elements outside this range.
226
Chapter 6 C Arrays
Error-Prevention Tip 6.2
Programs should validate the correctness of all input values to prevent erroneous information from affecting a program’s calculations.
Graphing Array Element Values with Histograms
Our next example (Fig. 6.8) reads numbers from an array and graphs the information in
the form of a bar chart or histogram—each number is printed, then a bar consisting of that
many asterisks is printed beside the number. The nested for statement (line 20) draws the
bars. Note the use of puts( "" ) to end each histogram bar (line 24).
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
// Fig. 6.8: fig06_08.c
// Displaying a histogram.
#include <stdio.h>
#define SIZE 10
// function main begins program execution
int main( void )
{
// use initializer list to initialize array n
int n[ SIZE ] = { 19, 3, 15, 7, 11, 9, 13, 5, 17, 1 };
size_t i; // outer for counter for array elements
int j; // inner for counter counts *s in each histogram bar
printf( "%s%13s%17s\n", "Element", "Value", "Histogram" );
// for each element of array n, output a bar of the histogram
for ( i = 0; i < SIZE; ++i ) {
printf( "%7u%13d
", i, n[ i ]) ;
for ( j = 1; j <= n[ i ]; ++j ) { // print one bar
printf( "%c", '*' );
} // end inner for
puts( "" ); // end a histogram bar
} // end outer for
} // end main
Element
0
1
2
3
4
5
6
7
8
9
Value
19
3
15
7
11
9
13
5
17
1
Fig. 6.8 | Displaying a histogram.
Histogram
*******************
***
***************
*******
***********
*********
*************
*****
*****************
*
6.4 Array Examples
227
Rolling a Die 6,000,000 Times and Summarizing the Results in an Array
In Chapter 5, we stated that we’d show a more elegant method of writing the dice-rolling
program of Fig. 5.12. The problem was to roll a single six-sided die 6,000,000 times to
test whether the random number generator actually produces random numbers. An array
version of this program is shown in Fig. 6.9.
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
// Fig. 6.9: fig06_09.c
// Roll a six-sided die 6,000,000 times
#include <stdio.h>
#include <stdlib.h>
#include <time.h>
#define SIZE 7
// function main begins program execution
int main( void )
{
size_t face; // random die value 1 - 6
unsigned int roll; // roll counter 1-6,000,000
unsigned int frequency[ SIZE ] = { 0 }; // clear counts
srand( time( NULL ) ); // seed random number generator
// roll die 6,000,000 times
for ( roll = 1; roll <= 6000000; ++roll ) {
face = 1 + rand() % 6;
++frequency[ face ]; // replaces entire switch of Fig. 5.8
} // end for
printf( "%s%17s\n", "Face", "Frequency" );
// output frequency elements 1-6 in tabular format
for ( face = 1; face < SIZE; ++face ) {
printf( "%4d%17d\n", face, frequency[ face ] );
} // end for
} // end main
Face
1
2
3
4
5
6
Frequency
999753
1000773
999600
999786
1000552
999536
Fig. 6.9 | Roll a six-sided die 6,000,000 times.
Using Character Arrays to Store and Manipulate Strings
We’ve discussed only integer arrays. However, arrays are capable of holding data of any
type. We now discuss storing strings in character arrays. So far, the only string-processing
capability we have is outputting a string with printf. A string such as "hello" is really an
array of individual characters in C.
228
Chapter 6 C Arrays
Character arrays have several unique features. A character array can be initialized using
a string literal. For example,
char string1[] = "first";
initializes the elements of array string1 to the individual characters in the string literal
"first". In this case, the size of array string1 is determined by the compiler based on the
length of the string. The string "first" contains five characters plus a special string-termination character called the null character. Thus, array string1 actually contains six elements. The character constant representing the null character is '\0'. All strings in C end
with this character. A character array representing a string should always be defined large
enough to hold the number of characters in the string and the terminating null character.
Character arrays also can be initialized with individual character constants in an initializer list, but this can be tedious. The preceding definition is equivalent to
char string1[] = { 'f', 'i', 'r', 's', 't', '\0' };
Because a string is really an array of characters, we can access individual characters in
a string directly using array subscript notation. For example, string1[0] is the character
'f' and string1[3] is the character 's'.
We also can input a string directly into a character array from the keyboard using
scanf and the conversion specifier %s. For example,
char string2[ 20 ];
creates a character array capable of storing a string of at most 19 characters and a terminating
null character. The statement
scanf( "%19s", string2 );
reads a string from the keyboard into string2. The name of the array is passed to scanf
without the preceding & used with nonstring variables. The & is normally used to provide
scanf with a variable’s location in memory so that a value can be stored there. In
Section 6.5, when we discuss passing arrays to functions, we’ll see that the value of an array
name is the address of the start of the array; therefore, the & is not necessary. Function scanf
will read characters until a space, tab, newline or end-of-file indicator is encountered. The
string string2 should be no longer than 19 characters to leave room for the terminating
null character. If the user types 20 or more characters, your program may crash or create
a security vulnerability. For this reason, we used the conversion specifier %19s so that
scanf reads a maximum of 19 characters and does not write characters into memory beyond the end of the array string2.
It’s your responsibility to ensure that the array into which the string is read is capable
of holding any string that the user types at the keyboard. Function scanf does not check
how large the array is. Thus, scanf can write beyond the end of the array.
A character array representing a string can be output with printf and the %s conversion specifier. The array string2 is printed with the statement
printf( "%s\n", string2 );
Function printf, like scanf, does not check how large the character array is. The characters of the string are printed until a terminating null character is encountered. [Consider
what would print if, for some reason, the terminating null character were missing.]
6.4 Array Examples
229
Figure 6.10 demonstrates initializing a character array with a string literal, reading a
string into a character array, printing a character array as a string and accessing individual
characters of a string. The program uses a for statement (line 23) to loop through the
string1 array and print the individual characters separated by spaces, using the %c conversion specifier. The condition in the for statement is true while the counter is less than the
size of the array and the terminating null character has not been encountered in the string.
In this program, we read only strings that do not contain whitespace characters. We’ll show
how to read strings with whitespace characters in Chapter 8. Notice that lines 18–19 contain
two string literals separated only by whitespace. The compiler automatically combines such
string literals into one—this is helpful for making long string literals more readable.
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
// Fig. 6.10: fig06_10.c
// Treating character arrays as strings.
#include <stdio.h>
#define SIZE 20
// function main begins program execution
int main( void )
{
char string1[ SIZE ]; // reserves 20 characters
char string2[] = "string literal"; // reserves 15 characters
size_t i; // counter
// read string from user into array string1
printf( "%s", "Enter a string (no longer than 19 characters): " );
scanf( "%19s", string1 ); // input no more than 19 characters
// output strings
printf( "string1 is: %s\nstring2 is: %s\n"
"string1 with spaces between characters is:\n",
string1, string2 );
// output characters until null character is reached
for ( i = 0; i < SIZE && string1[ i ] != '\0'; ++i ) {
printf( "%c ", string1[ i ] );
} // end for
puts( "" );
} // end main
Enter a
string1
string2
string1
H e l l
string (no longer than 19 characters): Hello there
is: Hello
is: string literal
with spaces between characters is:
o
Fig. 6.10 | Treating character arrays as strings.
Static Local Arrays and Automatic Local Arrays
Chapter 5 discussed the storage-class specifier static. A static local variable exists for
the duration of the program but is visible only in the function body. We can apply static
to a local array definition so the array is not created and initialized each time the function
230
Chapter 6 C Arrays
is called and the array is not destroyed each time the function is exited in the program. This
reduces program execution time, particularly for programs with frequently called functions that contain large arrays.
Performance Tip 6.2
In functions that contain automatic arrays where the function is in and out of scope frequently, make the array static so it’s not created each time the function is called.
Arrays that are static are initialized once at program startup. If you do not explicitly
initialize a static array, that array’s elements are initialized to zero by default.
Figure 6.11 demonstrates function staticArrayInit (lines 21–40) with a local static
array (line 24) and function automaticArrayInit (lines 43–62) with a local automatic array
(line 46). Function staticArrayInit is called twice (lines 12 and 16). The local static
array in the function is initialized to zero before program startup (line 24). The function
prints the array, adds 5 to each element and prints the array again. The second time the function is called, the static array contains the values stored during the first function call.
Function automaticArrayInit is also called twice (lines 13 and 17). The elements of
the automatic local array in the function are initialized with the values 1, 2 and 3 (line 46).
The function prints the array, adds 5 to each element and prints the array again. The
second time the function is called, the array elements are initialized to 1, 2 and 3 again
because the array has automatic storage duration.
Common Programming Error 6.6
Assuming that elements of a local static array are initialized to zero every time the function in which the array is defined is called.
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
// Fig. 6.11: fig06_11.c
// Static arrays are initialized to zero if not explicitly initialized.
#include <stdio.h>
void staticArrayInit( void ); // function prototype
void automaticArrayInit( void ); // function prototype
// function main begins program execution
int main( void )
{
puts( "First call to each function:" );
staticArrayInit();
automaticArrayInit();
puts( "\n\nSecond call to each function:" );
staticArrayInit();
automaticArrayInit();
} // end main
// function to demonstrate a static local array
void staticArrayInit( void )
{
Fig. 6.11 | Static arrays are initialized to zero if not explicitly initialized. (Part 1 of 3.)
6.4 Array Examples
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
// initializes elements to 0 first time function is called
static int array1[ 3 ];
size_t i; // counter
puts( "\nValues on entering staticArrayInit:" );
// output contents of array1
for ( i = 0; i <= 2; ++i ) {
printf( "array1[ %u ] = %d
} // end for
", i, array1[ i ] );
puts( "\nValues on exiting staticArrayInit:" );
// modify and output contents of array1
for ( i = 0; i <= 2; ++i ) {
printf( "array1[ %u ] = %d ", i, array1[ i ] += 5 );
} // end for
} // end function staticArrayInit
// function to demonstrate an automatic local array
void automaticArrayInit( void )
{
// initializes elements each time function is called
int array2[ 3 ] = { 1, 2, 3 };
size_t i; // counter
puts( "\n\nValues on entering automaticArrayInit:" );
// output contents of array2
for ( i = 0; i <= 2; ++i ) {
printf("array2[ %u ] = %d
} // end for
", i, array2[ i ] );
puts( "\nValues on exiting automaticArrayInit:" );
// modify and output contents of array2
for ( i = 0; i <= 2; ++i ) {
printf( "array2[ %u ] = %d ", i, array2[ i ] += 5 );
} // end for
} // end function automaticArrayInit
First call to each function:
Values on
array1[ 0
Values on
array1[ 0
entering staticArrayInit:
] = 0 array1[ 1 ] = 0 array1[ 2 ] = 0
exiting staticArrayInit:
] = 5 array1[ 1 ] = 5 array1[ 2 ] = 5
Values on
array2[ 0
Values on
array2[ 0
entering automaticArrayInit:
] = 1 array2[ 1 ] = 2 array2[ 2 ] = 3
exiting automaticArrayInit:
] = 6 array2[ 1 ] = 7 array2[ 2 ] = 8
Fig. 6.11 | Static arrays are initialized to zero if not explicitly initialized. (Part 2 of 3.)
231
232
Chapter 6 C Arrays
Second call to each function:
Values on
array1[ 0
Values on
array1[ 0
entering staticArrayInit:
] = 5 array1[ 1 ] = 5 array1[ 2 ] = 5
exiting staticArrayInit:
] = 10 array1[ 1 ] = 10 array1[ 2 ] = 10
Values on
array2[ 0
Values on
array2[ 0
entering automaticArrayInit:
] = 1 array2[ 1 ] = 2 array2[ 2 ] = 3
exiting automaticArrayInit:
] = 6 array2[ 1 ] = 7 array2[ 2 ] = 8
Fig. 6.11 | Static arrays are initialized to zero if not explicitly initialized. (Part 3 of 3.)
6.5 Passing Arrays to Functions
To pass an array argument to a function, specify the array’s name without any brackets.
For example, if array hourlyTemperatures has been defined as
int hourlyTemperatures[ HOURS_IN_A_DAY ];
the function call
modifyArray( hourlyTemperatures, HOURS_IN_A_DAY )
passes array hourlyTemperatures and its size to function modifyArray.
Recall that all arguments in C are passed by value. C automatically passes arrays to
functions by reference (again, we’ll see in Chapter 7 that this is not a contradiction)—the
called functions can modify the element values in the callers’ original arrays. The name of
the array evaluates to the address of the first element of the array. Because the starting
address of the array is passed, the called function knows precisely where the array is stored.
Therefore, when the called function modifies array elements in its function body, it’s modifying the actual elements of the array in their original memory locations.
Figure 6.12 demonstrates that an array name is really the address of the first element
of the array by printing array, &array[0] and &array using the %p conversion specifier—
a special conversion specifier for printing addresses. The %p conversion specifier normally
outputs addresses as hexadecimal numbers, but this is compiler dependent. Hexadecimal
(base 16) numbers consist of the digits 0 through 9 and the letters A through F (these letters are the hexadecimal equivalents of the decimal numbers 10–15). Appendix C provides
an in-depth discussion of the relationships among binary (base 2), octal (base 8), decimal
(base 10; standard integers) and hexadecimal integers. The output shows that array,
&array and &array[0] have the same value, namely 0012FF78. The output of this program
is system dependent, but the addresses are always identical for a particular execution of this
program on a particular computer.
Performance Tip 6.3
Passing arrays by reference makes sense for performance reasons. If arrays were passed by
value, a copy of each element would be passed. For large, frequently passed arrays, this
would be time consuming and would consume storage for the copies of the arrays.
6.5 Passing Arrays to Functions
1
2
3
4
5
6
7
8
9
10
11
12
233
// Fig. 6.12: fig06_12.c
// Array name is the same as the address of the array’s first element.
#include <stdio.h>
// function main begins program execution
int main( void )
{
char array[ 5 ]; // define an array of size 5
printf( "
array = %p\n&array[0] = %p\n
array, &array[ 0 ], &array );
} // end main
&array = %p\n",
array = 0012FF78
&array[0] = 0012FF78
&array = 0012FF78
Fig. 6.12 | Array name is the same as the address of the array’s first element.
Software Engineering Observation 6.2
It’s possible to pass an array by value (by using a simple trick we explain in Chapter 10).
Although entire arrays are passed by reference, individual array elements are passed by
value exactly as simple variables are. Such simple single pieces of data (such as individual
ints, floats and chars) are called scalars. To pass an element of an array to a function,
use the subscripted name of the array element as an argument in the function call. In
Chapter 7, we show how to pass scalars (i.e., individual variables and array elements) to
functions by reference.
For a function to receive an array through a function call, the function’s parameter list
must specify that an array will be received. For example, the function header for function
modifyArray (that we called earlier in this section) might be written as
void modifyArray( int b[], int size )
indicating that modifyArray expects to receive an array of integers in parameter b and the
number of array elements in parameter size. The size of the array is not required between
the array brackets. If it’s included, the compiler checks that it’s greater than zero, then ignores it. Specifying a negative size is a compilation error. Because arrays are automatically
passed by reference, when the called function uses the array name b, it will be referring to
the array in the caller (array hourlyTemperatures in the preceding call). In Chapter 7, we
introduce other notations for indicating that an array is being received by a function. As
we’ll see, these notations are based on the intimate relationship between arrays and pointers in C.
Difference Between Passing an Entire Array and Passing an Array Element
Figure 6.13 demonstrates the difference between passing an entire array and passing an array element. The program first prints the five elements of integer array a (lines 20–22).
Next, a and its size are passed to function modifyArray (line 27), where each of a’s ele-
234
Chapter 6 C Arrays
ments is multiplied by 2 (lines 53–55). Then a is reprinted in main (lines 32–34). As the
output shows, the elements of a are indeed modified by modifyArray. Now the program
prints the value of a[3] (line 38) and passes it to function modifyElement (line 40). Function modifyElement multiplies its argument by 2 (line 63) and prints the new value. When
a[3] is reprinted in main (line 43), it has not been modified, because individual array elements are passed by value.
There may be situations in your programs in which a function should not be allowed
to modify array elements. C provides the type qualifier const (for “constant”) that can be
used to prevent modification of array values in a function. When an array parameter is preceded by the const qualifier, the array elements become constant in the function body,
and any attempt to modify an element of the array in the function body results in a compile-time error. This enables you to correct a program so it does not attempt to modify
array elements.
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
// Fig. 6.13: fig06_13.c
// Passing arrays and individual array elements to functions.
#include <stdio.h>
#define SIZE 5
// function prototypes
void modifyArray( int b[], size_t size );
void modifyElement( int e );
// function main begins program execution
int main( void )
{
int a[ SIZE ] = { 0, 1, 2, 3, 4 }; // initialize array a
size_t i; // counter
puts( "Effects of passing entire array by reference:\n\nThe "
"values of the original array are:" );
// output original array
for ( i = 0; i < SIZE; ++i ) {
printf( "%3d", a[ i ] );
} // end for
puts( "" );
// pass array a to modifyArray by reference
modifyArray( a, SIZE );
puts( "The values of the modified array are:" );
// output modified array
for ( i = 0; i < SIZE; ++i ) {
printf( "%3d", a[ i ] );
} // end for
Fig. 6.13 | Passing arrays and individual array elements to functions. (Part 1 of 2.)
6.5 Passing Arrays to Functions
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
235
// output value of a[ 3 ]
printf( "\n\n\nEffects of passing array element "
"by value:\n\nThe value of a[3] is %d\n", a[ 3 ] );
modifyElement( a[ 3 ] ); // pass array element a[ 3 ] by value
// output value of a[ 3 ]
printf( "The value of a[ 3 ] is %d\n", a[ 3 ] );
} // end main
// in function modifyArray, "b" points to the original array "a"
// in memory
void modifyArray( int b[], size_t size )
{
size_t j; // counter
// multiply each array element by 2
for ( j = 0; j < size; ++j ) {
b[ j ] *= 2; // actually modifies original array
} // end for
} // end function modifyArray
// in function modifyElement, "e" is a local copy of array element
// a[ 3 ] passed from main
void modifyElement( int e )
{
// multiply parameter by 2
printf( "Value in modifyElement is %d\n", e *= 2 );
} // end function modifyElement
Effects of passing entire array by reference:
The values of
0 1 2 3
The values of
0 2 4 6
the original array are:
4
the modified array are:
8
Effects of passing array element by value:
The value of a[3] is 6
Value in modifyElement is 12
The value of a[ 3 ] is 6
Fig. 6.13 | Passing arrays and individual array elements to functions. (Part 2 of 2.)
Using the const Qualifier with Array Parameters
Figure 6.14 demonstrates the const qualifier. Function tryToModifyArray (line 19) is defined with parameter const int b[], which specifies that array b is constant and cannot be
modified. The output shows the error messages produced by the compiler—the errors may
be different for your compiler. Each of the function’s three attempts to modify array elements results in the compiler error “l-value specifies a const object.” The const qualifier is discussed in additional contexts in Chapter 7.
236
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
Chapter 6 C Arrays
// Fig. 6.14: fig06_14.c
// Using the const type qualifier with arrays.
#include <stdio.h>
void tryToModifyArray( const int b[] ); // function prototype
// function main begins program execution
int main( void )
{
int a[] = { 10, 20, 30 }; // initialize array a
tryToModifyArray( a );
printf("%d %d %d\n", a[ 0 ], a[ 1 ], a[ 2 ] );
} // end main
// in function tryToModifyArray, array b is const, so it cannot be
// used to modify the original array a in main.
void tryToModifyArray( const int b[] )
{
b[ 0 ] /= 2; // error
b[ 1 ] /= 2; // error
b[ 2 ] /= 2; // error
} // end function tryToModifyArray
fig06_14.c(21) : error C2166: l-value specifies const object
fig06_14.c(22) : error C2166: l-value specifies const object
fig06_14.c(23) : error C2166: l-value specifies const object
Fig. 6.14 | Using the const type qualifier with arrays.
Software Engineering Observation 6.3
The const type qualifier can be applied to an array parameter in a function definition to
prevent the original array from being modified in the function body. This is another
example of the principle of least privilege. A function should not be given the capability to
modify an array in the caller unless it’s absolutely necessary.
6.6 Sorting Arrays
Sorting data (i.e., placing the data into a particular order such as ascending or descending)
is one of the most important computing applications. A bank sorts all checks by account
number so that it can prepare individual bank statements at the end of each month. Telephone companies sort their lists of accounts by last name and, within that, by first name
to make it easy to find phone numbers. Virtually every organization must sort some data,
and in many cases massive amounts of it. Sorting data is an intriguing problem which has
attracted some of the most intense research efforts in the field of computer science. In this
chapter we discuss what is perhaps the simplest known sorting scheme. In Chapter 12 and
Appendix E, we investigate more complex schemes that yield better performance.
6.6 Sorting Arrays
237
Performance Tip 6.4
Often, the simplest algorithms perform poorly. Their virtue is that they’re easy to write, test
and debug. More complex algorithms are often needed to realize maximum performance.
Figure 6.15 sorts the values in the elements of the 10-element array a (line 10) into
ascending order. The technique we use is called the bubble sort or the sinking sort because
the smaller values gradually “bubble” their way upward to the top of the array like air bubbles rising in water, while the larger values sink to the bottom of the array. The technique
is to make several passes through the array. On each pass, successive pairs of elements (element 0 and element 1, then element 1 and element 2, etc.) are compared. If a pair is in
increasing order (or if the values are identical), we leave the values as they are. If a pair is
in decreasing order, their values are swapped in the array.
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
// Fig. 6.15: fig06_15.c
// Sorting an array's values into ascending order.
#include <stdio.h>
#define SIZE 10
// function main begins program execution
int main( void )
{
// initialize a
int a[ SIZE ] = { 2, 6, 4, 8, 10, 12, 89, 68, 45, 37 };
int pass; // passes counter
size_t i; // comparisons counter
int hold; // temporary location used to swap array elements
puts( "Data items in original order" );
// output original array
for ( i = 0; i < SIZE; ++i ) {
printf( "%4d", a[ i ] );
} // end for
// bubble sort
// loop to control number of passes
for ( pass = 1; pass < SIZE; ++pass ) {
// loop to control number of comparisons per pass
for ( i = 0; i < SIZE - 1; ++i ) {
// compare adjacent elements and swap them if first
// element is greater than second element
if ( a[ i ] > a[ i + 1 ] ) {
hold = a[ i ];
a[ i ] = a[ i + 1 ];
a[ i + 1 ] = hold;
} // end if
} // end inner for
} // end outer for
Fig. 6.15 | Sorting an array’s values into ascending order. (Part 1 of 2.)
238
38
39
40
41
42
43
44
45
46
47
Chapter 6 C Arrays
puts( "\nData items in ascending order" );
// output sorted array
for ( i = 0; i < SIZE; ++i ) {
printf( "%4d", a[ i ] );
} // end for
puts( "" );
} // end main
Data items in original order
2
6
4
8 10 12 89 68
Data items in ascending order
2
4
6
8 10 12 37 45
45
37
68
89
Fig. 6.15 | Sorting an array’s values into ascending order. (Part 2 of 2.)
First the program compares a[0] to a[1], then a[1] to a[2], then a[2] to a[3], and
so on until it completes the pass by comparing a[8] to a[9]. Although there are 10 elements, only nine comparisons are performed. Because of the way the successive comparisons are made, a large value may move down the array many positions on a single pass, but
a small value may move up only one position.
On the first pass, the largest value is guaranteed to sink to the bottom element of the
array, a[9]. On the second pass, the second-largest value is guaranteed to sink to a[8]. On
the ninth pass, the ninth-largest value sinks to a[1]. This leaves the smallest value in a[0],
so only nine passes of the array are needed to sort the array, even though there are ten elements.
The sorting is performed by the nested for loops (lines 24–37). If a swap is necessary,
it’s performed by the three assignments
hold = a[ i ];
a[ i ] = a[ i + 1 ];
a[ i + 1 ] = hold;
where the extra variable hold temporarily stores one of the two values being swapped. The
swap cannot be performed with only the two assignments
a[ i ] = a[ i + 1 ];
a[ i + 1 ] = a[ i ];
If, for example, a[i] is 7 and a[i + 1] is 5, after the first assignment both values will be 5
and the value 7 will be lost—hence the need for the extra variable hold.
The chief virtue of the bubble sort is that it’s easy to program. However, it runs slowly
because every exchange moves an element only one position closer to its final destination.
This becomes apparent when sorting large arrays. In the exercises, we’ll develop more efficient versions of the bubble sort. Far more efficient sorts than the bubble sort have been
developed. We’ll investigate a few of these in the exercises. More advanced courses investigate sorting and searching in greater depth.
6.7 Case Study: Computing Mean, Median and Mode Using Arrays
239
6.7 Case Study: Computing Mean, Median and Mode
Using Arrays
We now consider a larger example. Computers are commonly used for survey data analysis
to compile and analyze the results of surveys and opinion polls. Figure 6.16 uses array response initialized with 99 responses to a survey. Each response is a number from 1 to 9. The
program computes the mean, median and mode of the 99 values. Figure 6.17 contains a
sample run of this program. This example includes most of the common manipulations
usually required in array problems, including passing arrays to functions.
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
// Fig. 6.16: fig06_16.c
// Survey data analysis with arrays:
// computing the mean, median and mode of the data.
#include <stdio.h>
#define SIZE 99
// function prototypes
void mean( const unsigned int answer[] );
void median( unsigned int answer[] );
void mode( unsigned int freq[], unsigned const int answer[] ) ;
void bubbleSort( int a[] );
void printArray( unsigned const int a[] );
// function main begins program execution
int main( void )
{
unsigned int frequency[ 10 ] = { 0 }; // initialize array frequency
// initialize array response
unsigned int response[ SIZE ] =
{ 6, 7, 8, 9, 8, 7, 8, 9, 8, 9,
7, 8, 9, 5, 9, 8, 7, 8, 7, 8,
6, 7, 8, 9, 3, 9, 8, 7, 8, 7,
7, 8, 9, 8, 9, 8, 9, 7, 8, 9,
6, 7, 8, 7, 8, 7, 9, 8, 9, 2,
7, 8, 9, 8, 9, 8, 9, 7, 5, 3,
5, 6, 7, 2, 5, 3, 9, 4, 6, 4,
7, 8, 9, 6, 8, 7, 8, 9, 7, 8,
7, 4, 4, 2, 5, 3, 8, 7, 5, 6,
4, 5, 6, 1, 6, 5, 7, 8, 7 };
// process responses
mean( response );
median( response );
mode( frequency, response );
} // end main
// calculate average of all response values
void mean( const unsigned int answer[] )
{
Fig. 6.16 | Survey data analysis with arrays: computing the mean, median and mode of the data.
(Part 1 of 4.)
240
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
Chapter 6 C Arrays
size_t j; // counter for totaling array elements
unsigned int total = 0; // variable to hold sum of array elements
printf( "%s\n%s\n%s\n", "********", "
Mean", "********" );
// total response values
for ( j = 0; j < SIZE; ++j ) {
total += answer[ j ];
} // end for
printf( "The mean is the average value of the data\n"
"items. The mean is equal to the total of\n"
"all the data items divided by the number\n"
"of data items ( %u ). The mean value for\n"
"this run is: %u / %u = %.4f\n\n",
SIZE, total, SIZE, ( double ) total / SIZE );
} // end function mean
// sort array and determine median element's value
void median( unsigned int answer[] )
{
printf( "\n%s\n%s\n%s\n%s",
"********", " Median", "********",
"The unsorted array of responses is" );
printArray( answer ); // output unsorted array
bubbleSort( answer ); // sort array
printf( "%s", "\n\nThe sorted array is" );
printArray( answer ); // output sorted array
// display median element
printf( "\n\nThe median is element %u of\n"
"the sorted %u element array.\n"
"For this run the median is %u\n\n",
SIZE / 2, SIZE, answer[ SIZE / 2 ] );
} // end function median
// determine most frequent response
void mode( unsigned int freq[], const unsigned int answer[] )
{
size_t rating; // counter for accessing elements 1-9 of array freq
size_t j; // counter for summarizing elements 0-98 of array answer
unsigned int h; // counter for diplaying histograms freq array values
unsigned int largest = 0; // represents largest frequency
unsigned int modeValue = 0; // represents most frequent response
printf( "\n%s\n%s\n%s\n",
"********", " Mode", "********" );
Fig. 6.16 | Survey data analysis with arrays: computing the mean, median and mode of the data.
(Part 2 of 4.)
6.7 Case Study: Computing Mean, Median and Mode Using Arrays
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
241
// initialize frequencies to 0
for ( rating = 1; rating <= 9; ++rating ) {
freq[ rating ] = 0;
} // end for
// summarize frequencies
for ( j = 0; j < SIZE; ++j ) {
++freq[ answer[ j ] ];
} // end for
// output headers for result columns
printf( "%s%11s%19s\n\n%54s\n%54s\n\n",
"Response", "Frequency", "Histogram",
"1
1
2
2", "5
0
5
0
5" );
// output results
for ( rating = 1; rating <= 9; ++rating ) {
printf( "%8u%11u
", rating, freq[ rating ] );
// keep track of mode value and largest frequency value
if ( freq[ rating ] > largest ) {
largest = freq[ rating ];
modeValue = rating;
} // end if
// output histogram bar representing frequency value
for ( h = 1; h <= freq[ rating ]; ++h ) {
printf( "%s", "*" );
} // end inner for
puts( "" ); // being new line of output
} // end outer for
// display the mode value
printf( "\nThe mode is the most frequent value.\n"
"For this run the mode is %u which occurred"
" %u times.\n", modeValue, largest );
} // end function mode
// function that sorts an array with bubble sort algorithm
void bubbleSort( unsigned int a[] )
{
unsigned int pass; // pass counter
size_t j; // comparison counter
unsigned int hold; // temporary location used to swap elements
// loop to control number of passes
for ( pass = 1; pass < SIZE; ++pass ) {
Fig. 6.16 | Survey data analysis with arrays: computing the mean, median and mode of the data.
(Part 3 of 4.)
242
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
Chapter 6 C Arrays
// loop to control number of comparisons per pass
for ( j = 0; j < SIZE - 1; ++j ) {
// swap elements if out of order
if ( a[ j ] > a[ j + 1 ] ) {
hold = a[ j ];
a[ j ] = a[ j + 1 ];
a[ j + 1 ] = hold;
} // end if
} // end inner for
} // end outer for
} // end function bubbleSort
// output array contents (20 values per row)
void printArray( const unsigned int a[] )
{
size_t j; // counter
// output array contents
for ( j = 0; j < SIZE; ++j ) {
if ( j % 20 == 0 ) { // begin new line every 20 values
puts( "" );
} // end if
printf( "%2u", a[ j ] );
} // end for
} // end function printArray
Fig. 6.16 | Survey data analysis with arrays: computing the mean, median and mode of the data.
(Part 4 of 4.)
********
Mean
********
The mean is the average value of the data
items. The mean is equal to the total of
all the data items divided by the number
of data items ( 99 ). The mean value for
this run is: 681 / 99 = 6.8788
********
Median
********
The unsorted
6 7 8 9 8 7
6 7 8 9 3 9
6 7 8 7 8 7
5 6 7 2 5 3
7 4 4 2 5 3
array
8 9 8
8 7 8
9 8 9
9 4 6
8 7 5
of responses is
9 7 8 9 5 9 8 7
7 7 8 9 8 9 8 9
2 7 8 9 8 9 8 9
4 7 8 9 6 8 7 8
6 4 5 6 1 6 5 7
8
7
7
9
8
7
8
5
7
7
8
9
3
8
Fig. 6.17 | Sample run for the survey data analysis program. (Part 1 of 2.)
6.7 Case Study: Computing Mean, Median and Mode Using Arrays
The sorted
1 2 2 2 3
5 6 6 6 6
7 7 7 7 7
8 8 8 8 8
9 9 9 9 9
array
3 3 3
6 6 6
7 7 7
8 8 8
9 9 9
is
4 4
6 6
7 7
8 8
9 9
4
7
7
8
9
4
7
7
8
9
4
7
7
8
9
5
7
8
8
9
5
7
8
8
9
5
7
8
8
9
5
7
8
8
9
5
7
8
8
9
5
7
8
8
9
243
5
7
8
8
The median is element 49 of
the sorted 99 element array.
For this run the median is 7
********
Mode
********
Response
Frequency
Histogram
5
1
2
3
4
5
6
7
8
9
1
3
4
5
8
9
23
27
19
1
0
1
5
2
0
2
5
*
***
****
*****
********
*********
***********************
***************************
*******************
The mode is the most frequent value.
For this run the mode is 8 which occurred 27 times.
Fig. 6.17 | Sample run for the survey data analysis program. (Part 2 of 2.)
Mean
The mean is the arithmetic average of the 99 values. Function mean (Fig. 6.16, lines 39–57)
computes the mean by totaling the 99 elements and dividing the result by 99.
Median
The median is the middle value. Function median (lines 60–78) determines the median by
calling function bubbleSort (defined in lines 132–152) to sort the array of responses into
ascending order, then picking answer[SIZE / 2] (the middle element) of the sorted array.
When the number of elements is even, the median should be calculated as the mean of the
two middle elements. Function median does not currently provide this capability. Function printArray (lines 155–168) is called to output the response array.
Mode
The mode is the value that occurs most frequently among the 99 responses. Function mode
(lines 81–129) determines the mode by counting the number of responses of each type,
then selecting the value with the greatest count. This version of function mode does not
handle a tie (see Exercise 6.14). Function mode also produces a histogram to aid in determining the mode graphically.
244
Chapter 6 C Arrays
6.8 Searching Arrays
You’ll often work with large amounts of data stored in arrays. It may be necessary to determine whether an array contains a value that matches a certain key value. The process of
finding a particular element of an array is called searching. In this section we discuss two
searching techniques—the simple linear search technique and the more efficient (but
more complex) binary search technique. Exercise 6.32 and Exercise 6.33 ask you to implement recursive versions of the linear search and the binary search, respectively.
Searching an Array with Linear Search
The linear search (Fig. 6.18) compares each element of the array with the search key. Because the array is not in any particular order, it’s just as likely that the value will be found
in the first element as in the last. On average, therefore, the program will have to compare
the search key with half the elements of the array.
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
// Fig. 6.18: fig06_18.c
// Linear search of an array.
#include <stdio.h>
#define SIZE 100
// function prototype
size_t linearSearch( const int array[], int key, size_t size );
// function main begins program execution
int main( void )
{
int a[ SIZE ]; // create array a
size_t x; // counter for initializing elements 0-99 of array a
int searchKey; // value to locate in array a
size_t element; // variable to hold location of searchKey or -1
// create some data
for ( x = 0; x < SIZE; ++x ) {
a[ x ] = 2 * x;
} // end for
puts( "Enter integer search key:" );
scanf( "%d", &searchKey );
// attempt to locate searchKey in array a
element = linearSearch( a, searchKey, SIZE );
// display results
if ( element != -1 ) {
printf( "Found value in element %d\n", element );
} // end if
else {
puts( "Value not found" );
} // end else
} // end main
Fig. 6.18 | Linear search of an array. (Part 1 of 2.)
6.8 Searching Arrays
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
245
// compare key to every element of array until the location is found
// or until the end of array is reached; return subscript of element
// if key is found or -1 if key is not found
size_t linearSearch( const int array[], int key, size_t size )
{
size_t n; // counter
// loop through array
for ( n = 0; n < size; ++n ) {
if ( array[ n ] == key ) {
return n; // return location of key
} // end if
} // end for
return -1; // key not found
} // end function linearSearch
Enter integer search key:
36
Found value in element 18
Enter integer search key:
37
Value not found
Fig. 6.18 | Linear search of an array. (Part 2 of 2.)
Searching an Array with Binary Search
The linear searching method works well for small or unsorted arrays. However, for large
arrays linear searching is inefficient. If the array is sorted, the high-speed binary search technique can be used.
The binary search algorithm eliminates from consideration one-half of the elements in
a sorted array after each comparison. The algorithm locates the middle element of the array
and compares it to the search key. If they’re equal, the search key is found and the array
subscript of that element is returned. If they’re not equal, the problem is reduced to
searching one-half of the array. If the search key is less than the middle element of the array,
the first half of the array is searched, otherwise the second half is searched. If the search key
is not found in the specified subarray (piece of the original array), the algorithm is repeated
on one-quarter of the original array. The search continues until the search key is equal to
the middle element of a subarray, or until the subarray consists of one element that’s not
equal to the search key (i.e., the search key is not found).
In a worst case-scenario, searching an array of 1023 elements takes only 10 comparisons using a binary search. Repeatedly dividing 1,024 by 2 yields the values 512, 256, 128,
64, 32, 16, 8, 4, 2 and 1. The number 1,024 (210) is divided by 2 only 10 times to get the
value 1. Dividing by 2 is equivalent to one comparison in the binary search algorithm. An
array of 1,048,576 (220) elements takes a maximum of only 20 comparisons to find the
246
Chapter 6 C Arrays
search key. An array of one billion elements takes a maximum of only 30 comparisons to
find the search key. This is a tremendous increase in performance over the linear search
that required comparing the search key to an average of half of the array elements. For a
one-billion-element array, this is a difference between an average of 500 million comparisons and a maximum of 30 comparisons! The maximum comparisons for any array
can be determined by finding the first power of 2 greater than the number of array elements.
Figure 6.19 presents the iterative version of function binarySearch (lines 42–72).
The function receives four arguments—an integer array b to be searched, an integer
searchKey, the low array subscript and the high array subscript (these define the portion
of the array to be searched). If the search key does not match the middle element of a subarray, the low subscript or high subscript is modified so that a smaller subarray can be
searched. If the search key is less than the middle element, the high subscript is set to
middle - 1 and the search is continued on the elements from low to middle - 1. If the
search key is greater than the middle element, the low subscript is set to middle + 1 and the
search is continued on the elements from middle + 1 to high. The program uses an array
of 15 elements. The first power of 2 greater than the number of elements in this array is
16 (24), so no more than 4 comparisons are required to find the search key. The program
uses function printHeader (lines 75–94) to output the array subscripts and function
printRow (lines 98–118) to output each subarray during the binary search process. The
middle element in each subarray is marked with an asterisk (*) to indicate the element to
which the search key is compared.
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
// Fig. 6.19: fig06_19.c
// Binary search of a sorted array.
#include <stdio.h>
#define SIZE 15
// function prototypes
size_t binarySearch(const int b[], int searchKey, size_t low, size_t high);
void printHeader( void );
void printRow( const int b[], size_t low, size_t mid, size_t high );
// function main begins program execution
int main( void )
{
int a[ SIZE ]; // create array a
size_t i; // counter for initializing elements of array a
int key; // value to locate in array a
size_t result; // variable to hold location of key or -1
// create data
for ( i = 0; i < SIZE; ++i ) {
a[ i ] = 2 * i;
} // end for
printf( "%s", "Enter a number between 0 and 28: " );
scanf( "%d", &key );
Fig. 6.19 | Binary search of a sorted array. (Part 1 of 4.)
6.8 Searching Arrays
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
247
printHeader();
// search for key in array a
result = binarySearch( a, key, 0, SIZE - 1 );
// display results
if ( result != -1 ) {
printf( "\n%d found in array element %d\n", key, result );
} // end if
else {
printf( "\n%d not found\n", key );
} // end else
} // end main
// function to perform binary search of an array
size_t binarySearch(const int b[], int searchKey, size_t low, size_t high)
{
int middle; // variable to hold middle element of array
// loop until low subscript is greater than high subscript
while ( low <= high ) {
// determine middle element of subarray being searched
middle = ( low + high ) / 2;
// display subarray used in this loop iteration
printRow( b, low, middle, high );
// if searchKey matched middle element, return middle
if ( searchKey == b[ middle ] ) {
return middle;
} // end if
// if searchKey less than middle element, set new high
else if ( searchKey < b[ middle ] ) {
high = middle - 1; // search low end of array
} // end else if
// if searchKey greater than middle element, set new low
else {
low = middle + 1; // search high end of array
} // end else
} // end while
return -1; // searchKey not found
} // end function binarySearch
// Print a header for the output
void printHeader( void )
{
unsigned int i; // counter
Fig. 6.19 | Binary search of a sorted array. (Part 2 of 4.)
248
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
Chapter 6 C Arrays
puts( "\nSubscripts:" );
// output column head
for ( i = 0; i < SIZE; ++i ) {
printf( "%3u ", i );
} // end for
puts( "" ); // start new line of output
// output line of - characters
for ( i = 1; i <= 4 * SIZE; ++i ) {
printf( "%s", "-" );
} // end for
puts( "" ); // start new line of output
} // end function printHeader
// Print one row of output showing the current
// part of the array being processed.
void printRow( const int b[], size_t low, size_t mid, size_t high )
{
size_t i; // counter for iterating through array b
// loop through entire array
for ( i = 0; i < SIZE; ++i ) {
// display spaces if outside current subarray range
if ( i < low || i > high ) {
printf( "%s", "
" );
} // end if
else if ( i == mid ) { // display middle element
printf( "%3d*", b[ i ] ); // mark middle value
} // end else if
else { // display other elements in subarray
printf( "%3d ", b[ i ] );
} // end else
} // end for
puts( "" ); // start new line of output
} // end function printRow
Enter a number between 0 and 28: 25
Subscripts:
0
1
2
3
4
5
6
7
8
9 10 11 12 13 14
-----------------------------------------------------------0
2
4
6
8 10 12 14* 16 18 20 22 24 26 28
16 18 20 22* 24 26 28
24 26* 28
24*
25 not found
Fig. 6.19 | Binary search of a sorted array. (Part 3 of 4.)
6.9 Multidimensional Arrays
249
Enter a number between 0 and 28: 8
Subscripts:
0
1
2
3
4
5
6
7
8
9 10 11 12 13 14
-----------------------------------------------------------0
2
4
6
8 10 12 14* 16 18 20 22 24 26 28
0
2
4
6* 8 10 12
8 10* 12
8*
8 found in array element 4
Enter a number between 0 and 28: 6
Subscripts:
0
1
2
3
4
5
6
7
8
9 10 11 12 13 14
-----------------------------------------------------------0
2
4
6
8 10 12 14* 16 18 20 22 24 26 28
0
2
4
6* 8 10 12
6 found in array element 3
Fig. 6.19 | Binary search of a sorted array. (Part 4 of 4.)
6.9 Multidimensional Arrays
Arrays in C can have multiple subscripts. A common use of multiple-subscripted arrays,
which the C standard refers to as multidimensional arrays, is to represent tables of values
consisting of information arranged in rows and columns. To identify a particular table element, we must specify two subscripts: The first (by convention) identifies the element’s
row and the second (by convention) identifies the element’s column. Tables or arrays that
require two subscripts to identify a particular element are called double-subscripted arrays. Multidimensional arrays can have more than two subscripts.
Figure 6.20 illustrates a double-subscripted array, a. The array contains three rows
and four columns, so it’s said to be a 3-by-4 array. In general, an array with m rows and n
columns is called an m-by-n array.
Column 0
Column 1
Column 2
Column 3
Row 0 a[ 0 ][ 0 ]
a[ 0 ][ 1 ]
a[ 0 ][ 2 ]
a[ 0 ][ 3 ]
Row 1
a[ 1 ][ 0 ]
a[ 1 ][ 1 ]
a[ 1 ][ 2 ]
a[ 1 ][ 3 ]
Row 2
a[ 2 ][ 0 ]
a[ 2 ][ 1 ]
a[ 2 ][ 2 ]
a[ 2 ][ 3 ]
Column index
Row index
Array name
Fig. 6.20 | Double-subscripted array with three rows and four columns.
250
Chapter 6 C Arrays
Every element in array a is identified in Fig. 6.20 by an element name of the form
is the name of the array, and i and j are the subscripts that uniquely identify
each element in a. The names of the elements in row 0 all have a first subscript of 0; the
names of the elements in column 3 all have a second subscript of 3.
a[i][j]; a
Common Programming Error 6.7
Referencing a double-subscripted array element as a[ x, y ] instead of a[ x ][ y ] is
a logic error. C interprets a[ x, y ] as a[ y ] (because the comma in this context is
treated as a comma operator), so this programmer error is not a syntax error.
A multidimensional array can be initialized when it’s defined, much like a single-subscripted array. For example, a double-subscripted array int b[2][2] could be defined and
initialized with
int b[ 2 ][ 2 ] = { { 1, 2 }, { 3, 4 } };
The values are grouped by row in braces. The values in the first set of braces initialize row
0 and the values in the second set of braces initialize row 1. So, the values 1 and 2 initialize
elements b[0][0] and b[0][1], respectively, and the values 3 and 4 initialize elements
b[1][0] and b[1][1], respectively. If there are not enough initializers for a given row, the
remaining elements of that row are initialized to 0. Thus,
int b[ 2 ][ 2 ] = { { 1 }, { 3, 4 } };
would initialize b[0][0] to 1, b[0][1] to 0, b[1][0] to 3 and b[1][1] to 4. Figure 6.21
demonstrates defining and initializing double-subscripted arrays.
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
// Fig. 6.21: fig06_21.c
// Initializing multidimensional arrays.
#include <stdio.h>
void printArray( int a[][ 3 ] ); // function prototype
// function main
int main( void )
{
// initialize
int array1[ 2
int array2[ 2
int array3[ 2
begins program execution
array1, array2, array3
][ 3 ] = { { 1, 2, 3 }, { 4, 5, 6 } };
][ 3 ] = { 1, 2, 3, 4, 5 };
][ 3 ] = { { 1, 2 }, { 4 } };
puts( "Values in array1 by row are:" );
printArray( array1 );
puts( "Values in array2 by row are:" );
printArray( array2 );
puts( "Values in array3 by row are:" );
printArray( array3 );
} // end main
Fig. 6.21 | Initializing multidimensional arrays. (Part 1 of 2.)
6.9 Multidimensional Arrays
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
251
// function to output array with two rows and three columns
void printArray( int a[][ 3 ] )
{
size_t i; // row counter
size_t j; // column counter
// loop through rows
for ( i = 0; i <= 1; ++i ) {
// output column values
for ( j = 0; j <= 2; ++j ) {
printf( "%d ", a[ i ][ j ] );
} // end inner for
printf( "\n" ); // start new line of output
} // end outer for
} // end function printArray
Values in array1 by row are:
1 2 3
4 5 6
Values in array2 by row are:
1 2 3
4 5 0
Values in array3 by row are:
1 2 0
4 0 0
Fig. 6.21 | Initializing multidimensional arrays. (Part 2 of 2.)
The program defines three arrays of two rows and three columns (six elements
each). The definition of array1 (line 11) provides six initializers in two sublists. The first
sublist initializes row 0 of the array to the values 1, 2 and 3; and the second sublist initializes row 1 of the array to the values 4, 5 and 6.
If the braces around each sublist are removed from the array1 initializer list, the compiler initializes the elements of the first row followed by the elements of the second row.
The definition of array2 (line 12) provides five initializers. The initializers are assigned to
the first row, then the second row. Any elements that do not have an explicit initializer are
initialized to zero automatically, so array2[1][2] is initialized to 0.
The definition of array3 (line 13) provides three initializers in two sublists. The sublist for the first row explicitly initializes the first two elements of the first row to 1 and 2.
The third element is initialized to zero. The sublist for the second row explicitly initializes
the first element to 4. The last two elements are initialized to zero.
The program calls printArray (lines 26–41) to output each array’s elements. The
function definition specifies the array parameter as const int a[][3]. When we receive a
single-subscripted array as a parameter, the array brackets are empty in the function’s
parameter list. The first subscript of a multidimensional array is not required either, but
all subsequent subscripts are required. The compiler uses these subscripts to determine the
252
Chapter 6 C Arrays
locations in memory of elements in multidimensional arrays. All array elements are stored
consecutively in memory regardless of the number of subscripts. In a double-subscripted
array, the first row is stored in memory followed by the second row.
Providing the subscript values in a parameter declaration enables the compiler to tell
the function how to locate an element in the array. In a double-subscripted array, each row
is basically a single-subscripted array. To locate an element in a particular row, the compiler must know how many elements are in each row so that it can skip the proper number
of memory locations when accessing the array. Thus, when accessing a[1][2] in our
example, the compiler knows to skip the three elements of the first row to get to the second
row (row 1). Then, the compiler accesses element 2 of that row.
Many common array manipulations use for repetition statements. For example, the
following statement sets all the elements in row 2 of array a in Fig. 6.20 to zero:
for ( column = 0; column <= 3; ++column ) {
a[ 2 ][ column ] = 0;
}
We specified row 2, so the first subscript is always 2. The loop varies only the second (column) subscript. The preceding for statement is equivalent to the assignment statements:
a[
a[
a[
a[
2
2
2
2
][
][
][
][
0
1
2
3
]
]
]
]
=
=
=
=
0;
0;
0;
0;
The following nested for statement determines the total of all the elements in array a.
total = 0;
for ( row = 0; row <= 2; ++row ) {
for ( column = 0; column <= 3; ++column ) {
total += a[ row ][ column ];
}
}
The for statement totals the elements of the array one row at a time. The outer for statement begins by setting row (i.e., the row subscript) to 0 so that the elements of that row
may be totaled by the inner for statement. The outer for statement then increments row
to 1, so the elements of that row can be totaled. Then, the outer for statement increments
row to 2, so the elements of the third row can be totaled. When the nested for statement
terminates, total contains the sum of all the elements in the array a.
Two-Dimensonal Array Manipulations
Figure 6.22 performs several other common array manipulations on 3-by-4 array studentGrades using for statements. Each row of the array represents a student and each column represents a grade on one of the four exams the students took during the semester.
The array manipulations are performed by four functions. Function minimum (lines 41–
60) determines the lowest grade of any student for the semester. Function maximum (lines
63–82) determines the highest grade of any student for the semester. Function average
(lines 85–96) determines a particular student’s semester average. Function printArray
(lines 99–118) outputs the double-subscripted array in a neat, tabular format.
6.9 Multidimensional Arrays
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
253
// Fig. 6.22: fig06_22.c
// Double-subscripted array manipulations.
#include <stdio.h>
#define STUDENTS 3
#define EXAMS 4
// function prototypes
int minimum( int grades[][ EXAMS ], size_t pupils, size_t tests );
int maximum( int grades[][ EXAMS ], size_t pupils, size_t tests );
double average( const int setOfGrades[], size_t tests );
void printArray( int grades[][ EXAMS ], size_t pupils, size_t tests );
// function main begins program execution
int main( void )
{
size_t student; // student counter
// initialize student grades for three students (rows)
int studentGrades[ STUDENTS ][ EXAMS ] =
{ { 77, 68, 86, 73 },
{ 96, 87, 89, 78 },
{ 70, 90, 86, 81 } };
// output array studentGrades
puts( "The array is:" );
printArray( studentGrades, STUDENTS, EXAMS );
// determine smallest and largest grade values
printf( "\n\nLowest grade: %d\nHighest grade: %d\n",
minimum( studentGrades, STUDENTS, EXAMS ),
maximum( studentGrades, STUDENTS, EXAMS ) );
// calculate average grade for each student
for ( student = 0; student < STUDENTS; ++student ) {
printf( "The average grade for student %u is %.2f\n",
student, average( studentGrades[ student ], EXAMS ) );
} // end for
} // end main
// Find the minimum grade
int minimum( int grades[][ EXAMS ], size_t pupils, size_t tests )
{
size_t i; // student counter
size_t j; // exam counter
int lowGrade = 100; // initialize to highest possible grade
// loop through rows of grades
for ( i = 0; i < pupils; ++i ) {
// loop through columns of grades
for ( j = 0; j < tests; ++j ) {
if ( grades[ i ][ j ] < lowGrade ) {
Fig. 6.22 | Double-subscripted array manipulations. (Part 1 of 3.)
254
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
Chapter 6 C Arrays
lowGrade = grades[ i ][ j ];
} // end if
} // end inner for
} // end outer for
return lowGrade; // return minimum grade
} // end function minimum
// Find the maximum grade
int maximum( int grades[][ EXAMS ], size_t pupils, size_t tests )
{
size_t i; // student counter
size_t j; // exam counter
int highGrade = 0; // initialize to lowest possible grade
// loop through rows of grades
for ( i = 0; i < pupils; ++i ) {
// loop through columns of grades
for ( j = 0; j < tests; ++j ) {
if ( grades[ i ][ j ] > highGrade ) {
highGrade = grades[ i ][ j ];
} // end if
} // end inner for
} // end outer for
return highGrade; // return maximum grade
} // end function maximum
// Determine the average grade for a particular student
double average( const int setOfGrades[], size_t tests )
{
size_t i; // exam counter
int total = 0; // sum of test grades
// total all grades for one student
for ( i = 0; i < tests; ++i ) {
total += setOfGrades[ i ];
} // end for
return ( double ) total / tests; // average
} // end function average
// Print the array
void printArray( int grades[][ EXAMS ], size_t pupils, size_t tests )
{
size_t i; // student counter
size_t j; // exam counter
// output column heads
printf( "%s", "
[0]
[1]
[2]
Fig. 6.22 | Double-subscripted array manipulations. (Part 2 of 3.)
[3]" );
6.9 Multidimensional Arrays
255
107
// output grades in tabular format
108
for ( i = 0; i < pupils; ++i ) {
109
110
// output label for row
111
printf( "\nstudentGrades[%d] ", i );
112
113
// output grades for one student
114
for ( j = 0; j < tests; ++j ) {
115
printf( "%-5d", grades[ i ][ j ] );
116
} // end inner for
117
} // end outer for
118 } // end function printArray
The array is:
[0]
studentGrades[0] 77
studentGrades[1] 96
studentGrades[2] 70
[1]
68
87
90
[2]
86
89
86
[3]
73
78
81
Lowest grade: 68
Highest grade: 96
The average grade for student 0 is 76.00
The average grade for student 1 is 87.50
The average grade for student 2 is 81.75
Fig. 6.22 | Double-subscripted array manipulations. (Part 3 of 3.)
Functions
minimum, maximum
and printArray each receive three arguments—the
array (called grades in each function), the number of students (rows of
the array) and the number of exams (columns of the array). Each function loops through
array grades using nested for statements. The following nested for statement is from the
function minimum definition:
studentGrades
// loop through rows of grades
for ( i = 0; i < pupils; ++i ) {
// loop through columns of grades
for ( j = 0; j < tests; ++j ) {
if ( grades[ i ][ j ] < lowGrade ) {
lowGrade = grades[ i ][ j ];
} // end if
} // end inner for
} // end outer for
The outer for statement begins by setting i (i.e., the row subscript) to 0 so the elements
of that row (i.e., the grades of the first student) can be compared to variable lowGrade in
the body of the inner for statement. The inner for statement loops through the four
grades of a particular row and compares each grade to lowGrade. If a grade is less than
lowGrade, lowGrade is set to that grade. The outer for statement then increments the row
subscript to 1. The elements of that row are compared to variable lowGrade. The outer for
statement then increments the row subscript to 2. The elements of that row are compared
to variable lowGrade. When execution of the nested statement is complete, lowGrade contains the smallest grade in the double-subscripted array. Function maximum works similarly
to function minimum.
256
Chapter 6 C Arrays
Function average (lines 85–96) takes two arguments—a single-subscripted array of
test results for a particular student called setOfGrades and the number of test results in
the array. When average is called, the first argument studentGrades[student] is passed.
This causes the address of one row of the double-subscripted array to be passed to average.
The argument studentGrades[1] is the starting address of row 1 of the array. Remember
that a double-subscripted array is basically an array of single-subscripted arrays and that
the name of a single-subscripted array is the address of the array in memory. Function
average calculates the sum of the array elements, divides the total by the number of test
results and returns the floating-point result.
6.10 Variable-Length Arrays
In early versions of C, all arrays had constant size. But what if you don’t know an array’s
size at compilation time? To handle this, you’d have to use dynamic memory allocation
with malloc and related functions. The C standard allows you to handle arrays of unknown size using variable-length arrays (VLAs). These are not arrays whose size can
change—that would compromise the integrity of nearby locations in memory. A variablelength array is an array whose length, or size, is defined in terms of an expression evaluated
at execution time. The program of Fig. 6.23 declares and prints several VLAs. [Note: This
feature is not supported in Microsoft Visual C++.]
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
// Fig. 6.23: figG_14.c
// Using variable-length arrays in C99
#include <stdio.h>
// function prototypes
void print1DArray( int size, int arr[ size ] );
void print2DArray( int row, int col, int arr[ row ][ col ] );
int main( void )
{
int arraySize; // size of 1-D array
int row1, col1, row2, col2; // number of rows and columns in 2-D arrays
printf( "%s", "Enter size of a one-dimensional array: " );
scanf( "%d", &arraySize );
printf( "%s", "Enter number of rows and columns in a 2-D array: " );
scanf( "%d %d", &row1, &col1 );
printf( "%s",
"Enter number of rows and columns in another 2-D array: " );
scanf( "%d %d", &row2, &col2 );
int array[ arraySize ]; // declare 1-D variable-length array
int array2D1[ row1 ][ col1 ]; // declare 2-D variable-length array
int array2D2[ row2 ][ col2 ]; // declare 2-D variable-length array
Fig. 6.23 | Using variable-length arrays in C99. (Part 1 of 3.)
6.10 Variable-Length Arrays
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
257
// test sizeof operator on VLA
printf( "\nsizeof(array) yields array size of %d bytes\n",
sizeof( array ) );
// assign elements of 1-D VLA
for ( int i = 0; i < arraySize; ++i ) {
array[ i ] = i * i;
} // end for
// assign elements of first 2-D VLA
for ( int i = 0; i < row1; ++i ) {
for ( int j = 0; j < col1; ++j ) {
array2D1[ i ][ j ] = i + j;
} // end for
} // end for
// assign elements of second 2-D VLA
for ( int i = 0; i < row2; ++i ) {
for ( int j = 0; j < col2; ++j ) {
array2D2[ i ][ j ] = i + j;
} // end for
} // end for
puts( "\nOne-dimensional array:" );
print1DArray( arraySize, array ); // pass 1-D VLA to function
puts( "\nFirst two-dimensional array:" );
print2DArray( row1, col1, array2D1 ); // pass 2-D VLA to function
puts( "\nSecond two-dimensional array:" );
print2DArray( row2, col2, array2D2 ); // pass other 2-D VLA to function
} // end main
void print1DArray( int size, int array[ size ] )
{
// output contents of array
for ( int i = 0; i < size; i++ ) {
printf( "array[%d] = %d\n", i, array[ i ] );
} // end for
} // end function print1DArray
void print2DArray( int row, int col, int arr[ row ][ col ] )
{
// output contents of array
for ( int i = 0; i < row; ++i ) {
for ( int j = 0; j < col; ++j ) {
printf( "%5d", arr[ i ][ j ] );
} // end for
puts( "" );
} // end for
} // end function print2DArray
Fig. 6.23 | Using variable-length arrays in C99. (Part 2 of 3.)
258
Chapter 6 C Arrays
Enter size of a one-dimensional array: 6
Enter number of rows and columns in a 2-D array: 2 5
Enter number of rows and columns in another 2-D array: 4 3
sizeof(array) yields array size of 24 bytes
One-dimensional array:
array[0] = 0
array[1] = 1
array[2] = 4
array[3] = 9
array[4] = 16
array[5] = 25
First two-dimensional array:
0
1
2
3
4
1
2
3
4
5
Second two-dimensional array:
0
1
2
1
2
3
2
3
4
3
4
5
Fig. 6.23 | Using variable-length arrays in C99. (Part 3 of 3.)
First, we prompt the user for the desired sizes for a one-dimensional array and two
two-dimensional arrays (lines 14–22). Lines 24–26 then declare VLAs of the appropriate
size. This is valid as long as the variables representing the array sizes are of an integral type.
After declaring the arrays, we use the sizeof operator in lines 29–30 to make sure that
our VLA is of the proper length. In early versions of C sizeof was always a compile-time
operation, but when applied to a VLA, sizeof operates at runtime. The output window
shows that the sizeof operator returns a size of 24 bytes—four times that of the number
we entered because the size of an int on our machine is 4 bytes.
Next we assign values to the elements of our VLAs (lines 33–49). We use i < arraySize as our loop-continuation condition when filling the one-dimensional array. As with
fixed-length arrays, there is no protection against stepping outside the array bounds.
Lines 61–67 define function print1DArray that takes a one-dimensional VLA. The
syntax for passing VLAs as parameters to functions is the same as with a normal, fixedlength array. We use the variable size in the declaration of the array parameter, but no
checking is performed other than the variable being defined and of integral type—it’s
purely documentation for the programmer.
Function print2DArray (lines 69–79) takes a variable-length two-dimensional array
and displays it to the screen. Recall from Section 6.9 that, all but the first subscript of a
multidimensional array must be specified when declaring a function parameter. The same
restriction holds true for VLAs, except that the sizes can be specified by variables. The initial value of col passed to the function is used to convert from two-dimensional indices to
offsets into the contiguous memory the array is stored in, just as with a fixed-size array.
Changing the value of col inside the function will not cause any changes to the indexing,
but passing an incorrect value to the function will.
6.11 Secure C Programming
259
6.11 Secure C Programming
Bounds Checking for Array Subscripts
It’s important to ensure that every subscript you use to access an array element is within
the array’s bounds—that is, greater than or equal to 0 and less than the number of array
elements. A two-dimensional array’s row and column subscripts must be greater than or
equal to 0 and less than the numbers of rows and columns, respectively. This extends to
arrays with additional dimensions as well.
Allowing programs to read from or write to array elements outside the bounds of arrays
are common security flaws. Reading from out-of-bounds array elements can cause a program to crash or even appear to execute correctly while using bad data. Writing to an outof-bounds element (known as a buffer overflow) can corrupt a program’s data in memory,
crash a program and allow attackers to exploit the system and execute their own code.
As we stated in the chapter, C provides no automatic bounds checking for arrays, so you
must provide your own. For techniques that help you prevent such problems, see CERT
guideline ARR30-C at www.securecoding.cert.org.
scanf_s
Bounds checking is also important in string processing. When reading a string into a char
array, scanf does not prevent buffer overflows. If the number of characters input is greater
than or equal to the array’s length, scanf will write characters—including the string’s terminating null character ('\0')—beyond the end of the array. This might overwrite other
variables’ values, and eventually the program might overwrite the string’s '\0' if it writes
to those other variables.
Functions determine where strings end by looking for their terminating '\0' character. For example, function printf outputs a string by reading characters from the beginning of the string in memory and continuing until the string’s '\0' is encountered. If the
'\0' is missing, printf might read far beyond the end of the string until it encounters
some other '\0' in memory.
The C standard’s optional Annex K provides new, more secure, versions of many
string-processing and input/output functions, including scanf_s—a version of scanf that
performs additional checks to ensure that it does not write beyond the end of a character
array used to store a string. Assuming that myString is a 20-character array, the statement
scanf_s( "%19s", myString, 20 );
reads a string into myString. Function scanf_s requires two arguments for each %s in the
format string—a character array in which to place the input string and the number of array
elements. The second of these arguments is used by scanf_s to prevent buffer overflows.
For example, it’s possible to supply a field width for %s that’s too long for the underlying
character array, or to simply omit the field width entirely. If the number of characters input plus the terminating null character is larger than the number of array elements, the %s
conversion would fail. Because the preceding statement contains only one conversion
specifier, scanf_s would return 0 indicating that no conversions were performed, and
myString would be unaltered.
In general, if your compiler supports the functions from the C standard’s optional
Annex K, you should use them. We discuss additional Annex K functions in later Secure
C Programming sections.
260
Chapter 6 C Arrays
Don’t Use Strings Read from the User as Format-Control Strings
You might have noticed that throughout this book, we never use single-argument printfs.
Instead we use one of the following forms:
• When we need to output a '\n' after the string, we use function puts (which automatically outputs a '\n' after its single string argument), as in
puts( "Welcome to C!" );
•
When we need the cursor to remain on the same line as the string, we use function printf, as in
printf( "%s", "Enter first integer: " );
Because we were displaying string literals, we certainly could have used the one-argument
form of printf, as in
printf( "Welcome to C!\n" );
printf( "Enter first integer: " );
When printf evaluates the format-control string in its first (and possibly its only) argument, the function performs tasks based on the conversion specifier(s) in that string. If the
format-control string were obtained from the user, an attacker could supply malicious conversion specifiers that would be “executed” by the formatted output function. Now that you
know how to read strings into character arrays, it’s important to note that you should never
use as a printf’s format-control string a character array that might contain user input. For
more information, see CERT guideline FIO30-C at www.securecoding.cert.org.
Summary
Section 6.1 Introduction
• Arrays are data structures consisting of related data items of the same type.
• Arrays are “static” entities in that they remain the same size throughout program execution.
Section 6.2 Arrays
• An array is a group of memory locations related by the fact that they all have the same name and
the same type.
• To refer to a particular location or element in the array, specify the array’s name and the position
number of the particular element in the array.
• The first element in every array is the zeroth element. Thus, the first element of array c is referred
to as c[0], the second element is referred to as c[1], the seventh element is referred to as c[6],
and, in general, the ith element is referred to as c[i - 1].
• An array name, like other variable names, can contain only letters, digits and underscores and
cannot begin with a digit.
• The position number contained within square brackets is more formally called a subscript. A subscript must be an integer or an integer expression.
• The brackets used to enclose the subscript of an array are actually considered to be an operator
in C. They have the same level of precedence as the function call operator.
Summary
261
Section 6.3 Defining Arrays
• Arrays occupy space in memory. You specify the type of each element and the number of elements in the array so that the computer may reserve the appropriate amount of memory.
• An array of type char can be used to store a character string.
Section 6.4 Array Examples
• Type size_t represents an unsigned integral type. This type is recommended for any variable
that represents an array’s size or an array’s subscripts. The header <stddef.h> defines size_t and
is often included by other headers (such as <stdio.h>).
• The elements of an array can be initialized when the array is defined by following the definition
with an equals sign and braces, {}, containing a comma-separated list of initializers. If there are
fewer initializers than elements in the array, the remaining elements are initialized to zero.
• The statement int n[10] = {0}; explicitly initializes the first element to zero and initializes the
remaining nine elements to zero because there are fewer initializers than there are elements in the
array. It’s important to remember that automatic arrays are not automatically initialized to zero.
You must at least initialize the first element to zero for the remaining elements to be automatically zeroed. This method of initializing the array elements to 0 is performed before program
startup for static arrays and at runtime for automatic arrays.
• If the array size is omitted from a definition with an initializer list, the number of elements in the
array will be the number of elements in the initializer list.
• The #define preprocessor directive can be used to define a symbolic constant—an identifier that
the preprocessor replaces with replacement text before the program is compiled. When the program is preprocessed, all occurrences of the symbolic constant are replaced with the replacement
text. Using symbolic constants to specify array sizes makes programs more scalable.
• C has no array bounds checking to prevent a program from referring to an element that does not
exist. Thus, an executing program can “walk off” the end of an array without warning. You
should ensure that all array references remain within the bounds of the array.
• A string such as "hello" is really a static array of individual characters in C.
• A character array can be initialized using a string literal. In this case, the size of the array is determined by the compiler based on the length of the string.
• Every string contains a special string-termination character called the null character. The character constant representing the null character is '\0'.
• A character array representing a string should always be defined large enough to hold the number
of characters in the string and the terminating null character.
• Character arrays also can be initialized with individual character constants in an initializer list.
• Because a string is really an array of characters, we can access individual characters in a string directly using array subscript notation.
• You can input a string directly into a character array from the keyboard using scanf and the conversion specifier %s. The name of the character array is passed to scanf without the preceding &
used with nonstring variables.
• Function scanf reads characters from the keyboard until the first white-space character is encountered—it does not check the array size. Thus, scanf can write beyond the end of the array.
• A character array representing a string can be output with printf and the %s conversion specifier.
The characters of the string are printed until a terminating null character is encountered.
• A static local variable exists for the duration of the program but is visible only in the function
body. We can apply static to a local array definition so that the array is not created and initial-
262
Chapter 6 C Arrays
ized each time the function is called and the array is not destroyed each time the function is exited
in the program. This reduces program execution time, particularly for programs with frequently
called functions that contain large arrays.
• Arrays that are static are automatically initialized once before program startup. If you do not
explicitly initialize a static array, that array’s elements are initialized to zero by the compiler.
Section 6.5 Passing Arrays to Functions
• To pass an array argument to a function, specify the name of the array without any brackets.
• Unlike char arrays that contain strings, other array types do not have a special terminator. For
this reason, the size of an array is passed to a function, so that the function can process the proper
number of elements.
• C automatically passes arrays to functions by reference—the called functions can modify the element values in the callers’ original arrays. The name of the array evaluates to the address of the
first element of the array. Because the starting address of the array is passed, the called function
knows precisely where the array is stored. Therefore, when the called function modifies array elements in its function body, it’s modifying the actual elements of the array in their original memory locations.
• Although entire arrays are passed by reference, individual array elements are passed by value exactly as simple variables are.
• Such simple single pieces of data (such as individual ints, floats and chars) are called scalars.
• To pass an element of an array to a function, use the subscripted name of the array element as an
argument in the function call.
• For a function to receive an array through a function call, the function’s parameter list must specify that an array will be received. The size of the array is not required between the array brackets.
If it’s included, the compiler checks that it’s greater than zero, then ignores it.
• When an array parameter is preceded by the const qualifier, the elements of the array become
constant in the function body, and any attempt to modify an element of the array in the function
body results in a compile-time error.
Section 6.6 Sorting Arrays
• Sorting data (i.e., placing the data into a particular order such as ascending or descending) is one
of the most important computing applications.
• One sorting technique is called the bubble sort or the sinking sort, because the smaller values
gradually “bubble” their way upward to the top of the array like air bubbles rising in water, while
the larger values sink to the bottom of the array. The technique is to make several passes through
the array. On each pass, successive pairs of elements are compared. If a pair is in increasing order
(or if the values are identical), we leave the values as they are. If a pair is in decreasing order, their
values are swapped in the array.
• Because of the way the successive comparisons are made, a large value may move down the array
many positions on a single pass, but a small value may move up only one position.
• The chief virtue of the bubble sort is that it’s easy to program. However, the bubble sort runs
slowly. This becomes apparent when sorting large arrays.
Section 6.7 Case Study: Computing Mean, Median and Mode Using Arrays
• The mean is the arithmetic average of a set of values.
• The median is the “middle value” in a sorted set of values.
• The mode is the value that occurs most frequently in a set of values.
Summary
263
Section 6.8 Searching Arrays
• The process of finding a particular element of an array is called searching.
• The linear search compares each element of the array with the search key. Because the array is not
in any particular order, it’s just as likely that the value will be found in the first element as in the
last. On average, therefore, the search key will be compared with half the elements of the array.
• The linear searching method works well for small or unsorted arrays. For sorted arrays, the highspeed binary search technique can be used.
• The binary search algorithm eliminates from consideration one-half of the elements in a sorted array after each comparison. The algorithm locates the middle element of the array and compares it
to the search key. If they’re equal, the search key is found and the array subscript of that element
is returned. If they’re not equal, the problem is reduced to searching one-half of the array. If
the search key is less than the middle element of the array, the first half of the array is searched, otherwise the second half is searched. If the search key is not found in the specified subarray (piece of
the original array), the algorithm is repeated on one-quarter of the original array. The search continues until the search key is equal to the middle element of a subarray, or until the subarray consists
of one element that’s not equal to the search key (i.e., the search key is not found).
• When using a binary search, the maximum number of comparisons required for any array can
be determined by finding the first power of 2 greater than the number of array elements.
Section 6.9 Multidimensional Arrays
• A common use of multidimensional arrays is to represent tables of values consisting of information arranged in rows and columns. To identify a particular table element, we must specify two
subscripts: The first (by convention) identifies the element’s row and the second (by convention)
identifies the element’s column.
• Tables or arrays that require two subscripts to identify a particular element are called double-subscripted arrays.
• Multidimensional arrays can have more than two subscripts.
• A multidimensional array can be initialized when it’s defined, much like a single-subscripted array. The values are grouped by row in braces. If there are not enough initializers for a given row,
the remaining elements of that row are initialized to 0.
• The first subscript of a multidimensional array parameter declaration is not required, but all subsequent subscripts are required. The compiler uses these subscripts to determine the locations in
memory of elements in multidimensional arrays. All array elements are stored consecutively in
memory regardless of the number of subscripts. In a double-subscripted array, the first row is
stored in memory followed by the second row.
• Providing the subscript values in a parameter declaration enables the compiler to tell the function
how to locate an array element. In a double-subscripted array, each row is basically a single-subscripted array. To locate an element in a particular row, the compiler must know how many elements are
in each row so that it can skip the proper number of memory locations when accessing the array.
Section 6.10 Variable-Length Arrays
• A variable-length array is an array whose length, or size, is defined in terms of an expression evaluated at execution time.
• When applied to a variable-length array, sizeof operates at runtime.
• As with fixed-length arrays, there is no protection against stepping outside the array bounds of
variable-length arrays.
• The syntax for passing variable-length arrays as parameters to functions is the same as with a normal, fixed-length array.
264
Chapter 6 C Arrays
Terminology
#define preprocessor directive
array 217
array initializer 220
binary search 244
bubble sort 237
const keyword 234
double-subscripted array 249
element 217
key value 244
linear search 244
m-by-n array 249
multidimensional array 249
name 218
null character 228
position number 217
replacement text 222
scalable 222
scalar 233
search key 244
searching 244
sinking sort 237
subscript 218
survey data analysis 239
symbolic constant 222
table 249
value 218
variable-length array 256
zeroth element 217
Self-Review Exercises
6.1
6.2
6.3
6.4
Answer each of the following:
.
a) Lists and tables of values are stored in
b) The number used to refer to a particular element of an array is called its
.
c) A(n)
should be used to specify the size of an array because it makes the program more scalable.
the array.
d) The process of placing the elements of an array in order is called
e) Determining whether an array contains a certain key value is called
the array.
array.
f) An array that uses two subscripts is referred to as a(n)
State whether the following are true or false. If the answer is false, explain why.
a) An array can store many different types of values.
b) An array subscript can be of data type double.
c) If there are fewer initializers in an initializer list than the number of elements in the array,
C automatically initializes the remaining elements to the last value in the list of initializers.
d) It’s an error if an initializer list contains more initializers than there are elements in the array.
e) An individual array element that’s passed to a function as an argument of the form a[i]
and modified in the called function will contain the modified value in the calling function.
Follow the instructions below regarding an array called fractions.
a) Define a symbolic constant SIZE to be replaced with the replacement text 10.
b) Define an array with SIZE elements of type double and initialize the elements to 0.
c) Refer to array element 4.
d) Assign the value 1.667 to array element nine.
e) Assign the value 3.333 to the seventh element of the array.
f) Print array elements 6 and 9 with two digits of precision to the right of the decimal
point, and show the output that’s displayed on the screen.
g) Print all the elements of the array, using a for repetition statement. Assume the integer
variable x has been defined as a control variable for the loop. Show the output.
Write statements to accomplish the following:
a) Define table to be an integer array and to have 3 rows and 3 columns. Assume the symbolic constant SIZE has been defined to be 3.
b) How many elements does the array table contain? Print the total number of elements.
c) Use a for repetition statement to initialize each element of table to the sum of its subscripts. Assume the integer variables x and y are defined as control variables.
Answers to Self-Review Exercises
265
d) Print the values of each element of array table. Assume the array was initialized with
the definition:
int table[ SIZE ][ SIZE ] =
{ { 1, 8 }, { 2, 4, 6 }, { 5 } };
6.5
Find the error in each of the following program segments and correct the error.
a) #define SIZE 100;
b) SIZE = 10;
c) Assume int b[ 10 ] = { 0 }, i;
for ( i = 0; i <= 10; ++i ) {
b[ i ] = 1;
}
d) #include <stdio.h>;
e) Assume int a[ 2 ][ 2
] = { { 1, 2 }, { 3, 4 } };
a[ 1, 1 ] = 5;
f)
#define VALUE = 120
Answers to Self-Review Exercises
6.1
ed.
a) arrays. b) subscript. c) symbolic constant. d) sorting. e) searching. f) double-subscript-
6.2
a)
b)
c)
d)
e)
False. An array can store only values of the same type.
False. An array subscript must be an integer or an integer expression.
False. C automatically initializes the remaining elements to zero.
True.
False. Individual elements of an array are passed by value. If the entire array is passed to
a function, then any modifications to the elements will be reflected in the original.
6.3
a)
b)
c)
d)
e)
f)
#define SIZE 10
g)
for ( x = 0; x < SIZE; ++x ) {
double fractions[ SIZE ] = { 0.0 };
fractions[ 4 ]
fractions[ 9 ] = 1.667;
fractions[ 6 ] = 3.333;
printf( "%.2f %.2f\n", fractions[ 6 ], fractions[ 9 ] );
Output: 3.33
1.67.
printf( "fractions[%u] = %f\n", x, fractions[ x ] );
}
Output:
fractions[0] = 0.000000
fractions[1] = 0.000000
fractions[2] = 0.000000
fractions[3] = 0.000000
fractions[4] = 0.000000
fractions[5] = 0.000000
fractions[6] = 3.333000
fractions[7] = 0.000000
fractions[8] = 0.000000
fractions[9] = 1.667000
6.4
a) int table[ SIZE ][ SIZE ];
b) Nine elements. printf( "%d\n",
SIZE * SIZE );
266
Chapter 6 C Arrays
c)
for ( x = 0; x < SIZE; ++x ) {
for ( y = 0; y < SIZE; ++y ) {
table[ x ][ y ] = x + y;
}
}
d)
for ( x = 0; x < SIZE; ++x ) {
for ( y = 0; y < SIZE; ++y ) {
printf( "table[%d][%d] = %d\n", x, y, table[ x ][ y ] );
}
}
Output:
table[0][0]
table[0][1]
table[0][2]
table[1][0]
table[1][1]
table[1][2]
table[2][0]
table[2][1]
table[2][2]
6.5
=
=
=
=
=
=
=
=
=
1
8
0
2
4
6
5
0
0
a) Error: Semicolon at the end of the #define preprocessor directive.
Correction: Eliminate semicolon.
b) Error: Assigning a value to a symbolic constant using an assignment statement.
Correction: Assign a value to the symbolic constant in a #define preprocessor directive
without using the assignment operator as in #define SIZE 10.
c) Error: Referencing an array element outside the bounds of the array (b[ 10 ]).
Correction: Change the final value of the control variable to 9.
d) Error: Semicolon at the end of the #include preprocessor directive.
Correction: Eliminate semicolon.
e) Error: Array subscripting done incorrectly.
Correction: Change the statement to a[ 1 ][ 1 ] = 5;
f) Error: Assigning a value to a symbolic constant using an assignment statement.
Correction: Assign a value to the symbolic constant in a #define preprocessor
directive without using the assignment operator as in #define VALUE 120.
Exercises
6.6
Fill in the blanks in each of the following:
a) C stores lists of values in
.
b) The elements of an array are related by the fact that they
.
c) When referring to an array element, the position number contained within square
.
brackets is called a(n)
d) The names of the five elements of array p are
,
,
,
and
.
e) The contents of a particular element of an array is called the
of that element.
f) Naming an array, stating its type and specifying the number of elements in the array is
the array.
called
g) The process of placing the elements of an array into either ascending or descending order is called
.
of an element
h) In a double-subscripted array, the first subscript identifies the
and the second subscript identifies the
of an element.
Exercises
6.7
267
i) An m-by-n array contains
rows,
columns and
elements.
.
j) The name of the element in row 3 and column 5 of array d is
State which of the following are true and which are false. If false, explain why.
a) To refer to a particular location or element within an array, we specify the name of the
array and the value of the particular element.
b) An array definition reserves space for the array.
c) To indicate that 100 locations should be reserved for integer array p, write
p[ 100 ];
d) A C program that initializes the elements of a 15-element array to zero must contain
one for statement.
e) A C program that totals the elements of a double-subscripted array must contain nested
for statements.
f) The mean, median and mode of the following set of values are 5, 6 and 7, respectively:
1, 2, 5, 6, 7, 7, 7.
6.8
Write statements to accomplish each of the following:
a) Display the value of the seventh element of character array f.
b) Input a value into element 4 of single-subscripted floating-point array b.
c) Initialize each of the five elements of single-subscripted integer array g to 8.
d) Total the elements of floating-point array c of 100 elements.
e) Copy array a into the first portion of array b. Assume double a[11], b[34];
f) Determine and print the smallest and largest values contained in 99-element floatingpoint array w.
6.9
Consider a 2-by-5 integer array t.
a) Write a definition for t.
b) How many rows does t have?
c) How many columns does t have?
d) How many elements does t have?
e) Write the names of all the elements in the second row of t.
f) Write the names of all the elements in the third column of t.
g) Write a single statement that sets the element of t in row 1 and column 2 to zero.
h) Write a series of statements that initialize each element of t to zero. Do not use a repetition structure.
i) Write a nested for statement that initializes each element of t to zero.
j) Write a statement that inputs the values for the elements of t from the terminal.
k) Write a series of statements that determine and print the smallest value in array t.
l) Write a statement that displays the elements of the first row of t.
m) Write a statement that totals the elements of the fourth column of t.
n) Write a series of statements that print the array t in tabular format. List the column subscripts as headings across the top and list the row subscripts at the left of each row.
6.10 (Sales Commissions) Use a single-subscripted array to solve the following problem. A company pays its salespeople on a commission basis. The salespeople receive $200 per week plus 9% of
their gross sales for that week. For example, a salesperson who grosses $3,000 in sales in a week receives $200 plus 9% of $3,000, or a total of $470. Write a C program (using an array of counters)
that determines how many of the salespeople earned salaries in each of the following ranges (assume
that each salesperson’s salary is truncated to an integer amount):
a) $200–299
b) $300–399
c) $400–499
d) $500–599
268
Chapter 6 C Arrays
e)
f)
g)
h)
i)
$600–699
$700–799
$800–899
$900–999
$1000 and over
6.11 (Bubble Sort) The bubble sort presented in Fig. 6.15 is inefficient for large arrays. Make the
following simple modifications to improve its performance.
a) After the first pass, the largest number is guaranteed to be in the highest-numbered element of the array; after the second pass, the two highest numbers are “in place,” and
so on. Instead of making nine comparisons on every pass, modify the bubble sort to
make eight comparisons on the second pass, seven on the third pass and so on.
b) The data in the array may already be in the proper or near-proper order, so why make
nine passes if fewer will suffice? Modify the sort to check at the end of each pass whether
any swaps have been made. If none has been made, then the data must already be in the
proper order, so the program should terminate. If swaps have been made, then at least
one more pass is needed.
6.12
Write loops that perform each of the following single-subscripted array operations:
a) Initialize the 10 elements of integer array counts to zeros.
b) Add 1 to each of the 15 elements of integer array bonus.
c) Read the 12 values of floating-point array monthlyTemperatures from the keyboard.
d) Print the five values of integer array bestScores in column format.
6.13
Find the error(s) in each of the following statements:
a) Assume: char str[ 5 ];
scanf( "%s", str ); // User types hello
b) Assume: int
a[ 3 ];
printf( "$d
%d
c) double f[ 3 ] = {
d) Assume: double d[
%d\n", a[ 1 ], a[ 2 ], a[ 3 ] );
1.1, 10.01, 100.001, 1000.0001 };
2 ][ 10 ];
d[ 1, 9 ] = 2.345;
6.14 (Mean, Median and Mode Program Modifications) Modify the program of Fig. 6.16 so
function mode is capable of handling a tie for the mode value. Also modify function median so the
two middle elements are averaged in an array with an even number of elements.
6.15 (Duplicate Elimination) Use a single-subscripted array to solve the following problem.
Read in 20 numbers, each of which is between 10 and 100, inclusive. As each number is read, print
it only if it’s not a duplicate of a number already read. Provide for the “worst case” in which all 20
numbers are different. Use the smallest possible array to solve this problem.
6.16 Label the elements of 3-by-5 double-subscripted array sales to indicate the order in which
they’re set to zero by the following program segment:
for ( row = 0; row <= 2; ++row ) {
for ( column = 0; column <= 4; ++column ) {
sales[ row ][ column ] = 0;
}
}
6.17
1
2
3
4
What does the following program do?
// ex06_17.c
// What does this program do?
#include <stdio.h>
#define SIZE 10
Exercises
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
int whatIsThis( const int b[], size_t p ); // function prototype
// function main begins program execution
int main( void )
{
int x; // holds return value of function whatIsThis
// initialize array a
int a[ SIZE ] = { 1, 2, 3, 4, 5, 6, 7, 8, 9, 10 };
x = whatIsThis( a, SIZE );
printf( "Result is %d\n", x );
} // end main
// what does this function do?
int whatIsThis( const int b[], size_t p )
{
// base case
if ( 1 == p ) {
return b[ 0 ];
} // end if
else { // recursion step
return b[ p - 1 ] + whatIsThis( b, p - 1 );
} // end else
} // end function whatIsThis
6.18
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
269
What does the following program do?
// ex06_18.c
// What does this program do?
#include <stdio.h>
#define SIZE 10
// function prototype
void someFunction( const int b[], size_t startSubscript, size_t size );
// function main begins program execution
int main( void )
{
int a[ SIZE ] = { 8, 3, 1, 2, 6, 0, 9, 7, 4, 5 }; // initialize a
puts( "Answer is:" );
someFunction( a, 0, SIZE );
puts( "" );
} // end main
// What does this function do?
void someFunction( const int b[], size_t startSubscript, size_t size )
{
if ( startSubscript < size ) {
someFunction( b, startSubscript + 1, size );
printf( "%d ", b[ startSubscript ] );
} // end if
} // end function someFunction
6.19 (Dice Rolling) Write a program that simulates the rolling of two dice. The program should
use rand twice to roll the first die and second die, respectively. The sum of the two values should
270
Chapter 6 C Arrays
then be calculated. [Note: Because each die can show an integer value from 1 to 6, then the sum of
the two values will vary from 2 to 12, with 7 being the most frequent sum and 2 and 12 the least
frequent sums.] Figure 6.24 shows the 36 possible combinations of the two dice. Your program
should roll the two dice 36,000 times. Use a single-subscripted array to tally the numbers of times
each possible sum appears. Print the results in a tabular format. Also, determine if the totals are reasonable; i.e., there are six ways to roll a 7, so approximately one-sixth of all the rolls should be 7.
1
2
3
4
5
6
1
2
3
4
5
6
7
2
3
4
5
6
7
8
3
4
5
6
7
8
9
4
5
6
7
8
9
10
5
6
7
8
9
10
11
6
7
8
9
10
11
12
Fig. 6.24 | Dice-rolling outcomes.
6.20 (Game of Craps) Write a program that runs 1000 games of craps (without human intervention) and answers each of the following questions:
a) How many games are won on the first roll, second roll, …, twentieth roll and after the
twentieth roll?
b) How many games are lost on the first roll, second roll, …, twentieth roll and after the
twentieth roll?
c) What are the chances of winning at craps? [Note: You should discover that craps is one
of the fairest casino games. What do you suppose this means?]
d) What’s the average length of a game of craps?
e) Do the chances of winning improve with the length of the game?
6.21 (Airline Reservations System) A small airline has just purchased a computer for its new automated reservations system. The president has asked you to program the new system. You’ll write
a program to assign seats on each flight of the airline’s only plane (capacity: 10 seats).
Your program should display the following menu of alternatives:
Please type 1 for "first class"
Please type 2 for "economy"
If the person types 1, then your program should assign a seat in the first class section (seats 1–
5). If the person types 2, then your program should assign a seat in the economy section (seats 6–
10). Your program should then print a boarding pass indicating the person's seat number and
whether it’s in the first class or economy section of the plane.
Use a single-subscripted array to represent the seating chart of the plane. Initialize all the elements of the array to 0 to indicate that all seats are empty. As each seat is assigned, set the corresponding element of the array to 1 to indicate that the seat is no longer available.
Your program should, of course, never assign a seat that has already been assigned. When the
first class section is full, your program should ask the person if it’s acceptable to be placed in the
economy section (and vice versa). If yes, then make the appropriate seat assignment. If no, then
print the message "Next flight leaves in 3 hours."
Exercises
271
6.22 (Total Sales) Use a double-subscripted array to solve the following problem. A company has
four salespeople (1 to 4) who sell five different products (1 to 5). Once a day, each salesperson passes
in a slip for each different type of product sold. Each slip contains:
a) The salesperson number
b) The product number
c) The total dollar value of that product sold that day
Thus, each salesperson passes in between 0 and 5 sales slips per day. Assume that the information
from all of the slips for last month is available. Write a program that will read all this information
for last month’s sales and summarize the total sales by salesperson by product. All totals should be
stored in the double-subscripted array sales. After processing all the information for last month,
print the results in tabular format with each column representing a particular salesperson and each
row representing a particular product. Cross total each row to get the total sales of each product for
last month; cross total each column to get the total sales by salesperson for last month. Your tabular
printout should include these cross totals to the right of the totaled rows and to the bottom of the
totaled columns.
6.23 (Turtle Graphics) The Logo language made the concept of turtle graphics famous. Imagine
a mechanical turtle that walks around the room under the control of a C program. The turtle holds
a pen in one of two positions, up or down. While the pen is down, the turtle traces out shapes as it
moves; while the pen is up, the turtle moves about freely without writing anything. In this problem
you’ll simulate the operation of the turtle and create a computerized sketchpad as well.
Use a 50-by-50 array floor which is initialized to zeros. Read commands from an array that
contains them. Keep track of the current turtle position at all times and whether the pen is currently up or down. Assume that the turtle always starts at position 0, 0 of the floor with its pen up.
The set of turtle commands your program must process are shown in Fig. 6.25. Suppose that the
turtle is somewhere near the center of the floor. The following “program” would draw and print a
12-by-12 square:
2
5,12
3
5,12
3
5,12
3
5,12
1
6
9
As the turtle moves with the pen down, set the appropriate elements of array floor to 1s. When the
6 command (print) is given, wherever there’s a 1 in the array, display an asterisk, or some other
character you choose. Wherever there’s a zero, display a blank. Write a program to implement the
turtle graphics capabilities discussed here. Write several turtle graphics programs to draw interesting shapes. Add other commands to increase the power of your turtle graphics language.
Command
Meaning
1
Pen up
Pen down
Turn right
2
3
Fig. 6.25 | Turtle commands. (Part 1 of 2.)
272
Chapter 6 C Arrays
Command
Meaning
4
Turn left
Move forward 10 spaces (or a number other than 10)
Print the 50-by-50 array
End of data (sentinel)
5, 10
6
9
Fig. 6.25 | Turtle commands. (Part 2 of 2.)
6.24 (Knight’s Tour) One of the more interesting puzzlers for chess buffs is the Knight’s Tour
problem, originally proposed by the mathematician Euler. The question is this: Can the chess piece
called the knight move around an empty chessboard and touch each of the 64 squares once and only
once? We study this intriguing problem in depth here.
The knight makes L-shaped moves (over two in one direction and then over one in a perpendicular direction). Thus, from a square in the middle of an empty chessboard, the knight can
make eight different moves (numbered 0 through 7) as shown in Fig. 6.26.
0
1
2
3
4
5
6
7
0
1
2
2
0
3
4
5
1
3
K
4
7
5
6
6
7
Fig. 6.26 | The eight possible moves of the knight.
a) Draw an 8-by-8 chessboard on a sheet of paper and attempt a Knight’s Tour by hand.
Put a 1 in the first square you move to, a 2 in the second square, a 3 in the third, and so
on. Before starting the tour, estimate how far you think you’ll get, remembering that a
full tour consists of 64 moves. How far did you get? Were you close to the estimate?
b) Now let’s develop a program that will move the knight around a chessboard. The board
itself is represented by an 8-by-8 double-subscripted array board. Each square is initialized to zero. We describe each of the eight possible moves in terms of both its horizontal
and vertical components. For example, a move of type 0 as shown in Fig. 6.26 consists
of moving two squares horizontally to the right and one square vertically upward. Move
2 consists of moving one square horizontally to the left and two squares vertically upward. Horizontal moves to the left and vertical moves upward are indicated with negative numbers. The eight moves may be described by two single-subscripted arrays,
horizontal and vertical, as follows:
Exercises
horizontal[
horizontal[
horizontal[
horizontal[
horizontal[
horizontal[
horizontal[
horizontal[
0
1
2
3
4
5
6
7
]
]
]
]
]
]
]
]
=
=
=
=
=
=
=
=
vertical[
vertical[
vertical[
vertical[
vertical[
vertical[
vertical[
vertical[
]
]
]
]
]
]
]
]
=
=
=
=
=
=
=
=
-1
-2
-2
-1
1
2
2
1
0
1
2
3
4
5
6
7
273
2
1
-1
-2
-2
-1
1
2
Let the variables currentRow and currentColumn indicate the row and column of the
knight’s current position on the board. To make a move of type moveNumber, where
moveNumber is between 0 and 7, your program uses the statements
currentRow += vertical[ moveNumber ];
currentColumn += horizontal[ moveNumber ];
Keep a counter that varies from 1 to 64. Record the latest count in each square the
knight moves to. Remember to test each potential move to see if the knight has already
visited that square. And, of course, test every potential move to make sure that the
knight does not land off the chessboard. Now write a program to move the knight
around the chessboard. Run the program. How many moves did the knight make?
c) After attempting to write and run a Knight’s Tour program, you have probably developed some valuable insights. We’ll use these to develop a heuristic (or strategy) for moving the knight. Heuristics do not guarantee success, but a carefully developed heuristic
greatly improves the chance of success. You may have observed that the outer squares
are in some sense more troublesome than the squares nearer the center of the board. In
fact, the most troublesome, or inaccessible, squares are the four corners.
Intuition may suggest that you should attempt to move the knight to the most troublesome squares first and leave open those that are easiest to get to, so that when the
board gets congested near the end of the tour, there will be a greater chance of success.
We develop an “accessibility heuristic” by classifying each square according to how
accessible it is and always moving the knight to the square (within the knight’s L-shaped
moves, of course) that’s most inaccessible. We label a double-subscripted array accessibility with numbers indicating from how many squares each particular square is accessible. On a blank chessboard, the center squares are therefore rated as 8s, the corner squares
are rated as 2s, and the other squares have accessibility numbers of 3, 4, or 6 as follows:
2
3
4
4
4
4
3
2
3
4
6
6
6
6
4
3
4
6
8
8
8
8
6
4
4
6
8
8
8
8
6
4
4
6
8
8
8
8
6
4
4
6
8
8
8
8
6
4
3
4
6
6
6
6
4
3
2
3
4
4
4
4
3
2
Now write a version of the Knight’s Tour program using the accessibility heuristic.
At any time, the knight should move to the square with the lowest accessibility number. In case of a tie, the knight may move to any of the tied squares. Therefore, the tour
may begin in any of the four corners. [Note: As the knight moves around the chess-
274
Chapter 6 C Arrays
board, your program should reduce the accessibility numbers as more and more
squares become occupied. In this way, at any given time during the tour, each available
square’s accessibility number will remain equal to precisely the number of squares from
which that square may be reached.] Run this version of your program. Did you get a
full tour? (Optional: Modify the program to run 64 tours, one from each square of the
chessboard. How many full tours did you get?)
d) Write a version of the Knight’s Tour program which, when encountering a tie between
two or more squares, decides what square to choose by looking ahead to those squares
reachable from the “tied” squares. Your program should move to the square for which
the next move would arrive at a square with the lowest accessibility number.
6.25 (Knight’s Tour: Brute-Force Approaches) In Exercise 6.24 we developed a solution to the
Knight’s Tour problem. The approach used, called the “accessibility heuristic,” generates many solutions and executes efficiently.
As computers continue increasing in power, we’ll be able to solve many problems with sheer
computer power and relatively unsophisticated algorithms. Let’s call this approach brute-force
problem solving.
a) Use random number generation to enable the knight to walk around the chess board
(in its legitimate L-shaped moves, of course) at random. Your program should run one
tour and print the final chessboard. How far did the knight get?
b) Most likely, the preceding program produced a relatively short tour. Now modify your
program to attempt 1,000 tours. Use a single-subscripted array to keep track of the
number of tours of each length. When your program finishes attempting the 1000
tours, it should print this information in a tabular format. What was the best result?
c) Most likely, the preceding program gave you some “respectable” tours but no full tours.
Now “pull all the stops out” and simply let your program run until it produces a full
tour. [Caution: This version of the program could run for hours on a powerful computer.] Once again, keep a table of the number of tours of each length and print this table
when the first full tour is found. How many tours did your program attempt before producing a full tour? How much time did it take?
d) Compare the brute-force version of the Knight’s Tour with the accessibility-heuristic
version. Which required a more careful study of the problem? Which algorithm was
more difficult to develop? Which required more computer power? Could we be certain
(in advance) of obtaining a full tour with the accessibility-heuristic approach? Could we
be certain (in advance) of obtaining a full tour with the brute-force approach? Argue the
pros and cons of brute-force problem solving in general.
6.26 (Eight Queens) Another puzzler for chess buffs is the Eight Queens problem. Simply stated:
Is it possible to place eight queens on an empty chessboard so that no queen is “attacking” any other—that is, so that no two queens are in the same row, the same column, or along the same diagonal?
Use the kind of thinking developed in Exercise 6.24 to formulate a heuristic for solving the Eight
Queens problem. Run your program. [Hint: It’s possible to assign a numeric value to each square
of the chessboard indicating how many squares of an empty chessboard are “eliminated” once a
queen is placed in that square. For example, each of the four corners would be assigned the value
22, as in Fig. 6.27.]
Once these “elimination numbers” are placed in all 64 squares, an appropriate heuristic might
be: Place the next queen in the square with the smallest elimination number. Why is this strategy
intuitively appealing?
6.27 (Eight Queens: Brute-Force Approaches) In this problem you’ll develop several brute-force
approaches to solving the Eight Queens problem introduced in Exercise 6.26.
a) Solve the Eight Queens problem, using the random brute-force technique developed in
Exercise 6.25.
Recursion Exercises
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
275
*
*
*
*
*
*
*
Fig. 6.27 | The 22 squares eliminated by placing a queen in the upper-left corner.
b) Use an exhaustive technique (i.e., try all possible combinations of eight queens on the
chessboard).
c) Why do you suppose the exhaustive brute-force approach may not be appropriate for
solving the Eight Queens problem?
d) Compare and contrast the random brute-force and exhaustive brute-force approaches
in general.
6.28 (Duplicate Elimination) In Chapter 12, we explore the high-speed binary search tree data
structure. One feature of a binary search tree is that duplicate values are discarded when insertions
are made into the tree. This is referred to as duplicate elimination. Write a program that produces
20 random numbers between 1 and 20. The program should store all nonduplicate values in an array. Use the smallest possible array to accomplish this task.
6.29 (Knight’s Tour: Closed Tour Test) In the Knight’s Tour, a full tour occurs when the knight
makes 64 moves touching each square of the chessboard once and only once. A closed tour occurs
when the 64th move is one move away from the location in which the knight started the tour. Modify the Knight’s Tour program you wrote in Exercise 6.24 to test for a closed tour if a full tour has
occurred.
6.30 (The Sieve of Eratosthenes) A prime integer is any integer greater than 1 that can be divided
evenly only by itself and 1. The Sieve of Eratosthenes is a method of finding prime numbers. It
works as follows:
a) Create an array with all elements initialized to 1 (true). Array elements with prime subscripts will remain 1. All other array elements will eventually be set to zero.
b) Starting with array subscript 2 (subscript 1 is not prime), every time an array element is
found whose value is 1, loop through the remainder of the array and set to zero every
element whose subscript is a multiple of the subscript for the element with value 1. For
array subscript 2, all elements beyond 2 in the array that are multiples of 2 will be set to
zero (subscripts 4, 6, 8, 10, and so on.). For array subscript 3, all elements beyond 3 in
the array that are multiples of 3 will be set to zero (subscripts 6, 9, 12, 15, and so on.).
When this process is complete, the array elements that are still set to 1 indicate that the subscript is
a prime number. Write a program that uses an array of 1,000 elements to determine and print the
prime numbers between 1 and 999. Ignore element 0 of the array.
Recursion Exercises
6.31 (Palindromes) A palindrome is a string that’s spelled the same way forward and backward.
Some examples of palindromes are: “radar,” “able was i ere i saw elba,” and, if you ignore blanks, “a
man a plan a canal panama.” Write a recursive function testPalindrome that returns 1 if the string
276
Chapter 6 C Arrays
stored in the array is a palindrome and 0 otherwise. The function should ignore spaces and punctuation in the string.
6.32 (Linear Search) Modify the program of Fig. 6.18 to use a recursive linearSearch function
to perform the linear search of the array. The function should receive an integer array, the size of
the array and the search key as arguments. If the search key is found, return the array subscript; otherwise, return –1.
6.33 (Binary Search) Modify the program of Fig. 6.19 to use a recursive binarySearch function
to perform the binary search of the array. The function should receive an integer array, the starting
subscript, the ending subscript and the search key as arguments. If the search key is found, return
the array subscript; otherwise, return –1.
6.34 (Eight Queens) Modify the Eight Queens program you created in Exercise 6.26 to solve the
problem recursively.
6.35 (Print an Array) Write a recursive function printArray that takes an array and the size of
the array as arguments, prints the array, and returns nothing. The function should stop processing
and return when it receives an array of size zero.
6.36 (Print a String Backward) Write a recursive function stringReverse that takes a character
array as an argument, prints it back to front and returns nothing. The function should stop processing and return when the terminating null character of the string is encountered.
6.37 (Find the Minimum Value in an Array) Write a recursive function recursiveMinimum that
takes an integer array and the array size as arguments and returns the smallest element of the array.
The function should stop processing and return when it receives an array of one element.
6.38 (Project: Sudoku Puzzles) In Appendix D, Game Programming: Solving Sudoku, we’ll discuss various simple solution strategies and suggest what to do when these fail. We’ll also present various approaches for programming Sudoku puzzle creators and solvers in C. Unfortunately, Standard
C does not include graphics and GUI (graphical user interface) capabilities, so our representation
of the board won’t be as elegant as we could make it in Java and other programming languages that
support these capabilities. You may want to revisit your Sudoku programs after you study a game
programming library such as Allegro, which offers capabilities that will help you add graphics and
even sounds to your Sudoku programs.
Be sure to check out our Sudoku Resource Center at www.deitel.com/sudoku for downloads,
tutorials, books, e-books and more that will help you master the game. And—not for the faint of
heart—try fiendishly difficult Sudokus with tricky twists, a circular Sudoku and a variant of the
puzzle with five interlocking grids.
7
C Pointers
Addresses are given to us to
conceal our whereabouts.
—Saki (H. H. Munro)
By indirection find direction
out.
—William Shakespeare
Many things,
having full reference
To one consent,
may work contrariously.
—William Shakespeare
You will find it a very good
practice always to verify your
references, sir!
—Dr. Routh
Objectives
In this chapter, you’ll learn:
■
Pointers and pointer
operators.
■
To use pointers to pass
arguments to functions by
reference.
■
The close relationships
among pointers, arrays and
strings.
■
To use pointers to functions.
■
To define and use arrays of
strings.
278
Chapter 7 C Pointers
7.1 Introduction
7.2 Pointer Variable Definitions and
Initialization
7.3 Pointer Operators
7.4 Passing Arguments to Functions by
Reference
7.5 Using the const Qualifier with
Pointers
7.5.1 Converting a String to Uppercase
Using a Non-Constant Pointer to
Non-Constant Data
7.5.2 Printing a String One Character at a
Time Using a Non-Constant Pointer
to Constant Data
7.5.3 Attempting to Modify a Constant
Pointer to Non-Constant Data
7.5.4 Attempting to Modify a Constant
Pointer to Constant Data
7.6 Bubble Sort Using Pass-by-Reference
7.7 sizeof Operator
7.8 Pointer Expressions and Pointer
Arithmetic
7.9 Relationship between Pointers and
Arrays
7.10 Arrays of Pointers
7.11 Case Study: Card Shuffling and
Dealing Simulation
7.12 Pointers to Functions
7.13 Secure C Programming
Summary | Terminology | Self-Review Exercises | Answers to Self-Review Exercises | Exercises
Special Section: Building Your Own Computer | Array of Function Pointer Exercises |
Making a Difference
7.1 Introduction
In this chapter, we discuss one of the most powerful features of the C programming language, the pointer.1 Pointers are among C’s most difficult capabilities to master. Pointers
enable programs to simulate pass-by-reference, to pass functions between functions, and
to create and manipulate dynamic data structures, i.e., data structures that can grow and
shrink at execution time, such as linked lists, queues, stacks and trees. This chapter explains basic pointer concepts. Chapter 10 examines the use of pointers with structures.
Chapter 12 introduces dynamic memory management techniques and presents examples
of creating and using dynamic data structures.
7.2 Pointer Variable Definitions and Initialization
Pointers are variables whose values are memory addresses. Normally, a variable directly contains a specific value. A pointer, on the other hand, contains an address of a variable that
contains a specific value. In this sense, a variable name directly references a value, and a
pointer indirectly references a value (Fig. 7.1). Referencing a value through a pointer is
called indirection.
Declaring Pointers
Pointers, like all variables, must be defined before they can be used. The definition
int *countPtr, count;
1.
Pointers and pointer-based entities such as arrays and strings, when misused intentionally or accidentally, can lead to errors and security breaches. See our Secure C Programming Resource Center
(www.deitel.com/SecureC/) for articles, books, white papers and forums on this important topic.
7.3 Pointer Operators
279
count
7
countPtr
count
7
The name count directly references
a variable that contains the value 7
The pointer countPtr indirectly
references a variable that
contains the value 7
Fig. 7.1 | Directly and indirectly referencing a variable.
specifies that variable countPtr is of type int * (i.e., a pointer to an integer) and is read
(right to left), “countPtr is a pointer to int” or “countPtr points to an object of type int.”
Also, the variable count is defined to be an int, not a pointer to an int. The * applies only
to countPtr in the definition. When * is used in this manner in a definition, it indicates
that the variable being defined is a pointer. Pointers can be defined to point to objects of
any type. To prevent the ambiguity of declaring pointer and non-pointer variables in the
same declaration as shown above, you should always declare only one variable per declaration.
Common Programming Error 7.1
The asterisk (*) notation used to declare pointer variables does not distribute to all variable names in a declaration. Each pointer must be declared with the * prefixed to the
name; e.g., if you wish to declare xPtr and yPtr as int pointers, use int *xPtr, *yPtr;.
Good Programming Practice 7.1
We prefer to include the letters Ptr in pointer variable names to make it clear that these
variables are pointers and thus need to be handled appropriately.
Initializing and Assigning Values to Pointers
Pointers should be initialized when they’re defined, or they can be assigned a value. A
pointer may be initialized to NULL, 0 or an address. A pointer with the value NULL points
to nothing. NULL is a symbolic constant defined in the <stddef.h> header (and several other
headers, such as <stdio.h>). Initializing a pointer to 0 is equivalent to initializing a pointer
to NULL, but NULL is preferred. When 0 is assigned, it’s first converted to a pointer of the
appropriate type. The value 0 is the only integer value that can be assigned directly to a
pointer variable. Assigning a variable’s address to a pointer is discussed in Section 7.3.
Error-Prevention Tip 7.1
Initialize pointers to prevent unexpected results.
7.3 Pointer Operators
The &, or address operator, is a unary operator that returns the address of its operand. For
example, assuming the definitions
int y = 5;
int *yPtr;
280
Chapter 7 C Pointers
the statement
yPtr = &y;
assigns the address of the variable y to pointer variable yPtr. Variable yPtr is then said to
“point to” y. Figure 7.2 shows a schematic representation of memory after the preceding
assignment is executed.
yPtr
y
5
Fig. 7.2 | Graphical representation of a pointer pointing to an integer variable in memory.
Pointer Representation in Memory
Figure 7.3 shows the representation of the pointer in memory, assuming that integer variable y is stored at location 600000, and pointer variable yPtr is stored at location 500000.
The operand of the address operator must be a variable; the address operator cannot be applied to constants or expressions.
yPtr
location
500000
600000
y
location
600000
5
Fig. 7.3 | Representation of y and yPtr in memory.
The Indirection (*) Operator
The unary * operator, commonly referred to as the indirection operator or dereferencing
operator, returns the value of the object to which its operand (i.e., a pointer) points. For
example, the statement
printf( "%d", *yPtr );
prints the value of variable y, namely 5. Using * in this manner is called dereferencing a
pointer.
Common Programming Error 7.2
Dereferencing a pointer that has not been properly initialized or that has not been assigned
to point to a specific location in memory is an error. This could cause a fatal executiontime error, or it could accidentally modify important data and allow the program to run
to completion with incorrect results.
Demonstrating the & and * Operators
Figure 7.4 demonstrates the pointer operators & and *. The printf conversion specifier %p
outputs the memory location as a hexadecimal integer on most platforms. (See Appendix C
for more information on hexadecimal integers.) Notice that the address of a and the value
of aPtr are identical in the output, thus confirming that the address of a is indeed assigned
to the pointer variable aPtr (line 11). The & and * operators are complements of one another—when they’re both applied consecutively to aPtr in either order (line 21), the same
7.3 Pointer Operators
281
result is printed. Figure 7.5 lists the precedence and associativity of the operators introduced to this point.
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
// Fig. 7.4: fig07_04.c
// Using the & and * pointer operators.
#include <stdio.h>
int main( void )
{
int a; // a is an integer
int *aPtr; // aPtr is a pointer to an integer
a = 7;
aPtr = &a; // set aPtr to the address of a
printf( "The address of a is %p"
"\nThe value of aPtr is %p", &a, aPtr );
printf( "\n\nThe value of a is %d"
"\nThe value of *aPtr is %d", a, *aPtr );
printf( "\n\nShowing that * and & are complements of "
"each other\n&*aPtr = %p"
"\n*&aPtr = %p\n", &*aPtr, *&aPtr );
} // end main
The address of a is 0028FEC0
The value of aPtr is 0028FEC0
The value of a is 7
The value of *aPtr is 7
Showing that * and & are complements of each other
&*aPtr = 0028FEC0
*&aPtr = 0028FEC0
Fig. 7.4 | Using the & and * pointer operators.
Operators
()
[]
++
+
-
++
*
/
%
+
-
<
<=
==
!=
&&
||
>
(postfix)
--
s>=
!
-*
(postfix)
&
(type)
Associativity
Type
left to right
right to left
left to right
left to right
left to right
left to right
left to right
left to right
postfix
unary
multiplicative
additive
relational
equality
logical AND
logical OR
Fig. 7.5 | Operator precedence and associativity. (Part 1 of 2.)
282
Chapter 7 C Pointers
Operators
?:
=
+=
-=
*=
/=
%=
,
Associativity
Type
right to left
right to left
left to right
conditional
assignment
comma
Fig. 7.5 | Operator precedence and associativity. (Part 2 of 2.)
7.4 Passing Arguments to Functions by Reference
There are two ways to pass arguments to a function—pass-by-value and pass-by-reference. All arguments in C are passed by value. As we saw in Chapter 5, return may be used
to return one value from a called function to a caller (or to return control from a called
function without passing back a value). Many functions require the capability to modify
variables in the caller or to pass a pointer to a large data object to avoid the overhead of
passing the object by value (which incurs the time and memory overheads of making a
copy of the object).
In C, you use pointers and the indirection operator to simulate pass-by-reference.
When calling a function with arguments that should be modified, the addresses of the arguments are passed. This is normally accomplished by applying the address operator (&) to
the variable (in the caller) whose value will be modified. As we saw in Chapter 6, arrays are
not passed using operator & because C automatically passes the starting location in memory
of the array (the name of an array is equivalent to &arrayName[0]). When the address of
a variable is passed to a function, the indirection operator (*) may be used in the function
to modify the value at that location in the caller’s memory.
Pass-By-Value
The programs in Figs. 7.6 and 7.7 present two versions of a function that cubes an integer—cubeByValue and cubeByReference. Figure 7.6 passes the variable number by value
to function cubeByValue (line 14). The cubeByValue function cubes its argument and
passes the new value back to main using a return statement. The new value is assigned to
number in main (line 14).
1
2
3
4
5
6
7
8
9
10
11
12
// Fig. 7.6: fig07_06.c
// Cube a variable using pass-by-value.
#include <stdio.h>
int cubeByValue( int n ); // prototype
int main( void )
{
int number = 5; // initialize number
printf( "The original value of number is %d", number );
Fig. 7.6 | Cube a variable using pass-by-value. (Part 1 of 2.)
7.4 Passing Arguments to Functions by Reference
13
14
15
16
17
18
19
20
21
22
23
283
// pass number by value to cubeByValue
number = cubeByValue( number );
printf( "\nThe new value of number is %d\n", number );
} // end main
// calculate and return cube of integer argument
int cubeByValue( int n )
{
return n * n * n; // cube local variable n and return result
} // end function cubeByValue
The original value of number is 5
The new value of number is 125
Fig. 7.6 | Cube a variable using pass-by-value. (Part 2 of 2.)
Pass-By-Reference
Figure 7.7 passes the variable number by reference (line 15)—the address of number is
passed—to function cubeByReference. Function cubeByReference takes as a parameter
a pointer to an int called nPtr (line 21). The function dereferences the pointer and cubes
the value to which nPtr points (line 23), then assigns the result to *nPtr (which is really
number in main), thus changing the value of number in main. Figures 7.8 and 7.9 analyze
graphically and step-by-step the programs in Figs. 7.6 and 7.7, respectively.
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
// Fig. 7.7: fig07_07.c
// Cube a variable using pass-by-reference with a pointer argument.
#include <stdio.h>
void cubeByReference( int *nPtr ); // function prototype
int main( void )
{
int number = 5; // initialize number
printf( "The original value of number is %d", number );
// pass address of number to cubeByReference
cubeByReference( &number );
printf( "\nThe new value of number is %d\n", number );
} // end main
// calculate cube of *nPtr; actually modifies number in main
void cubeByReference( int *nPtr )
{
*nPtr = *nPtr * *nPtr * *nPtr; // cube *nPtr
} // end function cubeByReference
Fig. 7.7 | Cube a variable using pass-by-reference with a pointer argument. (Part 1 of 2.)
284
Chapter 7 C Pointers
The original value of number is 5
The new value of number is 125
Fig. 7.7 | Cube a variable using pass-by-reference with a pointer argument. (Part 2 of 2.)
A function receiving an address as an argument must define a pointer parameter to receive
the address. For example, in Fig. 7.7 the header for function cubeByReference (line 21) is:
void cubeByReference( int *nPtr )
The header specifies that cubeByReference receives the address of an integer variable as an
argument, stores the address locally in nPtr and does not return a value.
The function prototype for cubeByReference (line 6) contains int * in parentheses.
As with other variable types, it’s not necessary to include names of pointers in function prototypes. Names included for documentation purposes are ignored by the C compiler.
For a function that expects a single-subscripted array as an argument, the function’s
prototype and header can use the pointer notation shown in the parameter list of function
cubeByReference (line 21). The compiler does not differentiate between a function that
receives a pointer and one that receives a single-subscripted array. This, of course, means
that the function must “know” when it’s receiving an array or simply a single variable for
which it’s to perform pass-by-reference. When the compiler encounters a function parameter for a single-subscripted array of the form int b[], the compiler converts the parameter to the pointer notation int *b. The two forms are interchangeable.
Error-Prevention Tip 7.2
Use pass-by-value to pass arguments to a function unless the caller explicitly requires the
called function to modify the value of the argument variable in the caller’s environment.
This prevents accidental modification of the caller’s arguments and is another example of
the principle of least privilege.
7.5 Using the const Qualifier with Pointers
The const qualifier enables you to inform the compiler that the value of a particular variable should not be modified.
Software Engineering Observation 7.1
The const qualifier can be used to enforce the principle of least privilege in software
design. This can reduce debugging time and improper side effects, making a program
easier to modify and maintain.
Over the years, a large base of legacy code was written in early versions of C that did
not use const because it was not available. For this reason, there are significant opportunities for improvement by reengineering old C code.
Six possibilities exist for using (or not using) const with function parameters—two
with pass-by-value parameter passing and four with pass-by-reference parameter passing.
How do you choose one of the six possibilities? Let the principle of least privilege be your
guide. Always award a function enough access to the data in its parameters to accomplish
its specified task, but absolutely no more.
7.5 Using the const Qualifier with Pointers
285
Step 1: Before main calls cubeByValue:
int main( void )
{
int number = 5;
number
5
int cubeByValue( int n )
{
return n * n * n;
}
n
number = cubeByValue( number );
undefined
}
Step 2: After cubeByValue receives the call:
number
int main( void )
{
int number = 5;
5
int cubeByValue( int n )
{
return n * n * n;
}
n
number = cubeByValue( number );
}
5
Step 3: After cubeByValue cubes parameter n and before cubeByValue returns to main:
number
int main( void )
{
int number = 5;
5
int cubeByValue( int n )
{
125
return n * n * n;
n
}
number = cubeByValue( number );
}
5
Step 4: After cubeByValue returns to main and before assigning the result to number:
number
int main( void )
{
int number = 5;
5
125
int cubeByValue( int n )
{
return n * n * n;
}
number = cubeByValue( number );
n
undefined
}
Step 5: After main completes the assignment to number:
number
int main( void )
{
int number = 5;
125
125
125
number = cubeByValue( number );
}
Fig. 7.8 | Analysis of a typical pass-by-value.
int cubeByValue( int n )
{
return n * n * n;
}
n
undefined
286
Chapter 7 C Pointers
Step 1: Before main calls cubeByReference:
int main( void )
{
int number = 5;
number
5
cubeByReference( &number );
void cubeByReference( int *nPtr )
{
*nPtr = *nPtr * *nPtr * *nPtr;
}
nPtr
undefined
}
Step 2: After cubeByReference receives the call and before *nPtr is cubed:
int main( void )
{
int number = 5;
number
5
cubeByReference( &number );
void cubeByReference( int *nPtr )
{
*nPtr = *nPtr * *nPtr * *nPtr;
}
nPtr
call establishes this pointer
}
Step 3: After *nPtr is cubed and before program control returns to main:
int main( void )
{
int number = 5;
number
void cubeByReference( int *nPtr )
{
125
125
cubeByReference( &number );
}
*nPtr = *nPtr * *nPtr * *nPtr;
}
called function modifies caller’s
variable
nPtr
Fig. 7.9 | Analysis of a typical pass-by-reference with a pointer argument.
In Chapter 5, we explained that all function calls in C are pass-by-value—a copy of the
argument in the function call is made and passed to the function. If the copy is modified
in the function, the original value in the caller does not change. In many cases, a value
passed to a function is modified so the function can accomplish its task. However, in some
instances, the value should not be altered in the called function, even though it manipulates
only a copy of the original value.
Consider a function that takes a single-subscripted array and its size as arguments and
prints the array. Such a function should loop through the array and output each array element individually. The size of the array is used in the function body to determine the high
subscript of the array, so the loop can terminate when the printing is completed. Neither
the size of the array nor its contents should change in the function body.
Error-Prevention Tip 7.3
If a variable does not (or should not) change in the body of a function to which it’s passed,
the variable should be declared const to ensure that it’s not accidentally modified.
If an attempt is made to modify a value that’s declared const, the compiler catches it
and issues either a warning or an error, depending on the particular compiler.
7.5 Using the const Qualifier with Pointers
287
Common Programming Error 7.3
Being unaware that a function is expecting pointers as arguments for pass-by-reference
and passing arguments by value. Some compilers take the values assuming they’re pointers
and dereference the values as pointers. At runtime, memory-access violations or segmentation faults are often generated. Other compilers catch the mismatch in types between
arguments and parameters and generate error messages.
There are four ways to pass a pointer to a function: a non-constant pointer to nonconstant data, a constant pointer to nonconstant data, a non-constant pointer to constant data, and a constant pointer to constant data. Each of the four combinations provides different access privileges. These are discussed in the next several examples.
7.5.1 Converting a String to Uppercase Using a Non-Constant Pointer
to Non-Constant Data
The highest level of data access is granted by a non-constant pointer to non-constant data.
In this case, the data can be modified through the dereferenced pointer, and the pointer can
be modified to point to other data items. A declaration for a non-constant pointer to nonconstant data does not include const. Such a pointer might be used to receive a string as an
argument to a function that processes (and possibly modifies) each character in the string.
Function convertToUppercase of Fig. 7.10 declares its parameter, a non-constant pointer to
non-constant data called sPtr (char *sPtr), in line 19. The function processes the array
string (pointed to by sPtr) one character at a time. C standard library function toupper
(line 22) from the <ctype.h> header is called to convert each character to its corresponding
uppercase letter—if the original character is not a letter or is already uppercase, toupper returns the original character. Line 23 moves the pointer to the next character in the string.
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
// Fig. 7.10: fig07_10.c
// Converting a string to uppercase using a
// non-constant pointer to non-constant data.
#include <stdio.h>
#include <ctype.h>
void convertToUppercase( char *sPtr ); // prototype
int main( void )
{
char string[] = "cHaRaCters and $32.98"; // initialize char array
printf( "The string before conversion is: %s", string );
convertToUppercase( string );
printf( "\nThe string after conversion is: %s\n", string );
} // end main
// convert string to uppercase letters
void convertToUppercase( char *sPtr )
{
Fig. 7.10 | Converting a string to uppercase using a non-constant pointer to non-constant data.
(Part 1 of 2.)
288
21
22
23
24
25
Chapter 7 C Pointers
while ( *sPtr != '\0' ) { // current character is not '\0'
*sPtr = toupper( *sPtr ); // convert to uppercase
++sPtr; // make sPtr point to the next character
} // end while
} // end function convertToUppercase
The string before conversion is: cHaRaCters and $32.98
The string after conversion is: CHARACTERS AND $32.98
Fig. 7.10 | Converting a string to uppercase using a non-constant pointer to non-constant data.
(Part 2 of 2.)
7.5.2 Printing a String One Character at a Time Using a Non-Constant
Pointer to Constant Data
A non-constant pointer to constant data can be modified to point to any data item of the
appropriate type, but the data to which it points cannot be modified. Such a pointer might
be used to receive an array argument to a function that will process each element without
modifying the data. For example, function printCharacters (Fig. 7.11) declares parameter sPtr to be of type const char * (line 21). The declaration is read from right to left as
“sPtr is a pointer to a character constant.” The function uses a for statement to output
each character in the string until the null character is encountered. After each character is
printed, pointer sPtr is incremented to point to the next character in the string.
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
// Fig. 7.11: fig07_11.c
// Printing a string one character at a time using
// a non-constant pointer to constant data.
#include <stdio.h>
void printCharacters( const char *sPtr );
int main( void )
{
// initialize char array
char string[] = "print characters of a string";
puts( "The string is:" );
printCharacters( string );
puts( "" );
} // end main
// sPtr cannot modify the character to which it points,
// i.e., sPtr is a "read-only" pointer
void printCharacters( const char *sPtr )
{
Fig. 7.11 | Printing a string one character at a time using a non-constant pointer to constant
data. (Part 1 of 2.)
7.5 Using the const Qualifier with Pointers
23
24
25
26
27
289
// loop through entire string
for ( ; *sPtr != '\0'; ++sPtr ) { // no initialization
printf( "%c", *sPtr );
} // end for
} // end function printCharacters
The string is:
print characters of a string
Fig. 7.11 | Printing a string one character at a time using a non-constant pointer to constant
data. (Part 2 of 2.)
Figure 7.12 illustrates the attempt to compile a function that receives a non-constant
pointer (xPtr) to constant data. This function attempts to modify the data pointed to by
xPtr in line 18—which results in a compilation error. The actual error message you see
will be compiler specific.
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
// Fig. 7.12: fig07_12.c
// Attempting to modify data through a
// non-constant pointer to constant data.
#include <stdio.h>
void f( const int *xPtr ); // prototype
int main( void )
{
int y; // define y
f( &y ); // f attempts illegal modification
} // end main
// xPtr cannot be used to modify the
// value of the variable to which it points
void f( const int *xPtr )
{
*xPtr = 100; // error: cannot modify a const object
} // end function f
c:\examples\ch07\fig07_12.c(18) : error C2166: l-value specifies const object
Fig. 7.12 | Attempting to modify data through a non-constant pointer to constant data.
As you know, arrays are aggregate data types that store related data items of the same
type under one name. In Chapter 10, we’ll discuss another form of aggregate data type
called a structure (sometimes called a record in other languages). A structure is capable of
storing related data items of different data types under one name (e.g., storing information
about each employee of a company). When a function is called with an array as an argument, the array is automatically passed to the function by reference. However, structures
are always passed by value—a copy of the entire structure is passed. This requires the execution-time overhead of making a copy of each data item in the structure and storing it on
290
Chapter 7 C Pointers
the computer’s function call stack. When structure data must be passed to a function, we
can use pointers to constant data to get the performance of pass-by-reference and the protection of pass-by-value. When a pointer to a structure is passed, only a copy of the
address at which the structure is stored must be made. On a machine with four-byte
addresses, a copy of four bytes of memory is made rather than a copy of a possibly large
structure.
Performance Tip 7.1
Pass large objects such as structures using pointers to constant data to obtain the performance benefits of pass-by-reference and the security of pass-by-value.
If memory is low and execution efficiency is a concern, use pointers. If memory is in
abundance and efficiency is not a major concern, pass data by value to enforce the principle of least privilege. Remember that some systems do not enforce const well, so passby-value is still the best way to prevent data from being modified.
7.5.3 Attempting to Modify a Constant Pointer to Non-Constant Data
A constant pointer to non-constant data always points to the same memory location, and
the data at that location can be modified through the pointer. This is the default for an array
name. An array name is a constant pointer to the beginning of the array. All data in the
array can be accessed and changed by using the array name and array subscripting. A constant pointer to non-constant data can be used to receive an array as an argument to a function that accesses array elements using only array subscript notation. Pointers that are
declared const must be initialized when they’re defined (if the pointer is a function parameter, it’s initialized with a pointer that’s passed to the function). Figure 7.13 attempts
to modify a constant pointer. Pointer ptr is defined in line 12 to be of type int * const.
The definition is read from right to left as “ptr is a constant pointer to an integer.” The
pointer is initialized (line 12) with the address of integer variable x. The program attempts
to assign the address of y to ptr (line 15), but the compiler generates an error message.
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
// Fig. 7.13: fig07_13.c
// Attempting to modify a constant pointer to non-constant data.
#include <stdio.h>
int main( void )
{
int x; // define x
int y; // define y
// ptr is a constant pointer to an integer that can be modified
// through ptr, but ptr always points to the same memory location
int * const ptr = &x;
*ptr = 7; // allowed: *ptr is not const
ptr = &y; // error: ptr is const; cannot assign new address
} // end main
Fig. 7.13 | Attempting to modify a constant pointer to non-constant data. (Part 1 of 2.)
7.6 Bubble Sort Using Pass-by-Reference
291
c:\examples\ch07\fig07_13.c(15) : error C2166: l-value specifies const object
Fig. 7.13 | Attempting to modify a constant pointer to non-constant data. (Part 2 of 2.)
7.5.4 Attempting to Modify a Constant Pointer to Constant Data
The least access privilege is granted by a constant pointer to constant data. Such a pointer
always points to the same memory location, and the data at that memory location cannot
be modified. This is how an array should be passed to a function that only looks at the array
using array subscript notation and does not modify the array. Figure 7.14 defines pointer
variable ptr (line 13) to be of type const int *const, which is read from right to left as
“ptr is a constant pointer to an integer constant.” The figure shows the error messages generated when an attempt is made to modify the data to which ptr points (line 16) and when
an attempt is made to modify the address stored in the pointer variable (line 17).
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
// Fig. 7.14: fig07_14.c
// Attempting to modify a constant pointer to constant data.
#include <stdio.h>
int main( void )
{
int x = 5; // initialize x
int y; // define y
// ptr is
// points
// cannot
const int
a constant pointer to a constant integer. ptr always
to the same location; the integer at that location
be modified
*const ptr = &x; // initialization is OK
printf( "%d\n", *ptr );
*ptr = 7; // error: *ptr is const; cannot assign new value
ptr = &y; // error: ptr is const; cannot assign new address
} // end main
c:\examples\ch07\fig07_14.c(16) : error C2166: l-value specifies const object
c:\examples\ch07\fig07_14.c(17) : error C2166: l-value specifies const object
Fig. 7.14 | Attempting to modify a constant pointer to constant data.
7.6 Bubble Sort Using Pass-by-Reference
Let’s improve the bubble sort program of Fig. 6.15 to use two functions—bubbleSort and
swap. Function bubbleSort sorts the array. It calls function swap (line 50) to exchange the
array elements array[j] and array[j + 1] (Fig. 7.15). Remember that C enforces information hiding between functions, so swap does not have access to individual array elements
in bubbleSort. Because bubbleSort wants swap to have access to the array elements to be
swapped, bubbleSort passes each of these elements by reference to swap—the address of
each array element is passed explicitly. Although entire arrays are automatically passed by
reference, individual array elements are scalars and are ordinarily passed by value. There-
292
Chapter 7 C Pointers
fore, bubbleSort uses the address operator (&) on each of the array elements in the swap
call (line 50) to effect pass-by-reference as follows
swap( &array[ j ], &array[ j + 1 ] );
Function swap receives &array[j] in pointer variable element1Ptr (line 58). Even though
swap—because of information hiding—is not allowed to know the name array[j], swap
may use *element1Ptr as a synonym for array[j]—when swap references *element1Ptr,
it’s actually referencing array[j] in bubbleSort. Similarly, when swap references
*element2Ptr, it’s actually referencing array[j + 1] in bubbleSort. Even though swap
is not allowed to say
int hold = array[ j ];
array[ j ] = array[ j + 1 ];
array[ j + 1 ] = hold;
precisely the same effect is achieved by lines 60 through 62
int hold = *element1Ptr;
*element1Ptr = *element2Ptr;
*element2Ptr = hold;
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
// Fig. 7.15: fig07_15.c
// Putting values into an array, sorting the values into
// ascending order and printing the resulting array.
#include <stdio.h>
#define SIZE 10
void bubbleSort( int * const array, size_t size ); // prototype
int main( void )
{
// initialize array a
int a[ SIZE ] = { 2, 6, 4, 8, 10, 12, 89, 68, 45, 37 };
size_t i; // counter
puts( "Data items in original order" );
// loop through array a
for ( i = 0; i < SIZE; ++i ) {
printf( "%4d", a[ i ] );
} // end for
bubbleSort( a, SIZE ); // sort the array
puts( "\nData items in ascending order" );
// loop through array a
for ( i = 0; i < SIZE; ++i ) {
printf( "%4d", a[ i ] );
} // end for
Fig. 7.15 | Putting values into an array, sorting the values into ascending order and printing the
resulting array. (Part 1 of 2.)
7.6 Bubble Sort Using Pass-by-Reference
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
293
puts( "" );
} // end main
// sort an array of integers using bubble sort algorithm
void bubbleSort( int * const array, size_t size )
{
void swap( int *element1Ptr, int *element2Ptr ); // prototype
unsigned int pass; // pass counter
size_t j; // comparison counter
// loop to control passes
for ( pass = 0; pass < size - 1; ++pass ) {
// loop to control comparisons during each pass
for ( j = 0; j < size - 1; ++j ) {
// swap adjacent elements if they’re out of order
if ( array[ j ] > array[ j + 1 ] ) {
swap( &array[ j ], &array[ j + 1 ] );
} // end if
} // end inner for
} // end outer for
} // end function bubbleSort
// swap values at memory locations to which element1Ptr and
// element2Ptr point
void swap( int *element1Ptr, int *element2Ptr )
{
int hold = *element1Ptr;
*element1Ptr = *element2Ptr;
*element2Ptr = hold;
} // end function swap
Data items in original order
2
6
4
8 10 12 89 68
Data items in ascending order
2
4
6
8 10 12 37 45
45
37
68
89
Fig. 7.15 | Putting values into an array, sorting the values into ascending order and printing the
resulting array. (Part 2 of 2.)
Several features of function bubbleSort should be noted. The function header (line
36) declares array as int * const array rather than int array[] to indicate that bubbleSort receives a single-subscripted array as an argument (again, these notations are interchangeable). Parameter size is declared const to enforce the principle of least privilege.
Although parameter size receives a copy of a value in main, and modifying the copy
cannot change the value in main, bubbleSort does not need to alter size to accomplish its
task. The size of the array remains fixed during the execution of function bubbleSort.
Therefore, size is declared const to ensure that it’s not modified.
The prototype for function swap (line 38) is included in the body of function bubbleSort because bubbleSort is the only function that calls swap. Placing the prototype in
294
Chapter 7 C Pointers
bubbleSort restricts proper calls of swap to those made from bubbleSort. Other functions
that attempt to call swap do not have access to a proper function prototype, so the compiler
generates one automatically. This normally results in a prototype that does not match the
function header (and generates a compilation warning or error) because the compiler
assumes int for the return type and the parameter types.
Software Engineering Observation 7.2
Placing function prototypes in the definitions of other functions enforces the principle of
least privilege by restricting proper function calls to the functions in which the prototypes
appear.
Function bubbleSort receives the size of the array as a parameter (line 36). The function must know the size of the array to sort the array. When an array is passed to a function, the memory address of the first element of the array is received by the function. The
address, of course, does not convey the number of elements in the array. Therefore, you
must pass the array size to the function. Another common practice is to pass a pointer to
the beginning of the array and a pointer to the location just beyond the end of the array—
as you’ll learn in Section 7.8, the difference of the two pointers is the length of the array
and the resulting code is simpler.
In the program, the size of the array is explicitly passed to function bubbleSort. There
are two main benefits to this approach—software reusability and proper software engineering. By defining the function to receive the array size as an argument, we enable the
function to be used by any program that sorts single-subscripted integer arrays of any size.
Software Engineering Observation 7.3
When passing an array to a function, also pass the size of the array. This helps make the
function reusable in many programs.
We could have stored the array’s size in a global variable that’s accessible to the entire
program. This would be more efficient, because a copy of the size is not made to pass to
the function. However, other programs that require an integer array-sorting capability
may not have the same global variable, so the function cannot be used in those programs.
Software Engineering Observation 7.4
Global variables usually violate the principle of least privilege and can lead to poor
software engineering. Global variables should be used only to represent truly shared
resources, such as the time of day.
The size of the array could have been programmed directly into the function. This
restricts the use of the function to an array of a specific size and significantly reduces its
reusability. Only programs processing single-subscripted integer arrays of the specific size
coded into the function can use the function.
7.7 sizeof Operator
C provides the special unary operator sizeof to determine the size in bytes of an array (or
any other data type). When applied to the name of an array as in Fig. 7.16 (line 15), the
sizeof operator returns the total number of bytes in the array as type size_t. Variables
7.7 sizeof Operator
295
of type float on this computer are stored in 4 bytes of memory, and array is defined to
have 20 elements. Therefore, there are a total of 80 bytes in array.
Performance Tip 7.2
sizeof
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
is a compile-time operator, so it does not incur any execution-time overhead.
// Fig. 7.16: fig07_16.c
// Applying sizeof to an array name returns
// the number of bytes in the array.
#include <stdio.h>
#define SIZE 20
size_t getSize( float *ptr ); // prototype
int main( void )
{
float array[ SIZE ]; // create array
printf( "The number of bytes in the array is %u"
"\nThe number of bytes returned by getSize is %u\n",
sizeof( array ), getSize( array ) );
} // end main
// return size of ptr
size_t getSize( float *ptr )
{
return sizeof( ptr );
} // end function getSize
The number of bytes in the array is 80
The number of bytes returned by getSize is 4
Fig. 7.16 | Applying sizeof to an array name returns the number of bytes in the array.
The number of elements in an array also can be determined with
example, consider the following array definition:
sizeof.
For
double real[ 22 ];
Variables of type double normally are stored in 8 bytes of memory. Thus, array real contains a total of 176 bytes. To determine the number of elements in the array, the following
expression can be used:
sizeof( real ) / sizeof( real[ 0 ] )
The expression determines the number of bytes in array real and divides that value by the
number of bytes used in memory to store the first element of array real (a double value).
Even though function getSize receives an array of 20 elements as an argument, the
function’s parameter ptr is simply a pointer to the array’s first element. When you use
sizeof with a pointer, it returns the size of the pointer, not the size of the item to which it
points. The size of a pointer on our system is 4 bytes, so getSize returned 4. Also, the
296
Chapter 7 C Pointers
calculation shown above for determining the number of array elements using
works only when using the actual array, not when using a pointer to the array.
sizeof
Determining the Sizes of the Standard Types, an Array and a Pointer
Figure 7.17 calculates the number of bytes used to store each of the standard data types.
The results of this program are implementation dependent and often differ across platforms and
sometimes across different compilers on the same platform.
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
// Fig. 7.17: fig07_17.c
// Using operator sizeof to determine standard data type sizes.
#include <stdio.h>
int main( void )
{
char c;
short s;
int i;
long l;
long long ll;
float f;
double d;
long double ld;
int array[ 20 ]; // create array of 20 int elements
int *ptr = array; // create pointer to array
printf( "
sizeof c = %u\tsizeof(char) = %u"
"\n
sizeof s = %u\tsizeof(short) = %u"
"\n
sizeof i = %u\tsizeof(int) = %u"
"\n
sizeof l = %u\tsizeof(long) = %u"
"\n
sizeof ll = %u\tsizeof(long long) = %u"
"\n
sizeof f = %u\tsizeof(float) = %u"
"\n
sizeof d = %u\tsizeof(double) = %u"
"\n
sizeof ld = %u\tsizeof(long double) = %u"
"\n sizeof array = %u"
"\n
sizeof ptr = %u\n",
sizeof c, sizeof( char ), sizeof s, sizeof( short ), sizeof i,
sizeof( int ), sizeof l, sizeof( long ), sizeof ll,
sizeof( long long ), sizeof f, sizeof( float ), sizeof d,
sizeof( double ), sizeof ld, sizeof( long double ),
sizeof array, sizeof ptr );
} // end main
sizeof c
sizeof s
sizeof i
sizeof l
sizeof ll
sizeof f
sizeof d
sizeof ld
sizeof array
sizeof ptr
=
=
=
=
=
=
=
=
=
=
1
2
4
4
8
4
8
8
80
4
sizeof(char) = 1
sizeof(short) = 2
sizeof(int) = 4
sizeof(long) = 4
sizeof(long long) = 8
sizeof(float) = 4
sizeof(double) = 8
sizeof(long double) = 8
Fig. 7.17 | Using operator sizeof to determine standard data type sizes.
7.8 Pointer Expressions and Pointer Arithmetic
297
Portability Tip 7.1
The number of bytes used to store a particular data type may vary between systems. When
writing programs that depend on data type sizes and that will run on several computer
systems, use sizeof to determine the number of bytes used to store the data types.
Operator sizeof can be applied to any variable name, type or value (including the
value of an expression). When applied to a variable name (that’s not an array name) or a
constant, the number of bytes used to store the specific type of variable or constant is
returned. The parentheses are required when a type is supplied as sizeof’s operand.
7.8 Pointer Expressions and Pointer Arithmetic
Pointers are valid operands in arithmetic expressions, assignment expressions and comparison expressions. However, not all the operators normally used in these expressions are
valid in conjunction with pointer variables. This section describes the operators that can
have pointers as operands, and how these operators are used.
A limited set of arithmetic operations may be performed on pointers. A pointer may
be incremented (++) or decremented (--), an integer may be added to a pointer (+ or +=), an
integer may be subtracted from a pointer (- or -=) and one pointer may be subtracted from
another—this last operation is meaningful only when both pointers point to elements of
the same array.
Assume that array int v[5] has been defined and its first element is at location 3000
in memory. Assume pointer vPtr has been initialized to point to v[0]—i.e., the value of
vPtr is 3000. Figure 7.18 illustrates this situation for a machine with 4-byte integers. Variable vPtr can be initialized to point to array v with either of the statements
vPtr = v;
vPtr = &v[ 0 ];
Portability Tip 7.2
Because the results of pointer arithmetic depend on the size of the objects a pointer points
to, pointer arithmetic is machine dependent.
location
3000
3004
v[0]
3008
v[1]
3012
v[2]
3016
v[3]
v[4]
pointer variable vPtr
Fig. 7.18 | Array v and a pointer variable vPtr that points to v.
In conventional arithmetic, 3000 + 2 yields the value 3002. This is normally not the
case with pointer arithmetic. When an integer is added to or subtracted from a pointer,
298
Chapter 7 C Pointers
the pointer is not incremented or decremented simply by that integer, but by that integer
times the size of the object to which the pointer refers. The number of bytes depends on
the object’s data type. For example, the statement
vPtr += 2;
would produce 3008 (3000 + 2 * 4), assuming an integer is stored in 4 bytes of memory.
In the array v, vPtr would now point to v[2] (Fig. 7.19). If an integer is stored in 2 bytes
of memory, then the preceding calculation would result in memory location 3004 (3000 +
2 * 2). If the array were of a different data type, the preceding statement would increment
the pointer by twice the number of bytes that it takes to store an object of that data type.
When performing pointer arithmetic on a character array, the results will be consistent
with regular arithmetic, because each character is 1 byte long.
location
3000
3004
v[0]
3008
v[1]
3012
v[2]
3016
v[3]
v[4]
pointer variable vPtr
Fig. 7.19 | The pointer vPtr after pointer arithmetic.
If vPtr had been incremented to 3016, which points to v[4], the statement
vPtr -= 4;
would set vPtr back to 3000—the beginning of the array. If a pointer is being incremented
or decremented by one, the increment (++) and decrement (--) operators can be used. Either of the statements
++vPtr;
vPtr++;
increments the pointer to point to the next location in the array. Either of the statements
--vPtr;
vPtr--;
decrements the pointer to point to the previous element of the array.
Pointer variables may be subtracted from one another. For example, if vPtr contains
the location 3000, and v2Ptr contains the address 3008, the statement
x = v2Ptr - vPtr;
would assign to x the number of array elements from vPtr to v2Ptr, in this case 2 (not 8).
Pointer arithmetic is undefined unless performed on an array. We cannot assume that two
variables of the same type are stored contiguously in memory unless they’re adjacent elements of an array.
7.9 Relationship between Pointers and Arrays
299
Common Programming Error 7.4
Using pointer arithmetic on a pointer that does not refer to an element in an array.
Common Programming Error 7.5
Subtracting or comparing two pointers that do not refer to elements in the same array.
Common Programming Error 7.6
Running off either end of an array when using pointer arithmetic.
A pointer can be assigned to another pointer if both have the same type. The exception
to this rule is the pointer to void (i.e., void *), which is a generic pointer that can represent
any pointer type. All pointer types can be assigned a pointer to void, and a pointer to void
can be assigned a pointer of any type. In both cases, a cast operation is not required.
A pointer to void cannot be dereferenced. Consider this: The compiler knows that a
pointer to int refers to 4 bytes of memory on a machine with 4-byte integers, but a pointer
to void simply contains a memory location for an unknown data type—the precise number
of bytes to which the pointer refers is not known by the compiler. The compiler must know
the data type to determine the number of bytes to be dereferenced for a particular pointer.
Common Programming Error 7.7
Assigning a pointer of one type to a pointer of another type if neither is of type void * is
a syntax error.
Common Programming Error 7.8
Dereferencing a void * pointer is a syntax error.
Pointers can be compared using equality and relational operators, but such comparisons are meaningless unless the pointers point to elements of the same array. Pointer comparisons compare the addresses stored in the pointers. A comparison of two pointers
pointing to elements in the same array could show, for example, that one pointer points
to a higher-numbered element of the array than the other pointer does. A common use of
pointer comparison is determining whether a pointer is NULL.
7.9 Relationship between Pointers and Arrays
Arrays and pointers are intimately related in C and often may be used interchangeably. An
array name can be thought of as a constant pointer. Pointers can be used to do any operation
involving array subscripting.
Assume that integer array b[5] and integer pointer variable bPtr have been defined.
Because the array name (without a subscript) is a pointer to the first element of the array,
we can set bPtr equal to the address of the first element in array b with the statement
bPtr = b;
This statement is equivalent to taking the address of the array’s first element as follows:
bPtr = &b[ 0 ];
300
Chapter 7 C Pointers
Array element b[3] can alternatively be referenced with the pointer expression
*( bPtr + 3 )
The 3 in the expression is the offset to the pointer. When the pointer points to the array’s
first element, the offset indicates which array element should be referenced, and the offset
value is identical to the array subscript. This notation is referred to as pointer/offset notation. The parentheses are necessary because the precedence of * is higher than the precedence
of +. Without the parentheses, the above expression would add 3 to the value of the expression *bPtr (i.e., 3 would be added to b[0], assuming bPtr points to the beginning of the
array). Just as the array element can be referenced with a pointer expression, the address
&b[ 3 ]
can be written with the pointer expression
bPtr + 3
The array itself can be treated as a pointer and used in pointer arithmetic. For
example, the expression
*( b + 3 )
also refers to the array element b[3]. In general, all subscripted array expressions can be
written with a pointer and an offset. In this case, pointer/offset notation was used with the
name of the array as a pointer. The preceding statement does not modify the array name
in any way; b still points to the first element in the array.
Pointers can be subscripted like arrays. If bPtr has the value b, the expression
bPtr[ 1 ]
refers to the array element b[1]. This is referred to as pointer/subscript notation.
Remember that an array name is essentially a constant pointer; it always points to the
beginning of the array. Thus, the expression
b += 3
is invalid because it attempts to modify the value of the array name with pointer arithmetic.
Common Programming Error 7.9
Attempting to modify an array name with pointer arithmetic is a compilation error.
Figure 7.20 uses the four methods we’ve discussed for referring to array elements—
array subscripting, pointer/offset with the array name as a pointer, pointer subscripting,
and pointer/offset with a pointer—to print the four elements of the integer array b.
1
2
3
4
5
6
7
// Fig. 7.20: fig07_20.cpp
// Using subscripting and pointer notations with arrays.
#include <stdio.h>
#define ARRAY_SIZE 4
int main( void )
{
Fig. 7.20 | Using subscripting and pointer notations with arrays. (Part 1 of 3.)
7.9 Relationship between Pointers and Arrays
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
int b[] = { 10, 20, 30, 40 }; // create and initialize array b
int *bPtr = b; // create bPtr and point it to array b
size_t i; // counter
size_t offset; // counter
// output array b using array subscript notation
puts( "Array b printed with:\nArray subscript notation" );
// loop through array b
for ( i = 0; i < ARRAY_SIZE; ++i ) {
printf( "b[ %u ] = %d\n", i, b[ i ] );
} // end for
// output array b using array name and pointer/offset notation
puts( "\nPointer/offset notation where\n"
"the pointer is the array name" );
// loop through array b
for ( offset = 0; offset < ARRAY_SIZE; ++offset ) {
printf( "*( b + %u ) = %d\n", offset, *( b + offset ) );
} // end for
// output array b using bPtr and array subscript notation
puts( "\nPointer subscript notation" );
// loop through array b
for ( i = 0; i < ARRAY_SIZE; ++i ) {
printf( "bPtr[ %u ] = %d\n", i, bPtr[ i ] );
} // end for
// output array b using bPtr and pointer/offset notation
puts( "\nPointer/offset notation" );
// loop through array b
for ( offset = 0; offset < ARRAY_SIZE; ++offset ) {
printf( "*( bPtr + %u ) = %d\n", offset, *( bPtr + offset ) );
} // end for
} // end main
Array b printed with:
Array subscript notation
b[ 0 ] = 10
b[ 1 ] = 20
b[ 2 ] = 30
b[ 3 ] = 40
Pointer/offset notation where
the pointer is the array name
*( b + 0 ) = 10
*( b + 1 ) = 20
*( b + 2 ) = 30
*( b + 3 ) = 40
Fig. 7.20 | Using subscripting and pointer notations with arrays. (Part 2 of 3.)
301
302
Chapter 7 C Pointers
Pointer
bPtr[ 0
bPtr[ 1
bPtr[ 2
bPtr[ 3
subscript notation
] = 10
] = 20
] = 30
] = 40
Pointer/offset notation
*( bPtr + 0 ) = 10
*( bPtr + 1 ) = 20
*( bPtr + 2 ) = 30
*( bPtr + 3 ) = 40
Fig. 7.20 | Using subscripting and pointer notations with arrays. (Part 3 of 3.)
String Copying with Arrays and Pointers
To further illustrate the interchangeability of arrays and pointers, let’s look at the two
string-copying functions—copy1 and copy2—in the program of Fig. 7.21. Both functions
copy a string into a character array. After a comparison of the function prototypes for
copy1 and copy2, the functions appear identical. They accomplish the same task, but
they’re implemented differently.
Function copy1 uses array subscript notation to copy the string in s2 to the character
array s1. The function defines counter variable i as the array subscript. The for statement
header (line 29) performs the entire copy operation—its body is the empty statement. The
header specifies that i is initialized to zero and incremented by one on each iteration of
the loop. The expression s1[i] = s2[i] copies one character from s2 to s1. When the null
character is encountered in s2, it’s assigned to s1, and the value of the assignment becomes
the value assigned to the left operand (s1). The loop terminates when the null character is
assigned from s2 to s1 (false).
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
// Fig. 7.21: fig07_21.c
// Copying a string using array notation and pointer notation.
#include <stdio.h>
#define SIZE 10
void copy1( char * const s1, const char * const s2 ); // prototype
void copy2( char *s1, const char *s2 ); // prototype
int main( void )
{
char string1[ SIZE ]; // create array string1
char *string2 = "Hello"; // create a pointer to a string
char string3[ SIZE ]; // create array string3
char string4[] = "Good Bye"; // create a pointer to a string
copy1( string1, string2 );
printf( "string1 = %s\n", string1 );
Fig. 7.21 | Copying a string using array notation and pointer notation. (Part 1 of 2.)
7.10 Arrays of Pointers
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
303
copy2( string3, string4 );
printf( "string3 = %s\n", string3 );
} // end main
// copy s2 to s1 using array notation
void copy1( char * const s1, const char * const s2 )
{
size_t i; // counter
// loop through strings
for ( i = 0; ( s1[ i ] = s2[ i ] ) != '\0'; ++i ) {
; // do nothing in body
} // end for
} // end function copy1
// copy s2 to s1 using pointer notation
void copy2( char *s1, const char *s2 )
{
// loop through strings
for ( ; ( *s1 = *s2 ) != '\0'; ++s1, ++s2 ) {
; // do nothing in body
} // end for
} // end function copy2
string1 = Hello
string3 = Good Bye
Fig. 7.21 | Copying a string using array notation and pointer notation. (Part 2 of 2.)
Function copy2 uses pointers and pointer arithmetic to copy the string in s2 to the
character array s1. Again, the for statement header (line 38) performs the entire copy
operation. The header does not include any variable initialization. As in function copy1,
the expression (*s1 = *s2) performs the copy operation. Pointer s2 is dereferenced, and
the resulting character is assigned to the dereferenced pointer *s1. After the assignment in
the condition, the pointers are incremented to point to the next element of array s1 and
the next character of string s2, respectively. When the null character is encountered in s2,
it’s assigned to the dereferenced pointer s1 and the loop terminates.
The first argument to both copy1 and copy2 must be an array large enough to hold the
string in the second argument. Otherwise, an error may occur when an attempt is made to
write into a memory location that’s not part of the array. Also, the second parameter of
each function is declared as const char * (a constant string). In both functions, the
second argument is copied into the first argument—characters are read from it one at a
time, but the characters are never modified. Therefore, the second parameter is declared to
point to a constant value so that the principle of least privilege is enforced—neither function
requires the capability of modifying the second argument, so neither function is provided
with that capability.
7.10 Arrays of Pointers
Arrays may contain pointers. A common use of an array of pointers is to form an array of
strings, referred to simply as a string array. Each entry in the array is a string, but in C a
304
Chapter 7 C Pointers
string is essentially a pointer to its first character. So each entry in an array of strings is actually a pointer to the first character of a string. Consider the definition of string array
suit, which might be useful in representing a deck of cards.
const char *suit[ 4 ] = { "Hearts", "Diamonds", "Clubs", "Spades"
};
The suit[4] portion of the definition indicates an array of 4 elements. The char * portion of the declaration indicates that each element of array suit is of type “pointer to
char.” Qualifier const indicates that the strings pointed to by each element pointer will
not be modified. The four values to be placed in the array are "Hearts", "Diamonds",
"Clubs" and "Spades". Each is stored in memory as a null-terminated character string
that’s one character longer than the number of characters between quotes. The four strings
are 7, 9, 6 and 7 characters long, respectively. Although it appears as though these strings
are being placed in the suit array, only pointers are actually stored in the array
(Fig. 7.22). Each pointer points to the first character of its corresponding string. Thus,
even though the suit array is fixed in size, it provides access to character strings of any
length. This flexibility is one example of C’s powerful data-structuring capabilities.
suit[0]
'H'
'e'
'a'
'r'
't'
's'
'\0'
suit[1]
'D'
'i'
'a'
'm'
'o'
'n'
'd'
suit[2]
'C'
'l'
'u'
'b'
's'
'\0'
suit[3]
'S'
'p'
'a'
'd'
'e'
's'
's'
'\0'
'\0'
Fig. 7.22 | Graphical representation of the suit array.
The suits could have been placed in a two-dimensional array, in which each row
would represent a suit and each column would represent a letter from a suit name. Such a
data structure would have to have a fixed number of columns per row, and that number
would have to be as large as the largest string. Therefore, considerable memory could be
wasted when storing a large number of strings of which most were shorter than the longest
string. We use string arrays to represent a deck of cards in the next section.
7.11 Case Study: Card Shuffling and Dealing Simulation
In this section, we use random number generation to develop a card shuffling and dealing
simulation program. This program can then be used to implement programs that play
specific card games. To reveal some subtle performance problems, we’ve intentionally used
suboptimal shuffling and dealing algorithms. In this chapter’s exercises and in Chapter 10,
we develop more efficient algorithms.
Using the top-down, stepwise refinement approach, we develop a program that will
shuffle a deck of 52 playing cards and then deal each of the 52 cards. The top-down
approach is particularly useful in attacking larger, more complex problems than you’ve
seen in earlier chapters.
7.11 Case Study: Card Shuffling and Dealing Simulation
305
Hearts
Ace
Two
Three
Four
Five
Six
Seven
Eight
Nine
Ten
Jack
Queen
King
We use 4-by-13 double-subscripted array deck to represent the deck of playing cards
(Fig. 7.23). The rows correspond to the suits—row 0 corresponds to hearts, row 1 to diamonds, row 2 to clubs and row 3 to spades. The columns correspond to the face values of
the cards—columns 0 through 9 correspond to ace through ten respectively, and columns
10 through 12 correspond to jack, queen and king. We shall load string array suit with
character strings representing the four suits, and string array face with character strings
representing the thirteen face values.
0
1
2
3
4
5
6
7
8
9
10
11
12
0
Diamonds 1
Clubs
2
Spades
3
deck[2][12] represents the King of Clubs
Clubs
King
Fig. 7.23 | Double-subscripted array representation of a deck of cards.
This simulated deck of cards may be shuffled as follows. First the array deck is cleared
to zeros. Then, a row (0–3) and a column (0–12) are each chosen at random. The number
1 is inserted in array element deck[row][column] to indicate that this card will be the first
one dealt from the shuffled deck. This process continues with the numbers 2, 3, …, 52
being randomly inserted in the deck array to indicate which cards are to be placed second,
third, …, and fifty-second in the shuffled deck. As the deck array begins to fill with card
numbers, it’s possible that a card will be selected again—i.e., deck[row] [column] will be
nonzero when it’s selected. This selection is simply ignored and other rows and columns
are repeatedly chosen at random until an unselected card is found. Eventually, the numbers
1 through 52 will occupy the 52 slots of the deck array. At this point, the deck of cards is
fully shuffled.
This shuffling algorithm can execute indefinitely if cards that have already been shuffled are repeatedly selected at random. This phenomenon is known as indefinite postponement. In this chapter’s exercises, we discuss a better shuffling algorithm that
eliminates the possibility of indefinite postponement.
Performance Tip 7.3
Sometimes an algorithm that emerges in a “natural” way can contain subtle performance
problems, such as indefinite postponement. Seek algorithms that avoid indefinite postponement.
To deal the first card, we search the array for deck[row][column] equal to 1. This is
accomplished with nested for statements that vary row from 0 to 3 and column from 0 to
306
Chapter 7 C Pointers
12. What card does that element of the array correspond to? The suit array has been preloaded with the four suits, so to get the suit, we print the character string suit[row]. Similarly, to get the face value of the card, we print the character string face[column]. We
also print the character string " of ". Printing this information in the proper order enables
us to print each card in the form "King of Clubs", "Ace of Diamonds" and so on.
Let’s proceed with the top-down, stepwise refinement process. The top is simply
Shuffle and deal 52 cards
Our first refinement yields:
Initialize the suit array
Initialize the face array
Initialize the deck array
Shuffle the deck
Deal 52 cards
“Shuffle the deck” may be expanded as follows:
For each of the 52 cards
Place card number in randomly selected unoccupied slot of deck
“Deal 52 cards” may be expanded as follows:
For each of the 52 cards
Find card number in deck array and print face and suit of card
Incorporating these expansions yields our complete second refinement:
Initialize the suit array
Initialize the face array
Initialize the deck array
For each of the 52 cards
Place card number in randomly selected unoccupied slot of deck
For each of the 52 cards
Find card number in deck array and print face and suit of card
“Place card number in randomly selected unoccupied slot of deck” may be expanded
as:
Choose slot of deck randomly
While chosen slot of deck has been previously chosen
Choose slot of deck randomly
Place card number in chosen slot of deck
“Find card number in deck array and print face and suit of card” may be expanded as:
For each slot of the deck array
If slot contains card number
Print the face and suit of the card
7.11 Case Study: Card Shuffling and Dealing Simulation
307
Incorporating these expansions yields our third refinement:
Initialize the suit array
Initialize the face array
Initialize the deck array
For each of the 52 cards
Choose slot of deck randomly
While slot of deck has been previously chosen
Choose slot of deck randomly
Place card number in chosen slot of deck
For each of the 52 cards
For each slot of deck array
If slot contains desired card number
Print the face and suit of the card
This completes the refinement process. This program is more efficient if the shuffle
and deal portions of the algorithm are combined so that each card is dealt as it’s placed in
the deck. We’ve chosen to program these operations separately because normally cards are
dealt after they’re shuffled (not while they’re being shuffled).
The card shuffling and dealing program is shown in Fig. 7.24, and a sample execution
is shown in Fig. 7.25. Conversion specifier %s is used to print strings of characters in the
calls to printf. The corresponding argument in the printf call must be a pointer to char
(or a char array). The format specification "%5s of %-8s" (line 74) prints a character string
right justified in a field of five characters followed by " of " and a character string left justified in a field of eight characters. The minus sign in %-8s signifies left justification.
There’s a weakness in the dealing algorithm. Once a match is found, the two inner
for statements continue searching the remaining elements of deck for a match. We correct
this deficiency in this chapter’s exercises and in a Chapter 10 case study.
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
// Fig. 7.24: fig07_24.c
// Card shuffling and dealing.
#include <stdio.h>
#include <stdlib.h>
#include <time.h>
#define SUITS 4
#define FACES 13
#define CARDS 52
// prototypes
void shuffle( unsigned int wDeck[][ FACES ] ); // shuffling modifies wDeck
void deal( unsigned int wDeck[][ FACES ], const char *wFace[],
const char *wSuit[] ); // dealing doesn't modify the arrays
int main( void )
{
Fig. 7.24 | Card shuffling and dealing. (Part 1 of 3.)
308
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
Chapter 7 C Pointers
// initialize suit array
const char *suit[ SUITS ] =
{ "Hearts", "Diamonds", "Clubs", "Spades" };
// initialize face array
const char *face[ FACES ] =
{ "Ace", "Deuce", "Three", "Four",
"Five", "Six", "Seven", "Eight",
"Nine", "Ten", "Jack", "Queen", "King" };
// initialize deck array
unsigned int deck[ SUITS ][ FACES ] = { 0 };
srand( time( NULL ) ); // seed random-number generator
shuffle( deck ); // shuffle the deck
deal( deck, face, suit ); // deal the deck
} // end main
// shuffle cards in deck
void shuffle( unsigned int wDeck[][ FACES ] )
{
size_t row; // row number
size_t column; // column number
size_t card; // counter
// for each of the cards, choose slot of deck randomly
for ( card = 1; card <= CARDS; ++card ) {
// choose new random location until unoccupied slot found
do {
row = rand() % SUITS;
column = rand() % FACES;
} while( wDeck[ row ][ column ] != 0 ); // end do...while
// place card number in chosen slot of deck
wDeck[ row ][ column ] = card;
} // end for
} // end function shuffle
// deal cards in deck
void deal( unsigned int wDeck[][ FACES ], const char *wFace[],
const char *wSuit[] )
{
size_t card; // card counter
size_t row; // row counter
size_t column; // column counter
// deal each of the cards
for ( card = 1; card <= CARDS; ++card ) {
// loop through rows of wDeck
for ( row = 0; row < SUITS; ++row ) {
Fig. 7.24 | Card shuffling and dealing. (Part 2 of 3.)
7.12 Pointers to Functions
70
71
72
73
74
75
76
77
78
79
80
309
// loop through columns of wDeck for current row
for ( column = 0; column < FACES; ++column ) {
// if slot contains current card, display card
if ( wDeck[ row ][ column ] == card ) {
printf( "%5s of %-8s%c", wFace[ column ], wSuit[ row ],
card % 2 == 0 ? '\n' : '\t' ); // 2-column format
} // end if
} // end for
} // end for
} // end for
} // end function deal
Fig. 7.24 | Card shuffling and dealing. (Part 3 of 3.)
Nine
Queen
Queen
King
Jack
Seven
Three
Three
Queen
Six
Ace
Nine
Eight
Deuce
Deuce
Four
Four
Seven
King
Jack
Jack
Eight
Ace
Four
King
Three
of
of
of
of
of
of
of
of
of
of
of
of
of
of
of
of
of
of
of
of
of
of
of
of
of
of
Hearts
Spades
Hearts
Hearts
Diamonds
Hearts
Clubs
Diamonds
Diamonds
Diamonds
Spades
Diamonds
Spades
Clubs
Spades
Clubs
Spades
Diamonds
Spades
Hearts
Spades
Diamonds
Diamonds
Hearts
Diamonds
Hearts
Five
Three
Ace
Six
Five
King
Eight
Four
Five
Five
Six
Queen
Nine
Six
Jack
Eight
Seven
Seven
Ten
Ace
Ten
Deuce
Nine
Deuce
Ten
Ten
of
of
of
of
of
of
of
of
of
of
of
of
of
of
of
of
of
of
of
of
of
of
of
of
of
of
Clubs
Spades
Clubs
Spades
Spades
Clubs
Hearts
Diamonds
Diamonds
Hearts
Hearts
Clubs
Clubs
Clubs
Clubs
Clubs
Spades
Clubs
Diamonds
Hearts
Clubs
Diamonds
Spades
Hearts
Spades
Hearts
Fig. 7.25 | Sample run of card dealing program.
7.12 Pointers to Functions
A pointer to a function contains the address of the function in memory. In Chapter 6, we
saw that an array name is really the address in memory of the first element of the array.
Similarly, a function name is really the starting address in memory of the code that performs the function’s task. Pointers to functions can be passed to functions, returned from
functions, stored in arrays and assigned to other function pointers.
To illustrate the use of pointers to functions, Fig. 7.26 presents a modified version of
the bubble sort program in Fig. 7.15. The new version consists of main and functions
bubble, swap, ascending and descending. Function bubbleSort receives a pointer to a
310
Chapter 7 C Pointers
function—either function ascending or function descending—as an argument, in addition to an integer array and the size of the array. The program prompts the user to choose
whether the array should be sorted in ascending or in descending order. If the user enters 1,
a pointer to function ascending is passed to function bubble, causing the array to be
sorted into increasing order. If the user enters 2, a pointer to function descending is passed
to function bubble, causing the array to be sorted into decreasing order. The output of the
program is shown in Fig. 7.27.
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
// Fig. 7.26: fig07_26.c
// Multipurpose sorting program using function pointers.
#include <stdio.h>
#define SIZE 10
// prototypes
void bubble( int work[], size_t size, int (*compare)( int a, int b ) );
int ascending( int a, int b );
int descending( int a, int b );
int main( void )
{
int order; // 1 for ascending order or 2 for descending order
size_t counter; // counter
// initialize unordered array a
int a[ SIZE ] = { 2, 6, 4, 8, 10, 12, 89, 68, 45, 37 };
printf( "%s", "Enter 1 to sort in ascending order,\n"
"Enter 2 to sort in descending order: " );
scanf( "%d", &order );
puts( "\nData items in original order" );
// output original array
for ( counter = 0; counter < SIZE; ++counter ) {
printf( "%5d", a[ counter ] );
} // end for
// sort array in ascending order; pass function ascending as an
// argument to specify ascending sorting order
if ( order == 1 ) {
bubble( a, SIZE, ascending );
puts( "\nData items in ascending order" );
} // end if
else { // pass function descending
bubble( a, SIZE, descending );
puts( "\nData items in descending order" );
} // end else
// output sorted array
for ( counter = 0; counter < SIZE; ++counter ) {
printf( "%5d", a[ counter ] );
} // end for
Fig. 7.26 | Multipurpose sorting program using function pointers. (Part 1 of 2.)
7.12 Pointers to Functions
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
311
puts( "\n" );
} // end main
// multipurpose bubble sort; parameter compare is a pointer to
// the comparison function that determines sorting order
void bubble( int work[], size_t size, int (*compare)( int a, int b ) )
{
unsigned int pass; // pass counter
size_t count; // comparison counter
void swap( int *element1Ptr, int *element2ptr ); // prototype
// loop to control passes
for ( pass = 1; pass < size; ++pass ) {
// loop to control number of comparisons per pass
for ( count = 0; count < size - 1; ++count ) {
// if adjacent elements are out of order, swap them
if ( (*compare)( work[ count ], work[ count + 1 ] ) ) {
swap( &work[ count ], &work[ count + 1 ] );
} // end if
} // end for
} // end for
} // end function bubble
// swap values at memory locations to which element1Ptr and
// element2Ptr point
void swap( int *element1Ptr, int *element2Ptr )
{
int hold; // temporary holding variable
hold = *element1Ptr;
*element1Ptr = *element2Ptr;
*element2Ptr = hold;
} // end function swap
// determine whether elements are out of order for an ascending
// order sort
int ascending( int a, int b )
{
return b < a; // should swap if b is less than a
} // end function ascending
// determine whether elements are out of order for a descending
// order sort
int descending( int a, int b )
{
return b > a; // should swap if b is greater than a
} // end function descending
Fig. 7.26 | Multipurpose sorting program using function pointers. (Part 2 of 2.)
312
Chapter 7 C Pointers
Enter 1 to sort in ascending order,
Enter 2 to sort in descending order: 1
Data items in original order
2
6
4
8
10
12
Data items in ascending order
2
4
6
8
10
12
89
68
45
37
37
45
68
89
89
68
45
37
8
6
4
2
Enter 1 to sort in ascending order,
Enter 2 to sort in descending order: 2
Data items in original order
2
6
4
8
10
12
Data items in descending order
89
68
45
37
12
10
Fig. 7.27 | The outputs of the bubble sort program in Fig. 7.26.
The following parameter appears in the function header for bubble (line 51)
int (*compare)( int a, int b )
This tells bubble to expect a parameter (compare) that’s a pointer to a function that receives two integer parameters and returns an integer result. Parentheses are needed around
*compare to group the * with compare to indicate that compare is a pointer. If we had not
included the parentheses, the declaration would have been
int *compare( int a, int b )
which declares a function that receives two integers as parameters and returns a pointer to
an integer.
The function prototype for bubble is shown in line 7. The third parameter in the prototype could have been written as
int (*)( int, int );
without the function-pointer name and parameter names.
The function passed to bubble is called in an if statement (line 65) as follows:
if ( (*compare)( work[ count ], work[ count + 1 ] ) )
Just as a pointer to a variable is dereferenced to access the value of the variable, a pointer to
a function is dereferenced to use the function.
The call to the function could have been made without dereferencing the pointer as in
if ( compare( work[ count ], work[ count + 1 ] ) )
which uses the pointer directly as the function name. We prefer the first method of calling
a function through a pointer because it explicitly illustrates that compare is a pointer to a
function that’s dereferenced to call the function. The second method of calling a function
through a pointer makes it appear as if compare is an actual function. This may be confusing to a programmer reading the code who would like to see the definition of function
compare and finds that it’s never defined in the file.
7.12 Pointers to Functions
313
Using Function Pointers to Create a Menu-Driven System
A common use of function pointers is in text-based menu-driven systems. A user is prompted to select an option from a menu (possibly from 1 to 5) by typing the menu item’s number. Each option is serviced by a different function. Pointers to each function are stored in
an array of pointers to functions. The user’s choice is used as a subscript in the array, and
the pointer in the array is used to call the function.
Figure 7.28 provides a generic example of the mechanics of defining and using an
array of pointers to functions. We define three functions—function1, function2 and
function3—that each take an integer argument and return nothing. We store pointers to
these three functions in array f, which is defined in line 14.
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
// Fig. 7.28: fig07_28.c
// Demonstrating an array of pointers to functions.
#include <stdio.h>
// prototypes
void function1( int a );
void function2( int b );
void function3( int c );
int main( void )
{
// initialize array of 3 pointers to functions that each take an
// int argument and return void
void (*f[ 3 ])( int ) = { function1, function2, function3 };
size_t choice; // variable to hold user's choice
printf( "%s", "Enter a number between 0 and 2, 3 to end: " );
scanf( "%u", &choice );
// process user's choice
while ( choice >= 0 && choice < 3 ) {
// invoke function at location choice in array f and pass
// choice as an argument
(*f[ choice ])( choice );
printf( "%s", "Enter a number between 0 and 2, 3 to end: " );
scanf( "%u", &choice );
} // end while
puts( "Program execution completed." );
} // end main
void function1( int a )
{
printf( "You entered %d so function1 was called\n\n", a );
} // end function1
Fig. 7.28 | Demonstrating an array of pointers to functions. (Part 1 of 2.)
314
40
41
42
43
44
45
46
47
48
Chapter 7 C Pointers
void function2( int b )
{
printf( "You entered %d so function2 was called\n\n", b );
} // end function2
void function3( int c )
{
printf( "You entered %d so function3 was called\n\n", c );
} // end function3
Enter a number between 0 and 2, 3 to end: 0
You entered 0 so function1 was called
Enter a number between 0 and 2, 3 to end: 1
You entered 1 so function2 was called
Enter a number between 0 and 2, 3 to end: 2
You entered 2 so function3 was called
Enter a number between 0 and 2, 3 to end: 3
Program execution completed.
Fig. 7.28 | Demonstrating an array of pointers to functions. (Part 2 of 2.)
The definition is read beginning at the leftmost set of parentheses, “f is an array of 3
pointers to functions that each take an int as an argument and return void.” The array is
initialized with the names of the three functions. When the user enters a value between 0
and 2, the value is used as the subscript into the array of pointers to functions. In the function call (line 26), f[choice] selects the pointer at location choice in the array. The
pointer is dereferenced to call the function, and choice is passed as the argument to the function. Each function prints its argument’s value and its function name to demonstrate that
the function is called correctly. In this chapter’s exercises, you’ll develop several text-based,
menu-driven systems.
7.13 Secure C Programming
printf_s, scanf_s
and Other Secure Functions
Earlier Secure C Programming sections presented printf_s and scanf_s, and mentioned
other more secure versions of standard library functions that are described by Annex K of the
C standard. A key feature of functions like printf_s and scanf_s that makes them more
secure is that they have runtime constraints requiring their pointer arguments to be non-NULL.
The functions check these runtime constraints before attempting to use the pointers. Any
NULL pointer argument is considered to be a constraint violation and causes the function to
fail and return a status notification. In a scanf_s, if any of the pointer arguments (including
the format-control string) are NULL, the function returns EOF. In a printf_s, if the formatcontrol string or any argument that corresponds to a %s is NULL, the function stops outputting data and returns a negative number. For complete details of the Annex K functions, see
the C standard document or your compiler’s library documentation.
Summary
315
Other CERT Guidelines Regarding Pointers
Misused pointers lead to many of the most common security vulnerabilities in systems today. CERT provides various guidelines to help you prevent such problems. If you’re building industrial-strength C systems, you should familiarize yourself with the CERT C Secure
Coding Standard at www.securecoding.cert.org. The following guidelines apply to
pointer programming techniques that we presented in this chapter:
• EXP34-C: Dereferencing NULL pointers typically causes programs to crash, but
CERT has encountered cases in which dereferencing NULL pointers can allow attackers to execute code.
• DCL13-C: Section 7.5 discussed uses of const with pointers. If a function parameter points to a value that will not be changed by the function, const should be used
to indicate that the data is constant. For example, to represent a pointer to a string
that will not be modified, use const char * as the pointer parameter’s type.
• MSC16-C: This guideline discusses techniques for encrypting function pointers
to help prevent attackers from overwriting them and executing attack code.
Summary
Section 7.2 Pointer Variable Definitions and Initialization
• A pointer contains an address of another variable that contains a value. In this sense, a variable
name directly references a value, and a pointer indirectly references a value.
• Referencing a value through a pointer is called indirection.
• Pointers can be defined to point to objects of any type.
• Pointers should be initialized either when they’re defined or in an assignment statement. A pointer may be initialized to NULL, 0 or an address. A pointer with the value NULL points to nothing.
Initializing a pointer to 0 is equivalent to initializing a pointer to NULL, but NULL is preferred. The
value 0 is the only integer value that can be assigned directly to a pointer variable.
• NULL is a symbolic constant defined in the <stddef.h> header (and several other headers).
Section 7.3 Pointer Operators
•
•
•
•
The &, or address operator, is a unary operator that returns the address of its operand.
The operand of the address operator must be a variable.
The indirection operator * returns the value of the object to which its operand points.
The printf conversion specifier %p outputs a memory location as a hexadecimal integer on most
platforms.
Section 7.4 Passing Arguments to Functions by Reference
• All arguments in C are passed by value.
• C provides the capabilities for simulating pass-by-reference using pointers and the indirection
operator. To pass a variable by reference, apply the address operator (&) to the variable’s name.
316
Chapter 7 C Pointers
• When the address of a variable is passed to a function, the indirection operator (*) may be used
in the function to modify the value at that location in the caller’s memory.
• A function receiving an address as an argument must define a pointer parameter to receive the
address.
• The compiler does not differentiate between a function that receives a pointer and one that receives a single-subscripted array. A function must “know” when it’s receiving an array vs. a single
variable passed by reference.
• When the compiler encounters a function parameter for a single-subscripted array of the form
int b[], the compiler converts the parameter to the pointer notation int *b.
Section 7.5 Using the const Qualifier with Pointers
• The const qualifier indicates that the value of a particular variable should not be modified.
• If an attempt is made to modify a value that’s declared const, the compiler catches it and issues
either a warning or an error, depending on the particular compiler.
• There are four ways to pass a pointer to a function: a non-constant pointer to non-constant data,
a constant pointer to non-constant data, a non-constant pointer to constant data, and a constant
pointer to constant data.
• With a non-constant pointer to non-constant data, the data can be modified through the dereferenced pointer, and the pointer can be modified to point to other data items.
• A non-constant pointer to constant data can be modified to point to any data item of the appropriate type, but the data to which it points cannot be modified.
• A constant pointer to non-constant data always points to the same memory location, and the data
at that location can be modified through the pointer. This is the default for an array name.
• A constant pointer to constant data always points to the same memory location, and the data at
that memory location cannot be modified.
Section 7.7 sizeof Operator
• Unary operator sizeof determine the size in bytes of a variable or type at compilation time.
• When applied to the name of an array, sizeof returns the total number of bytes in the array.
• Type size_t is an integral type (unsigned int or unsigned long int) returned by operator
sizeof. Type size_t is defined in header <stddef.h>.
• Operator sizeof can be applied to any variable name, type or value.
• The parentheses used with sizeof are required if a type name is supplied as its operand.
Section 7.8 Pointer Expressions and Pointer Arithmetic
• A limited set of arithmetic operations may be performed on pointers. A pointer may be incremented (++) or decremented (--), an integer may be added to a pointer (+ or +=), an integer may
be subtracted from a pointer (- or -=) and one pointer may be subtracted from another.
• When an integer is added to or subtracted from a pointer, the pointer is incremented or decremented by that integer times the size of the object to which the pointer refers.
• Two pointers to elements of the same array may be subtracted from one another to determine
the number of elements between them.
• A pointer can be assigned to another pointer if both have the same type. An exception is the
pointer of type void * which can represent any pointer type. All pointer types can be assigned a
void * pointer, and a void * pointer can be assigned a pointer of any type.
• A void * pointer cannot be dereferenced.
Terminology
317
• Pointers can be compared using equality and relational operators, but such comparisons are
meaningless unless the pointers point to elements of the same array. Pointer comparisons compare the addresses stored in the pointers.
• A common use of pointer comparison is determining whether a pointer is NULL.
Section 7.9 Relationship between Pointers and Arrays
•
•
•
•
Arrays and pointers are intimately related in C and often may be used interchangeably.
An array name can be thought of as a constant pointer.
Pointers can be used to do any operation involving array subscripting.
When a pointer points to the beginning of an array, adding an offset to the pointer indicates
which element of the array should be referenced, and the offset value is identical to the array subscript. This is referred to as pointer/offset notation.
• An array name can be treated as a pointer and used in pointer arithmetic expressions that do not
attempt to modify the address of the pointer.
• Pointers can be subscripted exactly as arrays can. This is referred to as pointer/subscript notation.
• A parameter of type const char * typically represents a constant string.
Section 7.10 Arrays of Pointers
• Arrays may contain pointers. A common use of an array of pointers is to form an array of strings.
Each entry in the array is a string, but in C a string is essentially a pointer to its first character.
So, each entry in an array of strings is actually a pointer to the first character of a string.
Section 7.12 Pointers to Functions
• A pointer to a function contains the address of the function in memory. A function name is really
the starting address in memory of the code that performs the function’s task.
• Pointers to functions can be passed to functions, returned from functions, stored in arrays and
assigned to other function pointers.
• A pointer to a function is dereferenced to call the function. A function pointer can be used directly as the function name when calling the function.
• A common use of function pointers is in text-based, menu-driven systems.
Terminology
address operator (&) 279
array of pointers 303
array of strings 303
const qualifier 284
constant pointer to constant data 287
constant pointer to non-constant data 287
dereferencing a pointer 280
dereferencing operator (*) 280
function pointer 313
indefinite postponement 305
indirection 278
indirection operator (*) 280
non-constant pointer to constant data 287
non-constant pointer to non-constant data 287
offset to a pointer 300
pass-by-reference 282
pass-by-value 282
pointer 278
pointer arithmetic 297
pointer/offset notation 300
pointer/subscript notation 300
pointer subscripting 300
pointer to a function 309
pointer to void (void *) 299
principle of least privilege 284
record 289
sizeof operator 294
string array 304
structure 289
void * (pointer to void) 299
318
Chapter 7 C Pointers
Self-Review Exercises
7.1
7.2
Answer each of the following:
a) A pointer variable contains as its value the
of another variable.
b) The three values that can be used to initialize a pointer are
,
.
c) The only integer that can be assigned to a pointer is
.
and
State whether the following are true or false. If the answer is false, explain why.
a) The address operator (&) can be applied only to constants, to expressions and to variables declared with the storage-class register.
b) A pointer that’s declared to be void can be dereferenced.
c) Pointers of different types may not be assigned to one another without a cast operation.
7.3
Answer each of the following. Assume that single-precision floating-point numbers are
stored in 4 bytes, and that the starting address of the array is at location 1002500 in memory. Each
part of the exercise should use the results of previous parts where appropriate.
a) Define an array of type float called numbers with 10 elements, and initialize the elements to the values 0.0, 1.1, 2.2, …, 9.9. Assume the symbolic constant SIZE has been
defined as 10.
b) Define a pointer, nPtr, that points to an object of type float.
c) Print the elements of array numbers using array subscript notation. Use a for statement
and assume the integer control variable i has been defined. Print each number with 1
position of precision to the right of the decimal point.
d) Give two separate statements that assign the starting address of array numbers to the
pointer variable nPtr.
e) Print the elements of array numbers using pointer/offset notation with the pointer nPtr.
f) Print the elements of array numbers using pointer/offset notation with the array name
as the pointer.
g) Print the elements of array numbers by subscripting pointer nPtr.
h) Refer to element 4 of array numbers using array subscript notation, pointer/offset notation with the array name as the pointer, pointer subscript notation with nPtr and
pointer/offset notation with nPtr.
i) Assuming that nPtr points to the beginning of array numbers, what address is referenced
by nPtr + 8? What value is stored at that location?
j) Assuming that nPtr points to numbers[5], what address is referenced by nPtr –= 4?
What’s the value stored at that location?
7.4
For each of the following, write a statement that performs the indicated task. Assume that
floating-point variables number1 and number2 are defined and that number1 is initialized to 7.3.
a) Define the variable fPtr to be a pointer to an object of type float.
b) Assign the address of variable number1 to pointer variable fPtr.
c) Print the value of the object pointed to by fPtr.
d) Assign the value of the object pointed to by fPtr to variable number2.
e) Print the value of number2.
f) Print the address of number1. Use the %p conversion specifier.
g) Print the address stored in fPtr. Use the %p conversion specifier. Is the value printed the
same as the address of number1?
7.5
Do each of the following:
a) Write the function header for a function called exchange that takes two pointers to
floating-point numbers x and y as parameters and does not return a value.
b) Write the function prototype for the function in part (a).
Answers to Self-Review Exercises
319
c) Write the function header for a function called evaluate that returns an integer and
that takes as parameters integer x and a pointer to function poly. Function poly takes
an integer parameter and returns an integer.
d) Write the function prototype for the function in part (c).
7.6
Find the error in each of the following program segments. Assume
int *zPtr; // zPtr will reference array z
int *aPtr = NULL;
void *sPtr = NULL;
int number, i;
int z[ 5 ] = { 1, 2, 3, 4, 5 };
sPtr = z;
a)
b)
++zptr;
// use pointer to get first value of array; assume zPtr is initialized
number = zPtr;
c)
// assign array element 2 (the value 3) to number;
assume zPtr is initialized
number = *zPtr[ 2 ];
d)
// print entire array z; assume zPtr is initialized
for ( i = 0; i <= 5; ++i ) {
printf( "%d ", zPtr[ i ] );
}
e)
// assign the value pointed to by sPtr to number
number = *sPtr;
f)
++z;
Answers to Self-Review Exercises
7.1
a) address. b) 0, NULL, an address. c) 0.
7.2
a) False. The address operator can be applied only to variables. The address operator cannot be applied to variables declared with storage class register. b) False. A pointer to void cannot
be dereferenced, because there’s no way to know exactly how many bytes of memory to dereference.
c) False. Pointers of type void can be assigned pointers of other types, and pointers of type void can
be assigned to pointers of other types.
7.3
a)
b)
c)
float numbers[ SIZE ] = { 0.0, 1.1, 2.2, 3.3, 4.4, 5.5, 6.6, 7.7, 8.8, 9.9 };
float *nPtr;
for ( i = 0; i < SIZE; ++i ) {
printf( "%.1f ", numbers[ i ] );
}
d)
nPtr = numbers;
nPtr = &numbers[ 0 ];
e)
for ( i = 0; i < SIZE; ++i ) {
printf( "%.1f ", *( nPtr + i ) );
}
f)
for ( i = 0; i < SIZE; ++i ) {
printf( "%.1f ", *( numbers + i ) );
}
g)
for ( i = 0; i < SIZE; ++i ) {
printf( "%.1f ", nPtr[ i ] );
}
320
Chapter 7 C Pointers
h)
numbers[ 4 ]
*( numbers + 4 )
nPtr[ 4 ]
*( nPtr + 4 )
i) The address is 1002500 + 8 * 4 = 1002532. The value is 8.8.
j) The address of numbers[ 5 ] is 1002500 + 5 * 4 = 1002520.
The address of nPtr -= 4 is 1002520 - 4 * 4 = 1002504.
The value at that location is 1.1.
7.4
a)
b)
c)
d)
e)
f)
g)
float *fPtr;
fPtr = &number1;
printf( "The value of *fPtr is %f\n", *fPtr );
number2 = *fPtr;
printf( "The value of number2 is %f\n", number2 );
printf( "The address of number1 is %p\n", &number1 );
printf( "The address stored in fptr is %p\n", fPtr );
Yes, the value is the same.
7.5
7.6
a)
b)
c)
d)
void exchange( float *x, float *y )
void exchange( float *x, float *y );
int evaluate( int x, int (*poly)( int ) )
int evaluate( int x, int (*poly)( int ) );
a) Error: zPtr has not been initialized.
Correction: Initialize zPtr with zPtr = z; before performing the pointer arithmetic.
b) Error: The pointer is not dereferenced.
Correction: Change the statement to number = *zPtr;
c) Error: zPtr[ 2 ] is not a pointer and should not be dereferenced.
Correction: Change *zPtr[ 2 ] to zPtr[ 2 ].
d) Error: Referring to an array element outside the array bounds with pointer subscripting.
Correction: Change the operator <= in the for condition to <.
e) Error: Dereferencing a void pointer.
Correction: To dereference the pointer, it must first be cast to an integer pointer.
Change the statement to number = *( ( int * ) sPtr );
f) Error: Trying to modify an array name with pointer arithmetic.
Correction: Use a pointer variable instead of the array name to accomplish pointer
arithmetic, or subscript the array name to refer to a specific element.
Exercises
7.7
Answer each of the following:
a) The
operator returns the location in memory where its operand is stored.
b) The
operator returns the value of the object to which its operand points.
c) To simulate pass-by-reference when passing a nonarray variable to a function, it’s necof the variable to the function.
essary to pass the
7.8
State whether the following are true or false. If false, explain why.
a) Two pointers that point to different arrays cannot be compared meaningfully.
b) Because the name of an array is a pointer to the first element of the array, array names
may be manipulated in precisely the same manner as pointers.
7.9
Answer each of the following. Assume that unsigned integers are stored in 2 bytes and that
the starting address of the array is at location 1002500 in memory.
Exercises
321
a) Define an array of type unsigned int called values with five elements, and initialize the
elements to the even integers from 2 to 10. Assume the symbolic constant SIZE has been
defined as 5.
b) Define a pointer vPtr that points to an object of type unsigned int.
c) Print the elements of array values using array subscript notation. Use a for statement
and assume integer control variable i has been defined.
d) Give two separate statements that assign the starting address of array values to pointer
variable vPtr.
e) Print the elements of array values using pointer/offset notation.
f) Print the elements of array values using pointer/offset notation with the array name as
the pointer.
g) Print the elements of array values by subscripting the pointer to the array.
h) Refer to element 5 of array values using array subscript notation, pointer/offset notation with the array name as the pointer, pointer subscript notation, and pointer/offset
notation.
i) What address is referenced by vPtr + 3? What value is stored at that location?
j) Assuming vPtr points to values[4], what address is referenced by vPtr -= 4? What value is stored at that location?
7.10 For each of the following, write a single statement that performs the indicated task. Assume
that long integer variables value1 and value2 have been defined and that value1 has been initialized
to 200000.
a) Define the variable lPtr to be a pointer to an object of type long.
b) Assign the address of variable value1 to pointer variable lPtr.
c) Print the value of the object pointed to by lPtr.
d) Assign the value of the object pointed to by lPtr to variable value2.
e) Print the value of value2.
f) Print the address of value1.
g) Print the address stored in lPtr. Is the value printed the same as the address of value1?
7.11
Do each of the following:
a) Write the function header for function zero, which takes a long integer array parameter
bigIntegers and does not return a value.
b) Write the function prototype for the function in part (a).
c) Write the function header for function add1AndSum, which takes an integer array parameter oneTooSmall and returns an integer.
d) Write the function prototype for the function described in part (c).
Note: Exercises 7.12–7.15 are reasonably challenging. Once you have done these problems,
you ought to be able to implement most popular card games easily.
7.12 (Card Shuffling and Dealing) Modify the program in Fig. 7.24 so that the card-dealing
function deals a five-card poker hand. Then write the following additional functions:
a) Determine whether the hand contains a pair.
b) Determine whether the hand contains two pairs.
c) Determine whether the hand contains three of a kind (e.g., three jacks).
d) Determine whether the hand contains four of a kind (e.g., four aces).
e) Determine whether the hand contains a flush (i.e., all five cards of the same suit).
f) Determine whether the hand contains a straight (i.e., five cards of consecutive face values).
7.13 (Project: Card Shuffling and Dealing) Use the functions developed in Exercise 7.12 to
write a program that deals two five-card poker hands, evaluates each, and determines which is the
better hand.
322
Chapter 7 C Pointers
7.14 (Project: Card Shuffling and Dealing) Modify the program developed in Exercise 7.13 so
that it can simulate the dealer. The dealer’s five-card hand is dealt “face down” so the player cannot
see it. The program should then evaluate the dealer’s hand, and based on the quality of the hand,
the dealer should draw one, two or three more cards to replace the corresponding number of unneeded cards in the original hand. The program should then reevaluate the dealer’s hand. [Caution:
This is a difficult problem!]
7.15 (Project: Card Shuffling and Dealing) Modify the program developed in Exercise 7.14 so
that it can handle the dealer’s hand automatically, but the player is allowed to decide which cards
of the player's hand to replace. The program should then evaluate both hands and determine who
wins. Now use this new program to play 20 games against the computer. Who wins more games,
you or the computer? Have one of your friends play 20 games against the computer. Who wins more
games? Based on the results of these games, make appropriate modifications to refine your pokerplaying program (this, too, is a difficult problem). Play 20 more games. Does your modified program play a better game?
7.16 (Card Shuffling and Dealing Modification) In the card shuffling and dealing program of
Fig. 7.24, we intentionally used an inefficient shuffling algorithm that introduced the possibility of
indefinite postponement. In this problem, you’ll create a high-performance shuffling algorithm that
avoids indefinite postponement.
Modify the program of Fig. 7.24 as follows. Begin by initializing the deck array as shown in
Fig. 7.29. Modify the shuffle function to loop row-by-row and column-by-column through the
array, touching every element once. Each element should be swapped with a randomly selected element of the array. Print the resulting array to determine whether the deck is satisfactorily shuffled
(as in Fig. 7.30, for example). You may want your program to call the shuffle function several
times to ensure a satisfactory shuffle.
Unshuffled deck array
0
1
2
3
0
1
2
3
4
5
6
7
8
9
10
11
12
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
Fig. 7.29 | Unshuffled deck array.
Sample shuffled deck array
0
1
2
3
0
1
2
3
4
5
6
7
8
9
10
11
12
19
40
27
25
36
46
10
34
35
41
18
2
44
13
28
14
16
21
30
8
11
31
17
24
7
1
12
33
15
42
43
23
45
3
29
32
4
47
26
50
38
52
39
48
51
9
5
37
49
22
6
20
Fig. 7.30 | Sample shuffled deck array.
Although the approach in this problem improves the shuffling algorithm, the dealing algorithm
still requires searching the deck array for card 1, then card 2, then card 3, and so on. Worse yet, even
Exercises
323
after the dealing algorithm locates and deals the card, the algorithm still searches through the remainder of the deck. Modify the program of Fig. 7.24 so that once a card is dealt, no further attempts are
made to match that card number, and the program immediately proceeds with dealing the next card.
In Chapter 10, we develop a dealing algorithm that requires only one operation per card.
7.17 (Simulation: The Tortoise and the Hare) In this problem, you’ll recreate one of the truly
great moments in history, namely the classic race of the tortoise and the hare. You’ll use random
number generation to develop a simulation of this memorable event.
Our contenders begin the race at “square 1” of 70 squares. Each square represents a possible
position along the race course. The finish line is at square 70. The first contender to reach or pass
square 70 is rewarded with a pail of fresh carrots and lettuce. The course weaves its way up the side
of a slippery mountain, so occasionally the contenders lose ground.
There’s a clock that ticks once per second. With each tick of the clock, your program should
adjust the position of the animals according to the rules of Fig. 7.31.
Animal
Move type
Percentage of the time
Actual move
Tortoise
Fast plod
Slip
Slow plod
Sleep
Big hop
Big slip
Small hop
Small slip
50%
20%
30%
20%
20%
10%
30%
20%
3 squares to the right
6 squares to the left
1 square to the right
No move at all
9 squares to the right
12 squares to the left
1 square to the right
2 squares to the left
Hare
Fig. 7.31 | Tortoise and hare rules for adjusting positions.
Use variables to keep track of the positions of the animals (i.e., position numbers are 1–70).
Start each animal at position 1 (i.e., the “starting gate”). If an animal slips left before square 1,
move the animal back to square 1. Generate the percentages in the preceding table by producing a
random integer, i, in the range 1 ≤ i ≤ 10. For the tortoise, perform a “fast plod” when 1 ≤ i ≤ 5, a
“slip” when 6 ≤ i ≤ 7, or a “slow plod” when 8 ≤ i ≤ 10. Use a similar technique to move the hare.
Begin the race by printing
BANG !!!!!
AND THEY'RE OFF !!!!!
Then, for each tick of the clock (i.e., each repetition of a loop), print a 70-position line showing
the letter T in the position of the tortoise and the letter H in the position of the hare. Occasionally,
the contenders will land on the same square. In this case, the tortoise bites the hare and your program should print OUCH!!! beginning at that position. All print positions other than the T, the H, or
the OUCH!!! (in case of a tie) should be blank.
After each line is printed, test whether either animal has reached or passed square 70. If so,
then print the winner and terminate the simulation. If the tortoise wins, print TORTOISE WINS!!!
YAY!!! If the hare wins, print Hare wins. Yuch. If both animals win on the same tick of the clock,
you may want to favor the turtle (the “underdog”), or you may want to print It's a tie. If neither
animal wins, perform the loop again to simulate the next tick of the clock. When you’re ready to
run your program, assemble a group of fans to watch the race. You’ll be amazed at how involved
your audience gets!
324
Chapter 7 C Pointers
7.18 (Card Shuffling and Dealing Modification) Modify the card shuffling and dealing program
of Fig. 7.24 so the shuffling and dealing operations are performed by the same function (shuffleAndDeal). The function should contain one nested looping structure that’s similar to function shuffle in Fig. 7.24.
7.19
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
// ex07_19.c
// What does this program do?
#include <stdio.h>
#define SIZE 80
void mystery1( char *s1, const char *s2 ); // prototype
int main( void )
{
char string1[ SIZE ]; // create char array
char string2[ SIZE ]; // create char array
puts( "Enter two strings: " );
scanf( "%79s%79s" , string1, string2 );
mystery1( string1, string2 );
printf("%s", string1 );
} // end main
// What does this function do?
void mystery1( char *s1, const char *s2 )
{
while ( *s1 != '\0' ) {
++s1;
} // end while
for ( ; *s1 = *s2; ++s1, ++s2 ) {
; // empty statement
} // end for
} // end function mystery1
7.20
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
What does this program do, assuming that the user enters two strings of the same length?
What does this program do?
// ex07_20.c
// what does this program do?
#include <stdio.h>
#define SIZE 80
int mystery2( const char *s ); // prototype
int main( void )
{
char string[ SIZE ]; // create char array
puts( "Enter a string: ");
scanf( "%79s", string );
printf( "%d\n", mystery2( string ) );
} // end main
// What does this function do?
int mystery2( const char *s )
{
Exercises
325
int x; // counter
20
21
22
23
24
25
26
27
28
// loop through string
for ( x = 0; *s != '\0'; ++s ) {
++x;
} // end for
return x;
} // end function mystery2
7.21 Find the error in each of the following program segments. If the error can be corrected, explain how.
a) int *number;
printf( "%d\n", *number );
b)
float *realPtr;
long *integerPtr;
integerPtr = realPtr;
c)
int * x, y;
x = y;
d)
char s[] = "this is a character array";
int count;
for ( ; *s != '\0'; ++s)
printf( "%c ", *s );
e)
short *numPtr, result;
void *genericPtr = numPtr;
result = *genericPtr + 7;
f)
float x = 19.34;
float xPtr = &x;
printf( "%f\n", xPtr );
g)
char *s;
printf( "%s\n", s );
(Maze Traversal) The following grid is a double-subscripted array representation of a maze.
7.22
#
#
.
#
#
#
#
#
#
#
#
#
#
.
.
#
.
#
.
#
.
#
.
#
#
.
#
#
.
#
.
.
.
#
.
#
#
.
.
.
.
#
#
#
.
#
.
#
#
#
#
#
.
.
.
.
.
#
.
#
#
.
.
.
#
#
#
#
.
#
.
#
#
.
#
.
#
.
.
.
.
.
.
#
#
.
#
.
#
#
#
#
.
#
#
#
#
.
#
.
.
.
.
.
.
#
.
#
#
.
#
#
#
#
#
#
#
#
.
#
#
.
.
.
.
.
.
.
.
.
.
#
#
#
#
#
.
#
#
#
#
#
#
#
The # symbols represent the walls of the maze, and the periods (.) represent squares in the possible
paths through the maze.
There’s a simple algorithm for walking through a maze that guarantees finding the exit
(assuming there’s an exit). If there’s not an exit, you’ll arrive at the starting location again. Place
your right hand on the wall to your right and begin walking forward. Never remove your hand
from the wall. If the maze turns to the right, you follow the wall to the right. As long as you do not
remove your hand from the wall, eventually you’ll arrive at the exit of the maze. There may be a
shorter path than the one you have taken, but you’re guaranteed to get out of the maze.
326
Chapter 7 C Pointers
Write recursive function mazeTraverse to walk through the maze. The function should
receive as arguments a 12-by-12 character array representing the maze and the starting location of
the maze. As mazeTraverse attempts to locate the exit from the maze, it should place the character
X in each square in the path. The function should display the maze after each move so the user can
watch as the maze is solved.
7.23 (Generating Mazes Randomly) Write a function mazeGenerator that takes as an argument
a double-subscripted 12-by-12 character array and randomly produces a maze. The function should
also provide the starting and ending locations of the maze. Try your function mazeTraverse from
Exercise 7.22 using several randomly generated mazes.
7.24 (Mazes of Any Size) Generalize functions mazeTraverse and
Exercises 7.22–7.23 to process mazes of any width and height.
mazeGenerator
of
7.25 (Arrays of Pointers to Functions) Rewrite the program of Fig. 6.22 to use a menu-driven
interface. The program should offer the user four options as follows:
Enter a choice:
0 Print the array of grades
1 Find the minimum grade
2 Find the maximum grade
3 Print the average on all tests for each student
4 End program
One restriction on using arrays of pointers to functions is that all the pointers must have the same
type. The pointers must be to functions of the same return type that receive arguments of the same
type. For this reason, the functions in Fig. 6.22 must be modified so that they each return the same
type and take the same parameters. Modify functions minimum and maximum to print the minimum
or maximum value and return nothing. For option 3, modify function average of Fig. 6.22 to output the average for each student (not a specific student). Function average should return nothing
and take the same parameters as printArray, minimum and maximum. Store the pointers to the four
functions in array processGrades and use the choice made by the user as the subscript into the
array for calling each function.
7.26
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
What does this program do, assuming that the user enters two strings of the same length?
// ex07_26.c
// What does this program do?
#include <stdio.h>
#define SIZE 80
int mystery3( const char *s1, const char *s2 ); // prototype
int main( void )
{
char string1[ SIZE ]; // create char array
char string2[ SIZE ]; // create char array
puts( "Enter two strings: " );
scanf( "%79s%79s", string1 , string2 );
printf( "The result is %d\n", mystery3( string1, string2 ) );
} // end main
int mystery3( const char *s1, const char *s2 )
{
int result = 1;
Special Section: Building Your Own Computer
21
22
23
24
25
26
27
28
29
327
for ( ; *s1 != '\0' && *s2 != '\0'; ++s1, ++s2 ) {
if ( *s1 != *s2 ) {
result = 0;
} // end if
} // end for
return result;
} // end function mystery3
Special Section: Building Your Own Computer
In the next several exercises, we take a temporary diversion away from the world of high-level language programming. We “peel open” a computer and look at its internal structure. We introduce
machine-language programming and write several machine-language programs. To make this an
especially valuable experience, we then build a computer (through the technique of software-based
simulation) on which you can execute your machine-language programs!
7.27 (Machine-Language Programming) Let’s create a computer we’ll call the Simpletron. As its
name implies, it’s a simple machine, but as we’ll soon see, it’s a powerful one as well. The Simpletron
runs programs written in the only language it directly understands—that is, Simpletron Machine
Language, or SML for short.
The Simpletron contains an accumulator—a “special register” in which information is put
before the Simpletron uses that information in calculations or examines it in various ways. All
information in the Simpletron is handled in terms of words. A word is a signed four-digit decimal
number such as +3364, -1293, +0007, -0001 and so on. The Simpletron is equipped with a 100word memory, and these words are referenced by their location numbers 00, 01, …, 99.
Before running an SML program, we must load or place the program into memory. The first
instruction (or statement) of every SML program is always placed in location 00.
Each instruction written in SML occupies one word of the Simpletron’s memory, so instructions
are signed four-digit decimal numbers. We assume that the sign of an SML instruction is always plus,
but the sign of a data word may be either plus or minus. Each location in the Simpletron’s memory
may contain either an instruction, a data value used by a program or an unused (and hence undefined) area of memory. The first two digits of each SML instruction are the operation code, which
specifies the operation to be performed. SML operation codes are summarized in Fig. 7.32.
Operation code
Meaning
Input/output operations:
#define READ 10
#define WRITE 11
Read a word from the terminal into a specific location in
memory.
Write a word from a specific location in memory to the terminal.
Load/store operations:
#define LOAD 20
#define STORE 21
Load a word from a specific location in memory into the
accumulator.
Store a word from the accumulator into a specific location in
memory.
Fig. 7.32 | Simpletron Machine Language (SML) operation codes. (Part 1 of 2.)
328
Chapter 7 C Pointers
Operation code
Meaning
Arithmetic operations:
Add a word from a specific location in memory to the word in
the accumulator (leave result in accumulator).
Subtract a word from a specific location in memory from the
word in the accumulator (leave result in accumulator).
Divide a word from a specific location in memory into the word
in the accumulator (leave result in accumulator).
Multiply a word from a specific location in memory by the word
in the accumulator (leave result in accumulator).
#define ADD 30
#define SUBTRACT 31
#define DIVIDE 32
#define MULTIPLY 33
Transfer-of-control operations:
Branch to a specific location in memory.
Branch to a specific location in memory if the accumulator is
negative.
Branch to a specific location in memory if the accumulator is zero.
Halt—i.e., the program has completed its task.
#define BRANCH 40
#define BRANCHNEG 41
#define BRANCHZERO 42
#define HALT 43
Fig. 7.32 | Simpletron Machine Language (SML) operation codes. (Part 2 of 2.)
The last two digits of an SML instruction are the operand, which is the address of the memory
location containing the word to which the operation applies. Now let’s consider several simple
SML programs. The following SML program reads two numbers from the keyboard, and computes and prints their sum.
The instruction +1007 reads the first number from the keyboard and places it into location 07
(which has been initialized to zero). Then +1008 reads the next number into location 08. The load
instruction, +2007, puts the first number into the accumulator, and the add instruction, +3008,
adds the second number to the number in the accumulator. All SML arithmetic instructions leave
their results in the accumulator. The store instruction, +2109, places the result back into memory
location 09, from which the write instruction, +1109, takes the number and prints it (as a signed
four-digit decimal number). The halt instruction, +4300, terminates execution.
Example 1
Location
Number
Instruction
00
+1007
01
+1008
02
+2007
03
+3008
04
+2109
05
+1109
06
+4300
07
+0000
08
+0000
09
+0000
(Read A)
(Read B)
(Load A)
(Add B)
(Store C)
(Write C)
(Halt)
(Variable A)
(Variable B)
(Result C)
Special Section: Building Your Own Computer
329
The following SML program reads two numbers from the keyboard, and determines and
prints the larger value. Note the use of the instruction +4107 as a conditional transfer of control,
much the same as C’s if statement.
Example 2
Location
Number
Instruction
00
+1009
01
+1010
02
+2009
03
+3110
04
+4107
05
+1109
06
+4300
07
+1110
08
+4300
09
+0000
10
+0000
(Read A)
(Read B)
(Load A)
(Subtract B)
(Branch negative to 07)
(Write A)
(Halt)
(Write B)
(Halt)
(Variable A)
(Variable B)
Now write SML programs to accomplish each of the following tasks.
a) Use a sentinel-controlled loop to read positive integers and compute and print their
sum.
b) Use a counter-controlled loop to read seven numbers, some positive and some negative,
and compute and print their average.
c) Read a series of numbers and determine and print the largest number. The first number
read indicates how many numbers should be processed.
7.28 (A Computer Simulator) It may at first seem outrageous, but in this problem you’re going to
build your own computer. No, you won’t be soldering components together. Rather, you’ll use the
powerful technique of software-based simulation to create a software model of the Simpletron. You’ll not
be disappointed. Your Simpletron simulator will turn the computer you’re using into a Simpletron,
and you’ll actually be able to run, test and debug the SML programs you wrote in Exercise 7.27.
When you run your Simpletron simulator, it should begin by printing:
***
***
***
***
***
***
***
Welcome to Simpletron! ***
Please enter your program one instruction
(or data word) at a time. I will type the
location number and a question mark (?).
You then type the word for that location.
Type the sentinel -99999 to stop entering
your program. ***
***
***
***
***
***
Simulate the memory of the Simpletron with a single-subscripted array memory that has 100
elements. Now assume that the simulator is running, and let’s examine the dialog as we enter the
program of Example 2 of Exercise 7.27:
00
01
02
03
04
05
06
?
?
?
?
?
?
?
+1009
+1010
+2009
+3110
+4107
+1109
+4300
330
Chapter 7 C Pointers
07 ? +1110
08 ? +4300
09 ? +0000
10 ? +0000
11 ? -99999
*** Program loading completed ***
*** Program execution begins ***
The SML program has now been placed (or loaded) into the array memory. Now the Simpletron executes the SML program. It begins with the instruction in location 00 and continues
sequentially, unless directed to some other part of the program by a transfer of control.
Use the variable accumulator to represent the accumulator register. Use the variable instructionCounter to keep track of the location in memory that contains the instruction being performed. Use the variable operationCode to indicate the operation currently being performed—i.e.,
the left two digits of the instruction word. Use the variable operand to indicate the memory location on which the current instruction operates. Thus, if an instrucion has an operand, it’s the rightmost two digits of the instruction currently being performed. Do not execute instructions directly
from memory. Rather, transfer the next instruction to be performed from memory to a variable
called instructionRegister. Then “pick off ” the left two digits and place them in the variable
operationCode, and “pick off ” the right two digits and place them in operand.
When Simpletron begins execution, the special registers are initialized as follows:
accumulator
instructionCounter
instructionRegister
operationCode
operand
+0000
00
+0000
00
00
Now let’s “walk through” the execution of the first SML instruction, +1009 in memory location 00. This is called an instruction execution cycle.
The instructionCounter tells us the location of the next instruction to be performed. We
fetch the contents of that location from memory by using the C statement
instructionRegister = memory[ instructionCounter ];
The operation code and the operand are extracted from the instruction register by the statements
operationCode = instructionRegister / 100;
operand = instructionRegister % 100;
Now the Simpletron must determine that the operation code is actually a read (versus a write,
a load, and so on). A switch differentiates among the twelve operations of SML.
The switch statement simulates the behavior of various SML instructions as follows (we leave
the others to the reader):
scanf( "%d", &memory[ operand ] );
read:
load:
accumulator = memory[ operand ];
add:
accumulator += memory[ operand ];
Various branch instructions: We’ll discuss these shortly.
halt:
This instruction prints the message
*** Simpletron execution terminated ***
then prints the name and contents of each register as well as the complete contents of memory. Such
a printout is often called a computer dump. To help you program your dump function, a sample dump
format is shown in Fig. 7.33. A dump after executing a Simpletron program would show the actual
values of instructions and data values at the moment execution terminated. You can print leading 0s
in front of an integer that is shorter than its field width by placing the 0 formatting flag before the
field width in the format specifier as in "%02d". You can place a + or - sign before a value with the +
formatting flag. So to produce a number of the form +0000, you can use the format specifier "%+05d".
Special Section: Building Your Own Computer
REGISTERS:
accumulator
instructionCounter
instructionRegister
operationCode
operand
331
+0000
00
+0000
00
00
MEMORY:
0
10
20
30
40
50
60
70
80
90
0
+0000
+0000
+0000
+0000
+0000
+0000
+0000
+0000
+0000
+0000
1
+0000
+0000
+0000
+0000
+0000
+0000
+0000
+0000
+0000
+0000
2
+0000
+0000
+0000
+0000
+0000
+0000
+0000
+0000
+0000
+0000
3
+0000
+0000
+0000
+0000
+0000
+0000
+0000
+0000
+0000
+0000
4
+0000
+0000
+0000
+0000
+0000
+0000
+0000
+0000
+0000
+0000
5
+0000
+0000
+0000
+0000
+0000
+0000
+0000
+0000
+0000
+0000
6
+0000
+0000
+0000
+0000
+0000
+0000
+0000
+0000
+0000
+0000
7
+0000
+0000
+0000
+0000
+0000
+0000
+0000
+0000
+0000
+0000
8
+0000
+0000
+0000
+0000
+0000
+0000
+0000
+0000
+0000
+0000
9
+0000
+0000
+0000
+0000
+0000
+0000
+0000
+0000
+0000
+0000
Fig. 7.33 | Sample Simpletron dump format.
Let’s proceed with the execution of our program’s first instruction, namely the +1009 in location 00. As we’ve indicated, the switch statement simulates this by performing the C statement
scanf( "%d", &memory[ operand ] );
A question mark (?) should be displayed on the screen before the scanf is executed to prompt
the user for input. The Simpletron waits for the user to type a value and then press the Return key.
The value is then read into location 09.
At this point, simulation of the first instruction is completed. All that remains is to prepare
the Simpletron to execute the next instruction. Because the instruction just performed was not a
transfer of control, we need merely increment the instruction counter register as follows:
++instructionCounter;
This completes the simulated execution of the first instruction. The entire process (i.e., the
instruction execution cycle) begins anew with the fetch of the next instruction to be executed.
Now let’s consider how the branching instructions—the transfers of control—are simulated.
All we need to do is adjust the value in the instruction counter appropriately. Therefore, the
unconditional branch instruction (40) is simulated within the switch as
instructionCounter = operand;
The conditional “branch if accumulator is zero” instruction is simulated as
if ( accumulator == 0 ) {
instructionCounter = operand;
}
At this point, you should implement your Simpletron simulator and run the SML programs
you wrote in Exercise 7.27. You may embellish SML with additional features and provide for these
in your simulator.
Your simulator should check for various types of errors. During the program loading phase,
for example, each number the user types into the Simpletron’s memory must be in the range -9999
to +9999. Your simulator should use a while loop to test that each number entered is in this range,
and, if not, keep prompting the user to reenter the number until a correct number is entered.
During the execution phase, your simulator should check for serious errors, such as attempts
to divide by zero, attempts to execute invalid operation codes and accumulator overflows (i.e.,
332
Chapter 7 C Pointers
arithmetic operations resulting in values larger than +9999 or smaller than -9999). Such serious
errors are called fatal errors. When a fatal error is detected, print an error message such as:
*** Attempt to divide by zero ***
*** Simpletron execution abnormally terminated ***
and should print a full computer dump in the format we’ve discussed previously. This will help the
user locate the error in the program.
Implementation Note: When you implement the Simpletron Simulator, define the memory
array and all the registers as variables in main. The program should contain three other functions—
load, execute and dump. Function load reads the SML instructions from the user at the keyboard.
(Once you study file processing in Chapter 11, you’ll be able to read the SML instruction from a
file.) Function execute executes the SML program currently loaded in the memory array. Function
dump displays the contents of memory and all of the registers stored in main’s variables. Pass the memory array and registers to the other functions as necessary to complete their tasks. Functions load
and execute need to modify variables that are defined in main, so you’ll need to pass those variables
to the functions by reference using pointers. So, you’ll need to modify the statements we showed
throughout this problem description to use the appropriate pointer notations.
7.29 (Modifications to the Simpletron Simulator) In Exercise 7.28, you wrote a software simulation of a computer that executes programs written in Simpletron Machine Language (SML). In
this exercise, we propose several modifications and enhancements to the Simpletron Simulator. In
Exercises 12.26 and 12.27, we propose building a compiler that converts programs written in a
high-level programming language (a variation of BASIC) to Simpletron Machine Language. Some
of the following modifications and enhancements may be required to execute the programs produced by the compiler.
a) Extend the Simpletron Simulator’s memory to contain 1000 memory locations to enable the Simpletron to handle larger programs.
b) Allow the simulator to perform remainder calculations. This requires an additional
Simpletron Machine Language instruction.
c) Allow the simulator to perform exponentiation calculations. This requires an additional
Simpletron Machine Language instruction.
d) Modify the simulator to use hexadecimal values rather than integer values to represent
Simpletron Machine Language instructions.
e) Modify the simulator to allow output of a newline. This requires an additional Simpletron Machine Language instruction.
f) Modify the simulator to process floating-point values in addition to integer values.
g) Modify the simulator to handle string input. [Hint: Each Simpletron word can be divided into two groups, each holding a two-digit integer. Each two-digit integer represents the ASCII decimal equivalent of a character. Add a machine-language instruction
that will input a string and store it beginning at a specific Simpletron memory location.
The first half of the word at that location will be a count of the number of characters in
the string (i.e., the length of the string). Each succeeding half word contains one ASCII
character expressed as two decimal digits. The machine-language instruction converts
each character into its ASCII equivalent and assigns it to a half word.]
h) Modify the simulator to handle output of strings stored in the format of part (g). [Hint:
Add a machine-language instruction that prints a string beginning at a specified Simpletron memory location. The first half of the word at that location is the length of the
string in characters. Each succeeding half word contains one ASCII character expressed
as two decimal digits. The machine-language instruction checks the length and prints
the string by translating each two-digit number into its equivalent character.]
Array of Function Pointer Exercises
333
Array of Function Pointer Exercises
7.30 (Calculating Circle Circumference, Circle Area or Sphere Volume Using Function Pointers)
Using the techniques you learned in Fig. 7.28, create a text-based, menu-driven program that allows
the user to choose whether to calculate the circumference of a circle, the area of a circle or the volume of a sphere. The program should then input a radius from the user, perform the appropriate
calculation and display the result. Use an array of function pointers in which each pointer represents
a function that returns void and receives a double parameter. The corresponding functions should
each display messages indicating which calculation was performed, the value of the radius and the
result of the calculation.
7.31 (Calculator Using Function Pointers) Using the techniques you learned in Fig. 7.28, create
a text-based, menu-driven program that allows the user to choose whether to add, subtract, multiply
or divide two numbers. The program should then input two double values from the user, perform
the appropriate calculation and display the result. Use an array of function pointers in which each
pointer represents a function that returns void and receives two double parameters. The corresponding functions should each display messages indicating which calculation was performed, the
values of the parameters and the result of the calculation.
Making a Difference
7.32 (Polling) The Internet and the web are enabling more people to network, join a cause, voice
opinions, and so on. The U.S. presidential candidates in 2008 used the Internet intensively to get
out their messages and raise money for their campaigns. In this exercise, you’ll write a simple polling
program that allows users to rate five social-consciousness issues from 1 (least important) to 10
(most important). Pick five causes that are important to you (e.g., political issues, global environmental issues). Use a one-dimensional array topics (of type char *) to store the five causes. To summarize the survey responses, use a 5-row, 10-column two-dimensional array responses (of type
int), each row corresponding to an element in the topics array. When the program runs, it should
ask the user to rate each issue. Have your friends and family respond to the survey. Then have the
program display a summary of the results, including:
a) A tabular report with the five topics down the left side and the 10 ratings across the top,
listing in each column the number of ratings received for each topic.
b) To the right of each row, show the average of the ratings for that issue.
c) Which issue received the highest point total? Display both the issue and the point total.
d) Which issue received the lowest point total? Display both the issue and the point total.
7.33 (Carbon Footprint Calculator: Arrays of Function Pointers) Using arrays of function pointers, as you learned in this chapter, you can specify a set of functions that are called with the same
types of arguments and return the same type of data. Governments and companies worldwide are
becoming increasingly concerned with carbon footprints (annual releases of carbon dioxide into the
atmosphere) from buildings burning various types of fuels for heat, vehicles burning fuels for power,
and the like. Many scientists blame these greenhouse gases for the phenomenon called global warming. Create three functions that help calculate the carbon footprint of a building, a car and a bicycle,
respectively. Each function should input appropriate data from the user, then calculate and display
the carbon footprint. (Check out a few websites that explain how to calculate carbon footprints.)
Each function should receive no parameters and return void. Write a program that prompts the user
to enter the type of carbon footprint to calculate, then calls the corresponding function in the array
of function pointers. For each type of carbon footprint, display some identifying information and
the object’s carbon footprint.
8
Vigorous writing is concise. A
sentence should contain no
unnecessary words, a paragraph
no unnecessary sentences.
—William Strunk, Jr.
The difference between the
almost-right word and the right
word is really a large matter—
it’s the difference between the
lightning bug and the lightning.
—Mark Twain
Objectives
In this chapter, you’ll:
■
■
■
■
■
Use the functions of the
character-handling library
(<ctype.h>).
Use the string-conversion
functions of the general
utilities library
(<stdlib.h>).
Use the string and character
input/output functions of the
standard input/output library
(<stdio.h>).
Use the string-processing
functions of the stringhandling library
(<string.h>).
Use the memory-processing
functions of the stringhandling library
(<string.h>).
C Characters and Strings
8.1 Introduction
8.1 Introduction
8.2 Fundamentals of Strings and
Characters
8.3 Character-Handling Library
8.3.1 Functions isdigit, isalpha,
isalnum and isxdigit
8.3.2 Functions islower, isupper,
tolower and toupper
8.3.3 Functions isspace, iscntrl,
ispunct, isprint and isgraph
8.4 String-Conversion Functions
8.4.1 Function strtod
8.4.2 Function strtol
8.4.3 Function strtoul
8.5 Standard Input/Output Library
Functions
8.5.1
8.5.2
8.5.3
8.5.4
Functions fgets and putchar
Function getchar
Function sprintf
Function sscanf
8.6 String-Manipulation Functions of the
String-Handling Library
8.6.1 Functions strcpy and strncpy
8.6.2 Functions strcat and strncat
335
8.7 Comparison Functions of the StringHandling Library
8.8 Search Functions of the StringHandling Library
8.8.1
8.8.2
8.8.3
8.8.4
8.8.5
8.8.6
8.8.7
Function strchr
Function strcspn
Function strpbrk
Function strrchr
Function strspn
Function strstr
Function strtok
8.9 Memory Functions of the StringHandling Library
8.9.1
8.9.2
8.9.3
8.9.4
8.9.5
Function memcpy
Function memmove
Function memcmp
Function memchr
Function memset
8.10 Other Functions of the StringHandling Library
8.10.1 Function strerror
8.10.2 Function strlen
8.11 Secure C Programming
Summary | Terminology | Self-Review Exercises | Answers to Self-Review Exercises | Exercises |
Special Section: Advanced String-Manipulation Exercises |
A Challenging String-Manipulation Project | Making a Difference
8.1 Introduction
This chapter introduces the C standard library functions that facilitate string and character
processing. The functions enable programs to process characters, strings, lines of text and
blocks of memory. The chapter discusses the techniques used to develop editors, word processors, page-layout software, computerized typesetting systems and other kinds of text-processing software. The text manipulations performed by formatted input/output functions
like printf and scanf can be implemented using the functions discussed in this chapter.
8.2 Fundamentals of Strings and Characters
Characters are the fundamental building blocks of source programs. Every program is
composed of a sequence of characters that—when grouped together meaningfully—is interpreted by the computer as a series of instructions used to accomplish a task. A program
may contain character constants. A character constant is an int value represented as a
character in single quotes. The value of a character constant is the integer value of the character in the machine’s character set. For example, 'z' represents the integer value of z, and
'\n' the integer value of newline (122 and 10 in ASCII, respectively).
336
Chapter 8 C Characters and Strings
A string is a series of characters treated as a single unit. A string may include letters,
digits and various special characters such as +, -, *, / and $. String literals, or string constants, in C are written in double quotation marks as follows:
"John Q. Doe"
"99999 Main Street"
"Waltham, Massachusetts"
"(201) 555-1212"
(a name)
(a street address)
(a city and state)
(a telephone number)
A string in C is an array of characters ending in the null character ('\0'). A string is
accessed via a pointer to the first character in the string. The value of a string is the address
of its first character. Thus, in C, it’s appropriate to say that a string is a pointer—in fact,
a pointer to the string’s first character. In this sense, strings are like arrays, because an array
is also a pointer to its first element.
A character array or a variable of type char * can be initialized with a string in a definition. The definitions
char color[] = "blue";
const char *colorPtr = "blue";
each initialize a variable to the string "blue". The first definition creates a 5-element array
color containing the characters 'b', 'l', 'u', 'e' and '\0'. The second definition creates
pointer variable colorPtr that points to the string "blue" somewhere in memory.
Portability Tip 8.1
When a variable of type char * is initialized with a string literal, some compilers may
place the string in a location in memory where it cannot be modified. If you might need
to modify a string literal, it should be stored in a character array to ensure modifiability
on all systems.
The preceding array definition could also have been written
char color[] = { 'b', 'l', 'u', 'e', '\0' };
When defining a character array to contain a string, the array must be large enough to store
the string and its terminating null character. The preceding definition automatically determines the size of the array based on the number of initializers in the initializer list.
Common Programming Error 8.1
Not allocating sufficient space in a character array to store the null character that terminates a string is an error.
Common Programming Error 8.2
Printing a “string” that does not contain a terminating null character is an error.
Error-Prevention Tip 8.1
When storing a string of characters in a character array, be sure that the array is large
enough to hold the largest string that will be stored. C allows strings of any length to be
stored. If a string is longer than the character array in which it’s to be stored, characters
beyond the end of the array will overwrite data in memory following the array.
8.3 Character-Handling Library
337
A string can be stored in an array using scanf. For example, the following statement
stores a string in character array word[20]:
scanf( "%19s", word );
The string entered by the user is stored in word. Variable word is an array, which is, of
course, a pointer, so the & is not needed with argument word. Recall from Section 6.4 that
function scanf will read characters until a space, tab, newline or end-of-file indicator is
encountered. So, it’s possible that, without the field width 19 in the conversion specifier
%19s, the user input could exceed 19 characters and that your program might crash! For
this reason, you should always use a field width when using scanf to read into a char array.
The field width 19 in the preceding statement ensures that scanf reads a maximum of 19
characters and saves the last character for the string’s terminating null character. This prevents scanf from writing characters into memory beyond the end of s. (For reading input
lines of arbitrary length, there’s a nonstandard—yet widely supported—function readline, usually included in stdio.h.) For a character array to be printed properly as a string,
the array must contain a terminating null character.
Common Programming Error 8.3
Processing a single character as a string. A string is a pointer—probably a respectably large
integer. However, a character is a small integer (ASCII values range 0–255). On many
systems this causes an error, because low memory addresses are reserved for special purposes
such as operating-system interrupt handlers—so “access violations” occur.
Common Programming Error 8.4
Passing a character as an argument to a function when a string is expected (and vice versa)
is a compilation error.
8.3 Character-Handling Library
The character-handling library (<ctype.h>) includes several functions that perform useful tests and manipulations of character data. Each function receives an unsigned char
(represented as an int) or EOF as an argument. As we discussed in Chapter 4, characters
are often manipulated as integers, because a character in C is a one-byte integer. EOF normally has the value –1. Figure 8.1 summarizes the functions of the character-handling library.
Prototype
Function description
int isblank( int c );
Returns a true value if c is a blank character that separates
words in a line of text and 0 (false) otherwise. [Note: This function is not available in Microsoft Visual C++.]
Returns a true value if c is a digit and 0 (false) otherwise.
Returns a true value if c is a letter and 0 otherwise.
Returns a true value if c is a digit or a letter and 0 otherwise.
int isdigit( int c );
int isalpha( int c );
int isalnum( int c );
Fig. 8.1 | Character-handling library (<ctype.h>) functions. (Part 1 of 2.)
338
Chapter 8 C Characters and Strings
Prototype
Function description
int isxdigit( int c );
Returns a true value if c is a hexadecimal digit character and 0
otherwise. (See Appendix C for a detailed explanation of
binary numbers, octal numbers, decimal numbers and hexadecimal numbers.)
Returns a true value if c is a lowercase letter and 0 otherwise.
Returns a true value if c is an uppercase letter and 0 otherwise.
If c is an uppercase letter, tolower returns c as a lowercase letter.
Otherwise, tolower returns the argument unchanged.
If c is a lowercase letter, toupper returns c as an uppercase letter.
Otherwise, toupper returns the argument unchanged.
Returns a true value if c is a whitespace character—newline
('\n'), space (' '), form feed ('\f'), carriage return ('\r'),
horizontal tab ('\t') or vertical tab ('\v')—and 0 otherwise.
Returns a true value if c is a control character and 0 otherwise.
Returns a true value if c is a printing character other than a
space, a digit, or a letter and returns 0 otherwise.
Returns a true value if c is a printing character including a space
and returns 0 otherwise.
Returns a true value if c is a printing character other than a space
and returns 0 otherwise.
int islower( int c );
int isupper( int c );
int tolower( int c );
int toupper( int c );
int isspace( int c );
int iscntrl( int c );
int ispunct( int c );
int isprint( int c );
int isgraph( int c );
Fig. 8.1 | Character-handling library (<ctype.h>) functions. (Part 2 of 2.)
8.3.1 Functions isdigit, isalpha, isalnum and isxdigit
Figure 8.2 demonstrates functions isdigit, isalpha, isalnum and isxdigit. Function
isdigit determines whether its argument is a digit (0–9). Function isalpha determines
whether its argument is an uppercase (A–Z) or lowercase letter (a–z). Function isalnum
determines whether its argument is an uppercase letter, a lowercase letter or a digit. Function isxdigit determines whether its argument is a hexadecimal digit (A–F, a–f, 0–9).
1
2
3
4
5
6
7
8
9
10
11
12
13
// Fig. 8.2: fig08_02.c
// Using functions isdigit, isalpha, isalnum, and isxdigit
#include <stdio.h>
#include <ctype.h>
int main( void )
{
printf( "%s\n%s%s\n%s%s\n\n", "According to isdigit: ",
isdigit( '8' ) ? "8 is a " : "8 is not a ", "digit",
isdigit( '#' ) ? "# is a " : "# is not a ", "digit" );
printf( "%s\n%s%s\n%s%s\n%s%s\n%s%s\n\n",
"According to isalpha:",
Fig. 8.2 | Using functions isdigit, isalpha, isalnum and isxdigit. (Part 1 of 2.)
8.3 Character-Handling Library
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
isalpha(
isalpha(
isalpha(
isalpha(
'A'
'b'
'&'
'4'
)
)
)
)
?
?
?
?
"A
"b
"&
"4
is
is
is
is
a
a
a
a
"
"
"
"
:
:
:
:
"A
"b
"&
"4
is
is
is
is
not
not
not
not
a
a
a
a
",
",
",
",
339
"letter",
"letter",
"letter",
"letter" );
printf( "%s\n%s%s\n%s%s\n%s%s\n\n",
"According to isalnum:",
isalnum( 'A' ) ? "A is a " : "A is not a ",
"digit or a letter",
isalnum( '8' ) ? "8 is a " : "8 is not a ",
"digit or a letter",
isalnum( '#' ) ? "# is a " : "# is not a ",
"digit or a letter" );
printf( "%s\n%s%s\n%s%s\n%s%s\n%s%s\n%s%s\n",
"According to isxdigit:",
isxdigit( 'F' ) ? "F is a " : "F is not a ",
"hexadecimal digit",
isxdigit( 'J' ) ? "J is a " : "J is not a ",
"hexadecimal digit",
isxdigit( '7' ) ? "7 is a " : "7 is not a ",
"hexadecimal digit",
isxdigit( '$' ) ? "$ is a " : "$ is not a ",
"hexadecimal digit",
isxdigit( 'f' ) ? "f is a " : "f is not a ",
"hexadecimal digit" );
} // end main
According to isdigit:
8 is a digit
# is not a digit
According to isalpha:
A is a letter
b is a letter
& is not a letter
4 is not a letter
According to isalnum:
A is a digit or a letter
8 is a digit or a letter
# is not a digit or a letter
According to isxdigit:
F is a hexadecimal digit
J is not a hexadecimal digit
7 is a hexadecimal digit
$ is not a hexadecimal digit
f is a hexadecimal digit
Fig. 8.2 | Using functions isdigit, isalpha, isalnum and isxdigit. (Part 2 of 2.)
Figure 8.2 uses the conditional operator (?:) to determine whether the string " is a "
or the string " is not a " should be printed in the output for each character tested. For
example, the expression
340
Chapter 8 C Characters and Strings
isdigit( '8' ) ? "8 is a " : "8 is not a "
indicates that if '8' is a digit, the string "8 is a " is printed, and if '8' is not a digit (i.e.,
isdigit returns 0), the string "8 is not a " is printed.
8.3.2 Functions islower, isupper, tolower and toupper
Figure 8.3 demonstrates functions islower, isupper, tolower and toupper. Function
islower determines whether its argument is a lowercase letter (a–z). Function isupper
determines whether its argument is an uppercase letter (A–Z). Function tolower converts
an uppercase letter to a lowercase letter and returns the lowercase letter. If the argument is
not an uppercase letter, tolower returns the argument unchanged. Function toupper converts a lowercase letter to an uppercase letter and returns the uppercase letter. If the argument is not a lowercase letter, toupper returns the argument unchanged.
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
// Fig. 8.3: fig08_03.c
// Using functions islower, isupper, tolower, toupper
#include <stdio.h>
#include <ctype.h>
int main( void )
{
printf( "%s\n%s%s\n%s%s\n%s%s\n%s%s\n\n",
"According to islower:",
islower( 'p' ) ? "p is a " : "p is not
"lowercase letter",
islower( 'P' ) ? "P is a " : "P is not
"lowercase letter",
islower( '5' ) ? "5 is a " : "5 is not
"lowercase letter",
islower( '!' ) ? "! is a " : "! is not
"lowercase letter" );
a ",
a ",
a ",
a ",
printf( "%s\n%s%s\n%s%s\n%s%s\n%s%s\n\n",
"According to isupper:",
isupper( 'D' ) ? "D is an " : "D is not
"uppercase letter",
isupper( 'd' ) ? "d is an " : "d is not
"uppercase letter",
isupper( '8' ) ? "8 is an " : "8 is not
"uppercase letter",
isupper( '$' ) ? "$ is an " : "$ is not
"uppercase letter" );
printf( "%s%c\n%s%c\n%s%c\n%s%c\n",
"u converted to uppercase is ", toupper(
"7 converted to uppercase is ", toupper(
"$ converted to uppercase is ", toupper(
"L converted to lowercase is ", tolower(
} // end main
an ",
an ",
an ",
an ",
'u'
'7'
'$'
'L'
),
),
),
) );
Fig. 8.3 | Using functions islower, isupper, tolower and toupper. (Part 1 of 2.)
8.3 Character-Handling Library
341
According to islower:
p is a lowercase letter
P is not a lowercase letter
5 is not a lowercase letter
! is not a lowercase letter
According to isupper:
D is an uppercase letter
d is not an uppercase letter
8 is not an uppercase letter
$ is not an uppercase letter
u
7
$
L
converted
converted
converted
converted
to
to
to
to
uppercase
uppercase
uppercase
lowercase
is
is
is
is
U
7
$
l
Fig. 8.3 | Using functions islower, isupper, tolower and toupper. (Part 2 of 2.)
8.3.3 Functions isspace, iscntrl, ispunct, isprint and isgraph
Figure 8.4 demonstrates functions isspace, iscntrl, ispunct, isprint and isgraph.
Function isspace determines whether a character is one of the following whitespace characters: space (' '), form feed ('\f'), newline ('\n'), carriage return ('\r'), horizontal tab
('\t') or vertical tab ('\v'). Function iscntrl determines whether a character is one of
the following control characters: horizontal tab ('\t'), vertical tab ('\v'), form feed
('\f'), alert ('\a'), backspace ('\b'), carriage return ('\r') or newline ('\n'). Function
ispunct determines whether a character is a printing character other than a space, a digit
or a letter, such as $, #, (, ), [, ], {, }, ;, : or %. Function isprint determines whether a
character can be displayed on the screen (including the space character). Function isgraph
is the same as isprint, except that the space character is not included.
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
// Fig. 8.4: fig08_04.c
// Using functions isspace, iscntrl, ispunct, isprint, isgraph
#include <stdio.h>
#include <ctype.h>
int main( void )
{
printf( "%s\n%s%s%s\n%s%s%s\n%s%s\n\n",
"According to isspace:",
"Newline", isspace( '\n' ) ? " is a " : " is not a ",
"whitespace character", "Horizontal tab",
isspace( '\t' ) ? " is a " : " is not a ",
"whitespace character",
isspace( '%' ) ? "% is a " : "% is not a ",
"whitespace character" );
printf( "%s\n%s%s%s\n%s%s\n\n", "According to iscntrl:",
"Newline", iscntrl( '\n' ) ? " is a " : " is not a ",
Fig. 8.4 | Using functions isspace, iscntrl, ispunct, isprint and isgraph. (Part 1 of 2.)
342
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
Chapter 8 C Characters and Strings
"control character", iscntrl( '$' ) ? "$ is a " :
"$ is not a ", "control character" );
printf( "%s\n%s%s\n%s%s\n%s%s\n\n",
"According to ispunct:",
ispunct( ';' ) ? "; is a " : "; is not a ",
"punctuation character",
ispunct( 'Y' ) ? "Y is a " : "Y is not a ",
"punctuation character",
ispunct( '#' ) ? "# is a " : "# is not a ",
"punctuation character" );
printf( "%s\n%s%s\n%s%s%s\n\n", "According to isprint:",
isprint( '$' ) ? "$ is a " : "$ is not a ",
"printing character",
"Alert", isprint( '\a' ) ? " is a " : " is not a ",
"printing character" );
printf( "%s\n%s%s\n%s%s%s\n", "According to isgraph:",
isgraph( 'Q' ) ? "Q is a " : "Q is not a ",
"printing character other than a space",
"Space", isgraph( ' ' ) ? " is a " : " is not a ",
"printing character other than a space" );
} // end main
According to isspace:
Newline is a whitespace character
Horizontal tab is a whitespace character
% is not a whitespace character
According to iscntrl:
Newline is a control character
$ is not a control character
According to ispunct:
; is a punctuation character
Y is not a punctuation character
# is a punctuation character
According to isprint:
$ is a printing character
Alert is not a printing character
According to isgraph:
Q is a printing character other than a space
Space is not a printing character other than a space
Fig. 8.4 | Using functions isspace, iscntrl, ispunct, isprint and isgraph. (Part 2 of 2.)
8.4 String-Conversion Functions
This section presents the string-conversion functions from the general utilities library
(<stdlib.h>). These functions convert strings of digits to integer and floating-point values. Figure 8.5 summarizes the string-conversion functions. The C standard also includes
8.4 String-Conversion Functions
343
and strtoull for converting strings to long long int and unsigned long long
respectively. Note the use of const to declare variable nPtr in the function headers
(read from right to left as “nPtr is a pointer to a character constant”); const specifies that
the argument value will not be modified.
strtoll
int,
Function prototype
Function description
double strtod( const char *nPtr, char **endPtr );
Converts the string nPtr to double.
long strtol( const char *nPtr, char **endPtr, int base );
Converts the string nPtr to long.
unsigned long strtoul( const char *nPtr, char **endPtr, int base );
Converts the string nPtr to unsigned long.
Fig. 8.5 | String-conversion functions of the general utilities library.
8.4.1 Function strtod
Function strtod (Fig. 8.6) converts a sequence of characters representing a floating-point
value to double. The function returns 0 if it’s unable to convert any portion of its first argument to double. The function receives two arguments—a string (char *) and a pointer to a
string (char **). The string argument contains the character sequence to be converted to
double—any whitespace characters at the beginning of the string are ignored. The function
uses the char ** argument to modify a char * in the calling function (stringPtr) so that
it points to the location of the first character after the converted portion of the string or to the
entire string if no portion can be converted. Line 14
d = strtod( string, &stringPtr );
indicates that d is assigned the double value converted from string, and stringPtr is assigned the location of the first character after the converted value (51.2) in string.
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
// Fig. 8.6: fig08_06.c
// Using function strtod
#include <stdio.h>
#include <stdlib.h>
int main( void )
{
// initialize string pointer
const char *string = "51.2% are admitted"; // initialize string
double d; // variable to hold converted sequence
char *stringPtr; // create char pointer
d = strtod( string, &stringPtr );
Fig. 8.6 | Using function strtod. (Part 1 of 2.)
344
16
17
18
Chapter 8 C Characters and Strings
printf( "The string \"%s\" is converted to the\n", string );
printf( "double value %.2f and the string \"%s\"\n", d, stringPtr );
} // end main
The string "51.2% are admitted" is converted to the
double value 51.20 and the string "% are admitted"
Fig. 8.6 | Using function strtod. (Part 2 of 2.)
8.4.2 Function strtol
Function strtol (Fig. 8.7) converts to long int a sequence of characters representing an
integer. The function returns 0 if it’s unable to convert any portion of its first argument
to long int. The function receives three arguments—a string (char *), a pointer to a string
and an integer. The string argument contains the character sequence to be converted to
double—any whitespace characters at the beginning of the string are ignored. The function uses the char ** argument to modify a char * in the calling function (remainderPtr)
so that it points to the location of the first character after the converted portion of the string or
to the entire string if no portion can be converted. The integer specifies the base of the value being converted. Line 13
x = strtol( string, &remainderPtr, 0 );
indicates that x is assigned the long value converted from string. The second argument, remainderPtr, is assigned the remainder of string after the conversion. Using NULL for the
second argument causes the remainder of the string to be ignored. The third argument, 0, indicates that the value to be converted can be in octal (base 8), decimal (base 10) or hexadecimal (base 16) format. The base can be specified as 0 or any value between 2 and 36. (See
Appendix C for a detailed explanation of the octal, decimal and hexadecimal number systems.) Numeric representations of integers from base 11 to base 36 use the characters A–Z
to represent the values 10 to 35. For example, hexadecimal values can consist of the digits 0–
9 and the characters A–F. A base-11 integer can consist of the digits 0–9 and the character
A. A base-24 integer can consist of the digits 0–9 and the characters A–N. A base-36 integer
can consist of the digits 0–9 and the characters A–Z. The function returns 0 if it’s unable to
convert any portion of its first argument to a long int value.
1
2
3
4
5
6
7
8
9
10
11
12
// Fig. 8.7: fig08_07.c
// Using function strtol
#include <stdio.h>
#include <stdlib.h>
int main( void )
{
const char *string = "-1234567abc"; // initialize string pointer
char *remainderPtr; // create char pointer
long x; // variable to hold converted sequence
Fig. 8.7 | Using function strtol. (Part 1 of 2.)
8.4 String-Conversion Functions
13
14
15
16
17
18
19
20
21
The
The
The
The
345
x = strtol( string, &remainderPtr, 0 );
printf( "%s\"%s\"\n%s%ld\n%s\"%s\"\n%s%ld\n",
"The original string is ", string,
"The converted value is ", x,
"The remainder of the original string is ",
remainderPtr,
"The converted value plus 567 is ", x + 567 );
} // end main
original string is "-1234567abc"
converted value is -1234567
remainder of the original string is "abc"
converted value plus 567 is -1234000
Fig. 8.7 | Using function strtol. (Part 2 of 2.)
8.4.3 Function strtoul
Function strtoul (Fig. 8.8) converts to unsigned long int a sequence of characters representing an unsigned long int value. The function works identically to function strtol.
The statement
x = strtoul( string, &remainderPtr, 0 );
in line 12 of Fig. 8.8 indicates that x is assigned the unsigned long int value converted
from string. The second argument, &remainderPtr, is assigned the remainder of string
after the conversion. The third argument, 0, indicates that the value to be converted can
be in octal, decimal or hexadecimal format.
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
// Fig. 8.8: fig08_08.c
// Using function strtoul
#include <stdio.h>
#include <stdlib.h>
int main( void )
{
const char *string = "1234567abc"; // initialize string pointer
unsigned long int x; // variable to hold converted sequence
char *remainderPtr; // create char pointer
x = strtoul( string, &remainderPtr, 0 );
printf( "%s\"%s\"\n%s%lu\n%s\"%s\"\n%s%lu\n",
"The original string is ", string,
"The converted value is ", x,
"The remainder of the original string is ",
remainderPtr,
"The converted value minus 567 is ", x - 567 );
} // end main
Fig. 8.8 | Using function strtoul. (Part 1 of 2.)
346
The
The
The
The
Chapter 8 C Characters and Strings
original string is "1234567abc"
converted value is 1234567
remainder of the original string is "abc"
converted value minus 567 is 1234000
Fig. 8.8 | Using function strtoul. (Part 2 of 2.)
8.5 Standard Input/Output Library Functions
This section presents several functions from the standard input/output library (<stdio.h>)
specifically for manipulating character and string data. Figure 8.9 summarizes the character
and string input/output functions of the standard input/output library.
Function prototype
Function description
int getchar( void );
Inputs the next character from the standard input and
returns it as an integer.
char *fgets( char *s, int n, FILE *stream);
int putchar( int c );
int puts( const char *s );
Inputs characters from the specified stream into the array s
until a newline or end-of-file character is encountered, or
until n - 1 bytes are read. In this chapter, we specify the
stream as stdin—the standard input stream, which is typically used to read characters from the keyboard. A terminating null character is appended to the array. Returns the string
that was read into s.
Prints the character stored in c and returns it as an integer.
Prints the string s followed by a newline character. Returns a
nonzero integer if successful, or EOF if an error occurs.
int sprintf( char *s, const char *format, ... );
Equivalent to printf, except the output is stored in the array
s instead of printed on the screen. Returns the number of
characters written to s, or EOF if an error occurs. [Note: We
mention the more secure related functions snprintf and
snprintf_s in the Secure C Programming section of this
chapter and in Appendix F.]
int sscanf( char *s, const char *format, ... );
Equivalent to scanf, except the input is read from the array s
rather than from the keyboard. Returns the number of items
successfully read by the function, or EOF if an error occurs.
Fig. 8.9 | Standard input/output library character and string functions.
8.5.1 Functions fgets and putchar
Figure 8.10 uses functions fgets and putchar to read a line of text from the standard input
(keyboard) and recursively output the characters of the line in reverse order. Function fgets
reads characters from the standard input into its first argument—an array of chars—until a
8.5 Standard Input/Output Library Functions
347
newline or the end-of-file indicator is encountered, or until the maximum number of characters is read. The maximum number of characters is one fewer than the value specified in
fgets’s second argument. The third argument specifies the stream from which to read characters—in this case, we use the standard input stream (stdin). A null character ('\0') is appended to the array when reading terminates. Function putchar prints its character
argument. The program calls recursive function reverse to print the line of text backward.
If the first character of the array received by reverse is the null character '\0', reverse returns. Otherwise, reverse is called again with the address of the subarray beginning at element sPtr[1], and character sPtr[0] is output with putchar when the recursive call is
completed. The order of the two statements in the else portion of the if statement causes
reverse to walk to the terminating null character of the string before a character is printed.
As the recursive calls are completed, the characters are output in reverse order.
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
// Fig. 8.10: fig08_10.c
// Using functions fgets and putchar
#include <stdio.h>
#define SIZE 80
void reverse( const char * const sPtr ); // prototype
int main( void )
{
char sentence[ SIZE ]; // create char array
puts( "Enter a line of text:" );
// use fgets to read line of text
fgets( sentence, SIZE, stdin );
puts( "\nThe line printed backward is:" );
reverse( sentence );
} // end main
// recursively outputs characters in string in reverse order
void reverse( const char * const sPtr )
{
// if end of the string
if ( '\0' == sPtr[ 0 ] ) { // base case
return;
} // end if
else { // if not end of the string
reverse( &sPtr[ 1 ] ); // recursion step
putchar( sPtr[ 0 ] ); // use putchar to display character
} // end else
} // end function reverse
Enter a line of text:
Characters and Strings
The line printed backward is:
sgnirtS dna sretcarahC
Fig. 8.10 | Using functions fgets and putchar. (Part 1 of 2.)
348
Chapter 8 C Characters and Strings
Enter a line of text:
able was I ere I saw elba
The line printed backward is:
able was I ere I saw elba
Fig. 8.10 | Using functions fgets and putchar. (Part 2 of 2.)
8.5.2 Function getchar
Figure 8.11 uses functions getchar and puts to read characters from the standard input
into character array sentence and display the characters as a string. Function getchar
reads a character from the standard input and returns the character as an integer. As you
know, function puts takes a string as an argument and displays the string followed by a
newline character. The program stops inputting characters when either 79 characters have
been read or when getchar reads the newline character entered by the user to end the line
of text. A null character is appended to array sentence (line 20) so that the array may be
treated as a string. Then, line 24 uses puts to display the string contained in sentence.
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
// Fig. 8.11: fig08_11.c
// Using function getchar.
#include <stdio.h>
#define SIZE 80
int main( void )
{
int c; // variable to hold character input by user
char sentence[ SIZE ]; // create char array
int i = 0; // initialize counter i
// prompt user to enter line of text
puts( "Enter a line of text:" );
// use getchar to read each character
while ( i < SIZE - 1 && ( c = getchar() ) != '\n' ) {
sentence[ i++ ] = c;
} // end while
sentence[ i ] = '\0'; // terminate string
// use puts to display sentence
puts( "\nThe line entered was:" );
puts( sentence );
} // end main
Enter a line of text:
This is a test.
The line entered was:
This is a test.
Fig. 8.11 | Using function getchar.
8.5 Standard Input/Output Library Functions
349
8.5.3 Function sprintf
Figure 8.12 uses function sprintf to print formatted data into array s—an array of characters. The function uses the same conversion specifiers as printf (see Chapter 9 for a detailed discussion of formatting). The program inputs an int value and a double value to
be formatted and printed to array s. Array s is the first argument of sprintf. [Note: If your
system supports snprintf_s, then use that in preference to sprintf. If your system
doesn’t support snprintf_s but does support snprintf, then use that in preference to
sprintf. We discuss each of these functions in Appendix F.]
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
// Fig. 8.12: fig08_12.c
// Using function sprintf
#include <stdio.h>
#define SIZE 80
int main( void )
{
char s[ SIZE ]; // create char array
int x; // x value to be input
double y; // y value to be input
puts( "Enter an integer and a double:" );
scanf( "%d%lf", &x, &y );
sprintf( s, "integer:%6d\ndouble:%8.2f", x, y );
printf( "%s\n%s\n",
"The formatted output stored in array s is:", s );
} // end main
Enter an integer and a double:
298 87.375
The formatted output stored in array s is:
integer:
298
double:
87.38
Fig. 8.12 | Using function sprintf.
8.5.4 Function sscanf
Figure 8.13 uses function sscanf to read formatted data from character array s. The function uses the same conversion specifiers as scanf. The program reads an int and a double
from array s and stores the values in x and y, respectively. The values of x and y are printed.
Array s is the first argument of sscanf.
1
2
3
4
// Fig. 8.13: fig08_13.c
// Using function sscanf
#include <stdio.h>
Fig. 8.13 | Using function sscanf. (Part 1 of 2.)
350
5
6
7
8
9
10
11
12
13
14
15
Chapter 8 C Characters and Strings
int main( void )
{
char s[] = "31298 87.375"; // initialize array s
int x; // x value to be input
double y; // y value to be input
sscanf( s, "%d%lf", &x, &y );
printf( "%s\n%s%6d\n%s%8.3f\n",
"The values stored in character array s are:",
"integer:", x, "double:", y );
} // end main
The values stored in character array s are:
integer: 31298
double: 87.375
Fig. 8.13 | Using function sscanf. (Part 2 of 2.)
8.6 String-Manipulation Functions of the StringHandling Library
The string-handling library (<string.h>) provides many useful functions for manipulating string data (copying strings and concatenating strings), comparing strings, searching
strings for characters and other strings, tokenizing strings (separating strings into logical
pieces) and determining the length of strings. This section presents the string-manipulation functions of the string-handling library. The functions are summarized in Fig. 8.14.
Every function—except for strncpy—appends the null character to its result. [Note: Each
of these functions has a more secure version described in the optional Annex K of the C11
standard. We mention these in the Secure C Programming section of this chapter and in
Appendix F.]
Function prototype
Function description
char *strcpy( char *s1, const char *s2 )
Copies string s2 into array s1. The value of s1 is returned.
char *strncpy( char *s1, const char *s2, size_t n )
Copies at most n characters of string s2 into array s1. The value of s1
is returned.
char *strcat( char *s1, const char *s2 )
Appends string s2 to array s1. The first character of s2 overwrites the
terminating null character of s1. The value of s1 is returned.
char *strncat( char *s1, const char *s2, size_t n )
Appends at most n characters of string s2 to array s1. The first character of s2 overwrites the terminating null character of s1. The value
of s1 is returned.
Fig. 8.14 | String-manipulation functions of the string-handling library.
8.6 String-Manipulation Functions of the String-Handling Library
351
Functions strncpy and strncat specify a parameter of type size_t. Function strcpy
copies its second argument (a string) into its first argument—a character array that you must
ensure is large enough to store the string and its terminating null character, which is also
copied. Function strncpy is equivalent to strcpy, except that strncpy specifies the number
of characters to be copied from the string into the array. Function strncpy does not necessarily
copy the terminating null character of its second argument. This occurs only if the number of characters to be copied is at least one more than the length of the string. For example, if "test" is the
second argument, a terminating null character is written only if the third argument to
strncpy is at least 5 (four characters in "test" plus a terminating null character). If the third
argument is larger than 5, null characters are appended to the array until the total number of
characters specified by the third argument are written.
Error-Prevention Tip 8.2
When using functions from the string-handling library, include the <string.h> header.
Common Programming Error 8.5
Not appending a terminating null character to the first argument of a strncpy when the
third argument is less than or equal to the length of the string in the second argument.
8.6.1 Functions strcpy and strncpy
Figure 8.15 uses strcpy to copy the entire string in array x into array y and uses strncpy
to copy the first 14 characters of array x into array z. A null character ('\0') is appended
to array z, because the call to strncpy in the program does not write a terminating null character (the third argument is less than the string length of the second argument).
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
// Fig. 8.15: fig08_15.c
// Using functions strcpy and strncpy
#include <stdio.h>
#include <string.h>
#define SIZE1 25
#define SIZE2 15
int main( void )
{
char x[] = "Happy Birthday to You"; // initialize char array x
char y[ SIZE1 ]; // create char array y
char z[ SIZE2 ]; // create char array z
// copy
printf(
"The
"The
contents of x into y
"%s%s\n%s%s\n",
string in array x is: ", x,
string in array y is: ", strcpy( y, x ) );
// copy first 14 characters of x into z. Does not copy null
// character
strncpy( z, x, SIZE2 - 1 );
Fig. 8.15 | Using functions strcpy and strncpy. (Part 1 of 2.)
352
23
24
25
Chapter 8 C Characters and Strings
z[ SIZE2 - 1 ] = '\0'; // terminate string in z
printf( "The string in array z is: %s\n", z );
} // end main
The string in array x is: Happy Birthday to You
The string in array y is: Happy Birthday to You
The string in array z is: Happy Birthday
Fig. 8.15 | Using functions strcpy and strncpy. (Part 2 of 2.)
8.6.2 Functions strcat and strncat
Function strcat appends its second argument (a string) to its first argument (a character
array containing a string). The first character of the second argument replaces the null ('\0')
that terminates the string in the first argument. You must ensure that the array used to store the
first string is large enough to store the first string, the second string and the terminating null character copied from the second string. Function strncat appends a specified number of characters from the second string to the first string. A terminating null character is automatically
appended to the result. Figure 8.16 demonstrates function strcat and function strncat.
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
// Fig. 8.16: fig08_16.c
// Using functions strcat and strncat
#include <stdio.h>
#include <string.h>
int main( void
{
char s1[ 20
char s2[] =
char s3[ 40
)
] = "Happy "; // initialize char array s1
"New Year "; // initialize char array s2
] = ""; // initialize char array s3 to empty
printf( "s1 = %s\ns2 = %s\n", s1, s2 );
// concatenate s2 to s1
printf( "strcat( s1, s2 ) = %s\n", strcat( s1, s2 ) );
// concatenate first 6 characters of s1 to s3. Place '\0'
// after last character
printf( "strncat( s3, s1, 6 ) = %s\n", strncat( s3, s1, 6 ) );
// concatenate s1 to s3
printf( "strcat( s3, s1 ) = %s\n", strcat( s3, s1 ) );
} // end main
s1 = Happy
s2 = New Year
strcat( s1, s2 ) = Happy New Year
strncat( s3, s1, 6 ) = Happy
strcat( s3, s1 ) = Happy Happy New Year
Fig. 8.16 | Using functions strcat and strncat.
8.7 Comparison Functions of the String-Handling Library
353
8.7 Comparison Functions of the String-Handling Library
This section presents the string-handling library’s string-comparison functions, strcmp and
strncmp. Figure 8.17 contains their prototypes and a brief description of each function.
Function prototype
Function description
int strcmp( const char *s1, const char *s2 );
Compares the string s1 with the string s2. The function returns 0,
less than 0 or greater than 0 if s1 is equal to, less than or greater
than s2, respectively.
int strncmp( const char *s1, const char *s2, size_t n );
Compares up to n characters of the string s1 with the string s2. The
function returns 0, less than 0 or greater than 0 if s1 is equal to, less
than or greater than s2, respectively.
Fig. 8.17 | String-comparison functions of the string-handling library.
Figure 8.18 compares three strings using strcmp and strncmp. Function strcmp compares its first string argument with its second string argument, character by character. The
function returns 0 if the strings are equal, a negative value if the first string is less than the
second string and a positive value if the first string is greater than the second string. Function strncmp is equivalent to strcmp, except that strncmp compares up to a specified
number of characters. Function strncmp does not compare characters following a null
character in a string. The program prints the integer value returned by each function call.
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
// Fig. 8.18: fig08_18.c
// Using functions strcmp and strncmp
#include <stdio.h>
#include <string.h>
int main( void )
{
const char *s1 = "Happy New Year"; // initialize char pointer
const char *s2 = "Happy New Year"; // initialize char pointer
const char *s3 = "Happy Holidays"; // initialize char pointer
printf("%s%s\n%s%s\n%s%s\n\n%s%2d\n%s%2d\n%s%2d\n\n",
"s1 = ", s1, "s2 = ", s2, "s3 = ", s3,
"strcmp(s1, s2) = ", strcmp( s1, s2 ),
"strcmp(s1, s3) = ", strcmp( s1, s3 ),
"strcmp(s3, s1) = ", strcmp( s3, s1 ) );
printf("%s%2d\n%s%2d\n%s%2d\n",
"strncmp(s1, s3, 6) = ", strncmp( s1, s3, 6 ),
"strncmp(s1, s3, 7) = ", strncmp( s1, s3, 7 ),
"strncmp(s3, s1, 7) = ", strncmp( s3, s1, 7 ) );
} // end main
Fig. 8.18 | Using functions strcmp and strncmp. (Part 1 of 2.)
354
Chapter 8 C Characters and Strings
s1 = Happy New Year
s2 = Happy New Year
s3 = Happy Holidays
strcmp(s1, s2) = 0
strcmp(s1, s3) = 1
strcmp(s3, s1) = -1
strncmp(s1, s3, 6) = 0
strncmp(s1, s3, 7) = 6
strncmp(s3, s1, 7) = -6
Fig. 8.18 | Using functions strcmp and strncmp. (Part 2 of 2.)
Common Programming Error 8.6
Assuming that strcmp and strncmp return 1 when their arguments are equal is a logic
error. Both functions return 0 (strangely, the equivalent of C's false value) for equality.
Therefore, when comparing two strings for equality, the result of function strcmp or
strncmp should be compared with 0 to determine whether the strings are equal.
To understand just what it means for one string to be “greater than” or “less than”
another, consider the process of alphabetizing a series of last names. The reader would, no
doubt, place “Jones” before “Smith,” because the first letter of “Jones” comes before the first
letter of “Smith” in the alphabet. But the alphabet is more than just a list of 26 letters—it’s
an ordered list of characters. Each letter occurs in a specific position within the list. “Z” is
more than merely a letter of the alphabet; “Z” is specifically the 26th letter of the alphabet.
How do the string comparison functions know that one particular letter comes before
another? All characters are represented inside the computer as numeric codes in character
sets such as ASCII and Unicode; when the computer compares two strings, it actually compares the numeric codes of the characters in the strings.
8.8 Search Functions of the String-Handling Library
This section presents the functions of the string-handling library used to search strings for
characters and other strings. The functions are summarized in Fig. 8.19. The functions
strcspn and strspn return size_t. [Note: Function strtok has a more secure version described in optional Annex K of the C11 standard. We mention this in the Secure C Programming section of this chapter and in Appendix F.]
Function prototypes and descriptions
char *strchr( const char *s, int c );
Locates the first occurrence of character c in string s. If c is found, a pointer to c in s is
returned. Otherwise, a NULL pointer is returned.
size_t strcspn( const char *s1, const char *s2 );
Determines and returns the length of the initial segment of string s1 consisting of characters not contained in string s2.
Fig. 8.19 | Search functions of the string-handling library. (Part 1 of 2.)
8.8 Search Functions of the String-Handling Library
355
Function prototypes and descriptions
size_t strspn( const char *s1, const char *s2 );
Determines and returns the length of the initial segment of string s1 consisting only of
characters contained in string s2.
char *strpbrk( const char *s1, const char *s2 );
Locates the first occurrence in string s1 of any character in string s2. If a character from
string s2 is found, a pointer to the character in string s1 is returned. Otherwise, a NULL
pointer is returned.
char *strrchr( const char *s, int c );
Locates the last occurrence of c in string s. If c is found, a pointer to c in string s is
returned. Otherwise, a NULL pointer is returned.
char *strstr( const char *s1, const char *s2 );
Locates the first occurrence in string s1 of string s2. If the string is found, a pointer to the
string in s1 is returned. Otherwise, a NULL pointer is returned.
char *strtok( char *s1, const char *s2 );
A sequence of calls to strtok breaks string s1 into tokens—logical pieces such as words
in a line of text—separated by characters contained in string s2. The first call contains
s1 as the first argument, and subsequent calls to continue tokenizing the same string
contain NULL as the first argument. A pointer to the current token is returned by each
call. If there are no more tokens when the function is called, NULL is returned.
Fig. 8.19 | Search functions of the string-handling library. (Part 2 of 2.)
8.8.1 Function strchr
Function strchr searches for the first occurrence of a character in a string. If the character
is found, strchr returns a pointer to the character in the string; otherwise, strchr returns
NULL. Figure 8.20 searches for the first occurrences of 'a' and 'z' in "This is a test".
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
// Fig. 8.20: fig08_20.c
// Using function strchr
#include <stdio.h>
#include <string.h>
int main( void )
{
const char *string = "This is a test"; // initialize char pointer
char character1 = 'a'; // initialize character1
char character2 = 'z'; // initialize character2
// if character1 was found in string
if ( strchr( string, character1 ) != NULL ) {
printf( "\'%c\' was found in \"%s\".\n",
character1, string );
} // end if
Fig. 8.20 | Using function strchr. (Part 1 of 2.)
356
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
Chapter 8 C Characters and Strings
else { // if character1 was not found
printf( "\'%c\' was not found in \"%s\".\n",
character1, string );
} // end else
// if character2 was found in string
if ( strchr( string, character2 ) != NULL ) {
printf( "\'%c\' was found in \"%s\".\n",
character2, string );
} // end if
else { // if character2 was not found
printf( "\'%c\' was not found in \"%s\".\n",
character2, string );
} // end else
} // end main
'a' was found in "This is a test".
'z' was not found in "This is a test".
Fig. 8.20 | Using function strchr. (Part 2 of 2.)
8.8.2 Function strcspn
Function strcspn (Fig. 8.21) determines the length of the initial part of the string in its
first argument that does not contain any characters from the string in its second argument.
The function returns the length of the segment.
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
// Fig. 8.21: fig08_21.c
// Using function strcspn
#include <stdio.h>
#include <string.h>
int main( void )
{
// initialize two char pointers
const char *string1 = "The value is 3.14159";
const char *string2 = "1234567890";
printf( "%s%s\n%s%s\n\n%s\n%s%u\n",
"string1 = ", string1, "string2 = ", string2,
"The length of the initial segment of string1",
"containing no characters from string2 = ",
strcspn( string1, string2 ) );
} // end main
string1 = The value is 3.14159
string2 = 1234567890
The length of the initial segment of string1
containing no characters from string2 = 13
Fig. 8.21 | Using function strcspn.
8.8 Search Functions of the String-Handling Library
357
8.8.3 Function strpbrk
Function strpbrk searches its first string argument for the first occurrence of any character
in its second string argument. If a character from the second argument is found, strpbrk
returns a pointer to the character in the first argument; otherwise, strpbrk returns NULL.
Figure 8.22 shows a program that locates the first occurrence in string1 of any character
from string2.
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
// Fig. 8.22: fig08_22.c
// Using function strpbrk
#include <stdio.h>
#include <string.h>
int main( void )
{
const char *string1 = "This is a test"; // initialize char pointer
const char *string2 = "beware"; // initialize char pointer
printf( "%s\"%s\"\n'%c'%s\n\"%s\"\n",
"Of the characters in ", string2,
*strpbrk( string1, string2 ),
" appears earliest in ", string1 );
} // end main
Of the characters in "beware"
'a' appears earliest in
"This is a test"
Fig. 8.22 | Using function strpbrk.
8.8.4 Function strrchr
Function strrchr searches for the last occurrence of the specified character in a string. If
the character is found, strrchr returns a pointer to the character in the string; otherwise,
strrchr returns NULL. Figure 8.23 shows a program that searches for the last occurrence
of the character 'z' in the string "A zoo has many animals including zebras".
1
2
3
4
5
6
7
8
9
10
11
12
// Fig. 8.23: fig08_23.c
// Using function strrchr
#include <stdio.h>
#include <string.h>
int main( void )
{
// initialize char pointer
const char *string1 = "A zoo has many animals including zebras";
int c = 'z'; // character to search for
Fig. 8.23 | Using function strrchr. (Part 1 of 2.)
358
13
14
15
16
17
Chapter 8 C Characters and Strings
printf( "%s\n%s'%c'%s\"%s\"\n",
"The remainder of string1 beginning with the",
"last occurrence of character ", c,
" is: ", strrchr( string1, c ) );
} // end main
The remainder of string1 beginning with the
last occurrence of character 'z' is: "zebras"
Fig. 8.23 | Using function strrchr. (Part 2 of 2.)
8.8.5 Function strspn
Function strspn (Fig. 8.24) determines the length of the initial part of the string in its
first argument that contains only characters from the string in its second argument. The
function returns the length of the segment.
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
// Fig. 8.24: fig08_24.c
// Using function strspn
#include <stdio.h>
#include <string.h>
int main( void )
{
// initialize two char pointers
const char *string1 = "The value is 3.14159";
const char *string2 = "aehi lsTuv";
printf( "%s%s\n%s%s\n\n%s\n%s%u\n",
"string1 = ", string1, "string2 = ", string2,
"The length of the initial segment of string1",
"containing only characters from string2 = ",
strspn( string1, string2 ) );
} // end main
string1 = The value is 3.14159
string2 = aehi lsTuv
The length of the initial segment of string1
containing only characters from string2 = 13
Fig. 8.24 | Using function strspn.
8.8.6 Function strstr
Function strstr searches for the first occurrence of its second string argument in its first
string argument. If the second string is found in the first string, a pointer to the location
of the string in the first argument is returned. Figure 8.25 uses strstr to find the string
"def" in the string "abcdefabcdef".
8.8 Search Functions of the String-Handling Library
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
359
// Fig. 8.25: fig08_25.c
// Using function strstr
#include <stdio.h>
#include <string.h>
int main( void )
{
const char *string1 = "abcdefabcdef"; // string to search
const char *string2 = "def"; // string to search for
printf( "%s%s\n%s%s\n\n%s\n%s%s\n",
"string1 = ", string1, "string2 = ", string2,
"The remainder of string1 beginning with the",
"first occurrence of string2 is: ",
strstr( string1, string2 ) );
} // end main
string1 = abcdefabcdef
string2 = def
The remainder of string1 beginning with the
first occurrence of string2 is: defabcdef
Fig. 8.25 | Using function strstr.
8.8.7 Function strtok
Function strtok (Fig. 8.26) is used to break a string into a series of tokens. A token is a
sequence of characters separated by delimiters (usually spaces or punctuation marks, but a
delimiter can be any character). For example, in a line of text, each word can be considered
a token, and the spaces and punctuation separating the words can be considered delimiters.
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
// Fig. 8.26: fig08_26.c
// Using function strtok
#include <stdio.h>
#include <string.h>
int main( void )
{
// initialize array string
char string[] = "This is a sentence with 7 tokens";
char *tokenPtr; // create char pointer
printf( "%s\n%s\n\n%s\n",
"The string to be tokenized is:", string,
"The tokens are:" );
tokenPtr = strtok( string, " " ); // begin tokenizing sentence
Fig. 8.26 | Using function strtok. (Part 1 of 2.)
360
18
19
20
21
22
23
Chapter 8 C Characters and Strings
// continue tokenizing sentence until tokenPtr becomes NULL
while ( tokenPtr != NULL ) {
printf( "%s\n", tokenPtr );
tokenPtr = strtok( NULL, " " ); // get next token
} // end while
} // end main
The string to be tokenized is:
This is a sentence with 7 tokens
The tokens are:
This
is
a
sentence
with
7
tokens
Fig. 8.26 | Using function strtok. (Part 2 of 2.)
Multiple calls to strtok are required to tokenize a string—i.e., break it into tokens
(assuming that the string contains more than one token). The first call to strtok contains
two arguments: a string to be tokenized, and a string containing characters that separate
the tokens. In line 16, the statement
tokenPtr = strtok( string, " " ); // begin tokenizing sentence
assigns tokenPtr a pointer to the first token in string. The second argument, " ", indicates that tokens are separated by spaces. Function strtok searches for the first character
in string that’s not a delimiting character (space). This begins the first token. The function then finds the next delimiting character in the string and replaces it with a null ('\0')
character to terminate the current token. Function strtok saves a pointer to the next character following the token in string and returns a pointer to the current token.
Subsequent strtok calls in line 21 continue tokenizing string. These calls contain
NULL as their first argument. The NULL argument indicates that the call to strtok should
continue tokenizing from the location in string saved by the last call to strtok. If no
tokens remain when strtok is called, strtok returns NULL. You can change the delimiter
string in each new call to strtok. Figure 8.26 uses strtok to tokenize the string "This is
a sentence with 7 tokens". Each token is printed separately. Function strtok modifies
the input string by placing '\0' at the end of each token; therefore, a copy of the string
should be made if the string will be used again in the program after the calls to strtok.
[Note: Also see CERT recommendation STR06-C.]
8.9 Memory Functions of the String-Handling Library
The string-handling library functions presented in this section manipulate, compare and
search blocks of memory. The functions treat blocks of memory as character arrays and
can manipulate any block of data. Figure 8.27 summarizes the memory functions of the
string-handling library. In the function discussions, “object” refers to a block of data.
8.9 Memory Functions of the String-Handling Library
361
[Note: Each of these functions has a more secure version described in optional Annex K of
the C11 standard. We mention these in the Secure C Programming section of this chapter
and in Appendix F.]
Function prototype
Function description
void *memcpy( void *s1, const void *s2, size_t n );
Copies n characters from the object pointed to by s2 into the object
pointed to by s1. A pointer to the resulting object is returned.
void *memmove( void *s1, const void *s2, size_t n );
Copies n characters from the object pointed to by s2 into the object
pointed to by s1. The copy is performed as if the characters were
first copied from the object pointed to by s2 into a temporary array
and then from the temporary array into the object pointed to by
s1. A pointer to the resulting object is returned.
int memcmp( const void *s1, const void *s2, size_t n );
Compares the first n characters of the objects pointed to by s1 and
s2. The function returns 0, less than 0 or greater than 0 if s1 is
equal to, less than or greater than s2.
void *memchr( const void *s, int c, size_t n );
Locates the first occurrence of c (converted to unsigned char) in the
first n characters of the object pointed to by s. If c is found, a
pointer to c in the object is returned. Otherwise, NULL is returned.
void *memset( void *s, int c, size_t n );
Copies c (converted to unsigned char) into the first n characters of
the object pointed to by s. A pointer to the result is returned.
Fig. 8.27 | Memory functions of the string-handling library.
The pointer parameters are declared void * so they can be used to manipulate
memory for any data type. In Chapter 7, we saw that a pointer to any data type can be
assigned directly to a pointer of type void *, and a pointer of type void * can be assigned
directly to a pointer to any data type. For this reason, these functions can receive pointers
to any data type. Because a void * pointer cannot be dereferenced, each function receives
a size argument that specifies the number of characters (bytes) the function will process.
For simplicity, the examples in this section manipulate character arrays (blocks of characters). The functions in Fig. 8.27 do not check for terminating null characters.
8.9.1 Function memcpy
Function memcpy copies a specified number of characters from the object pointed to by its
second argument into the object pointed to by its first argument. The function can receive
a pointer to any type of object. The result of this function is undefined if the two objects
overlap in memory (i.e., if they are parts of the same object)—in such cases, use memmove.
Figure 8.28 uses memcpy to copy the string in array s2 to array s1.
362
Chapter 8 C Characters and Strings
Performance Tip 8.1
memcpy is more efficient than strcpy when you know the size of the string you are copying.
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
// Fig. 8.28: fig08_28.c
// Using function memcpy
#include <stdio.h>
#include <string.h>
int main( void )
{
char s1[ 17 ]; // create char array s1
char s2[] = "Copy this string"; // initialize char array s2
memcpy( s1, s2, 17 );
printf( "%s\n%s\"%s\"\n",
"After s2 is copied into s1 with memcpy,",
"s1 contains ", s1 );
} // end main
After s2 is copied into s1 with memcpy,
s1 contains "Copy this string"
Fig. 8.28 | Using function memcpy.
8.9.2 Function memmove
Function memmove, like memcpy, copies a specified number of bytes from the object pointed
to by its second argument into the object pointed to by its first argument. Copying is performed as if the bytes were copied from the second argument into a temporary character
array, then copied from the temporary array into the first argument. This allows characters
from one part of a string to be copied into another part of the same string. Figure 8.29 uses
memmove to copy the last 10 bytes of array x into the first 10 bytes of array x.
Common Programming Error 8.7
String-manipulation functions other than memmove that copy characters have undefined
results when copying takes place between parts of the same string.
1
2
3
4
5
6
7
8
9
// Fig. 8.29: fig08_29.c
// Using function memmove
#include <stdio.h>
#include <string.h>
int main( void )
{
char x[] = "Home Sweet Home"; // initialize char array x
Fig. 8.29 | Using function memmove. (Part 1 of 2.)
8.9 Memory Functions of the String-Handling Library
10
11
12
13
363
printf( "%s%s\n", "The string in array x before memmove is: ", x );
printf( "%s%s\n", "The string in array x after memmove is: ",
(char *) memmove( x, &x[ 5 ], 10 ) );
} // end main
The string in array x before memmove is: Home Sweet Home
The string in array x after memmove is: Sweet Home Home
Fig. 8.29 | Using function memmove. (Part 2 of 2.)
8.9.3 Function memcmp
Function memcmp (Fig. 8.30) compares the specified number of characters of its first argument
with the corresponding characters of its second argument. The function returns a value
greater than 0 if the first argument is greater than the second, returns 0 if the arguments
are equal and returns a value less than 0 if the first argument is less than the second.
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
// Fig. 8.30: fig08_30.c
// Using function memcmp
#include <stdio.h>
#include <string.h>
int main( void )
{
char s1[] = "ABCDEFG"; // initialize char array s1
char s2[] = "ABCDXYZ"; // initialize char array s2
printf( "%s%s\n%s%s\n\n%s%2d\n%s%2d\n%s%2d\n",
"s1 = ", s1, "s2 = ", s2,
"memcmp( s1, s2, 4 ) = ", memcmp( s1, s2, 4 ),
"memcmp( s1, s2, 7 ) = ", memcmp( s1, s2, 7 ),
"memcmp( s2, s1, 7 ) = ", memcmp( s2, s1, 7 ) );
} // end main
s1 = ABCDEFG
s2 = ABCDXYZ
memcmp( s1, s2, 4 ) = 0
memcmp( s1, s2, 7 ) = -1
memcmp( s2, s1, 7 ) = 1
Fig. 8.30 | Using function memcmp.
8.9.4 Function memchr
Function memchr searches for the first occurrence of a byte, represented as unsigned char,
in the specified number of bytes of an object. If the byte is found, a pointer to the byte in
the object is returned; otherwise, a NULL pointer is returned. Figure 8.31 searches for the
character (byte) 'r' in the string "This is a string".
364
1
2
3
4
5
6
7
8
9
10
11
12
13
Chapter 8 C Characters and Strings
// Fig. 8.31: fig08_31.c
// Using function memchr
#include <stdio.h>
#include <string.h>
int main( void )
{
const char *s = "This is a string"; // initialize char pointer
printf( "%s\'%c\'%s\"%s\"\n",
"The remainder of s after character ", 'r',
" is found is ", (char *) memchr( s, 'r', 16 ) );
} // end main
The remainder of s after character 'r' is found is "ring"
Fig. 8.31 | Using function memchr.
8.9.5 Function memset
Function memset copies the value of the byte in its second argument into the first n bytes
of the object pointed to by its first argument, where n is specified by the third argument.
Figure 8.32 uses memset to copy 'b' into the first 7 bytes of string1.
Performance Tip 8.2
Use memset to set an array’s values to 0 rather than looping through the array’s elements
and assigning 0 to each element. For example, in Fig. 6.3, we could have initialized the
10-element array n with memset(n, 0, 10);. Many hardware architectures have a block
copy or clear instruction that the compiler can use to optimize memset for high performance zeroing of memory.
1
2
3
4
5
6
7
8
9
10
11
12
13
// Fig. 8.32: fig08_32.c
// Using function memset
#include <stdio.h>
#include <string.h>
int main( void )
{
char string1[ 15 ] = "BBBBBBBBBBBBBB"; // initialize string1
printf( "string1 = %s\n", string1 );
printf( "string1 after memset = %s\n",
(char *) memset( string1, 'b', 7 ) );
} // end main
string1 = BBBBBBBBBBBBBB
string1 after memset = bbbbbbbBBBBBBB
Fig. 8.32 | Using function memset.
8.10 Other Functions of the String-Handling Library
365
8.10 Other Functions of the String-Handling Library
The two remaining functions of the string-handling library are
Figure 8.33 summarizes the strerror and strlen functions.
Function prototype
strerror
and
strlen.
Function description
char *strerror( int errornum );
Maps errornum into a full text string in a compiler- and locale-specific manner (e.g. the message may appear in different languages
based on its location). A pointer to the string is returned.
size_t strlen( const char *s );
Determines the length of string s. The number of characters preceding the terminating null character is returned.
Fig. 8.33 | Other functions of the string-handling library.
8.10.1 Function strerror
Function strerror takes an error number and creates an error message string. A pointer
to the string is returned. Figure 8.34 demonstrates strerror.
1
2
3
4
5
6
7
8
9
// Fig. 8.34: fig08_34.c
// Using function strerror
#include <stdio.h>
#include <string.h>
int main( void )
{
printf( "%s\n", strerror( 2 ) );
} // end main
No such file or directory
Fig. 8.34 | Using function strerror.
8.10.2 Function strlen
Function strlen takes a string as an argument and returns the number of characters in the
string—the terminating null character is not included in the length. Figure 8.35 demonstrates function strlen.
1
2
3
4
// Fig. 8.35: fig08_35.c
// Using function strlen
#include <stdio.h>
#include <string.h>
Fig. 8.35 | Using function strlen. (Part 1 of 2.)
366
5
6
7
8
9
10
11
12
13
14
15
16
17
Chapter 8 C Characters and Strings
int main( void )
{
// initialize 3 char pointers
const char *string1 = "abcdefghijklmnopqrstuvwxyz";
const char *string2 = "four";
const char *string3 = "Boston";
printf("%s\"%s\"%s%u\n%s\"%s\"%s%u\n%s\"%s\"%s%u\n",
"The length of ", string1, " is ", strlen( string1 ),
"The length of ", string2, " is ", strlen( string2 ),
"The length of ", string3, " is ", strlen( string3 ) );
} // end main
The length of "abcdefghijklmnopqrstuvwxyz" is 26
The length of "four" is 4
The length of "Boston" is 6
Fig. 8.35 | Using function strlen. (Part 2 of 2.)
8.11 Secure C Programming
Secure String-Processing Functions
In this chapter, we presented functions sprintf, strcpy, strncpy, strcat, strncat, strtok, strlen, memcpy, memmove and memset. More secure versions of these and many other
string-processing and input/output functions are described by the C11 standard’s optional
Annex K. If your C compiler supports Annex K, you should use the secure versions of these
functions. Among other things, the more secure versions help prevent buffer overflows by
requiring an additional parameter that specifies the number of elements in the target array
and by ensuring that pointer arguments are non-NULL.
Reading Numeric Inputs and Input Validation
It’s important to validate the data that you input into a program. For example, when you
ask the user to enter an integer in the range 1–100 then attempt to read that integer using
scanf, there are several possible problems. The user could enter an integer that’s outside
the program’s required range, an integer that’s outside the allowed range for integers on
that computer, a non-integer numeric value or a non-numeric value.
You can use various functions that you learned in this chapter to fully validate such
input. For example, you could
• use fgets to read the input as a line of text
• convert the string to a number using strtol and ensure that the conversion was
successful, then
• ensure that the value is in range.
For more information and techniques for converting input to numeric values, see CERT
guideline INT05-C at www.securecoding.cert.org.
Summary
367
Summary
Section 8.2 Fundamentals of Strings and Characters
• Characters are the fundamental building blocks of source programs. Every program is composed
of a sequence of characters that—when grouped together meaningfully—is interpreted by the
computer as a series of instructions used to accomplish a task.
• A character constant is an int value represented as a character in single quotes. The value of a
character constant is the character’s integer value in the machine’s character set.
• A string is a series of characters treated as a single unit. A string may include letters, digits and
various special characters such as +, -, *, / and $. String literals, or string constants, in C are written in double quotation marks.
• A string in C is an array of characters ending in the null character ('\0').
• A string is accessed via a pointer to its first character. The value of a string is the address of its
first character.
• A character array or a variable of type char * can be initialized with a string in a definition.
• When defining a character array to contain a string, the array must be large enough to store the
string and its terminating null character.
• A string can be stored in an array using scanf. Function scanf will read characters until a space,
tab, newline or end-of-file indicator is encountered.
• For a character array to be printed as a string, the array must contain a terminating null character.
Section 8.3 Character-Handling Library
•
•
•
•
•
•
•
•
•
•
•
•
•
Function islower determines whether its argument is a lowercase letter (a–z).
Function isupper determines whether its argument is an uppercase letter (A–Z).
Function isdigit determines whether its argument is a digit (0–9).
Function isalpha determines whether its argument is an uppercase letter (A–Z) or a lowercase
letter (a–z).
Function isalnum determines whether its argument is an uppercase letter (A–Z), a lowercase letter
(a–z) or a digit (0–9).
Function isxdigit determines whether its argument is a hexadecimal digit (A–F, a–f, 0–9).
Function toupper converts a lowercase letter to uppercase and returns the uppercase letter.
Function tolower converts an uppercase letter to lowercase and returns the lowercase letter.
Function isspace determines whether its argument is one of the following whitespace characters:
' ' (space), '\f', '\n', '\r', '\t' or '\v'.
Function iscntrl determines whether its argument is one of the following control characters:
'\t', '\v', '\f', '\a', '\b', '\r' or '\n'.
Function ispunct determines whether its argument is a printing character other than a space, a
digit or a letter.
Function isprint determines whether its argument is any printing character including the space
character.
Function isgraph determines whether its argument is a printing character other than the space
character.
Section 8.4 String-Conversion Functions
• Function strtod converts a sequence of characters representing a floating-point value to double.
The function receives two arguments—a string (char *) and a pointer to char *. The string con-
368
Chapter 8 C Characters and Strings
tains the character sequence to be converted, and the location specified by the pointer to char *
is assigned the address of the remainder of the string after the conversion, or to the entire string
if no portion of the string can be converted.
• Function strtol converts a sequence of characters representing an integer to long. The function
receives three arguments—a string (char *), a pointer to char * and an integer. The string contains
the character sequence to be converted, the location specified by the pointer to char * is assigned
the address of the remainder of the string after the conversion, or to the entire string if no portion
of the string can be converted. The integer specifies the base of the value being converted.
• Function strtoul converts a sequence of characters representing an integer to unsigned long
int. The function works identically to strtol.
Section 8.5 Standard Input/Output Library Functions
• Function fgets reads characters until a newline character or the end-of-file indicator is encountered. The arguments to fgets are an array of type char, the maximum number of characters that
can be read and the stream from which to read. A null character ('\0') is appended to the array
after reading terminates.
• Function putchar prints its character argument.
• Function getchar reads a single character from the standard input and returns it as an integer. If
the end-of-file indicator is encountered, getchar returns EOF.
• Function puts takes a string (char *) as an argument and prints the string followed by a newline
character.
• Function sprintf uses the same conversion specifications as function printf to print formatted
data into an array of type char.
• Function sscanf uses the same conversion specifications as function scanf to read formatted data
from a string.
Section 8.6 String-Manipulation Functions of the String-Handling Library
• Function strcpy copies its second argument (a string) into its first argument (a character array).
You must ensure that the array is large enough to store the string and its terminating null character.
• Function strncpy is equivalent to strcpy, except that a call to strncpy specifies the number of
characters to be copied from the string into the array. The terminating null character will be copied only if the number of characters to be copied is one more than the length of the string.
• Function strcat appends its second string argument—including the terminating null character—to its first string argument. The first character of the second string replaces the null ('\0')
character of the first string. You must ensure that the array used to store the first string is large
enough to store both the first string and the second string.
• Function strncat appends a specified number of characters from the second string to the first
string. A terminating null character is appended to the result.
Section 8.7 Comparison Functions of the String-Handling Library
• Function strcmp compares its first string argument to its second string argument, character by
character. It returns 0 if the strings are equal, returns a negative value if the first string is less than
the second and returns a positive value if the first string is greater than the second.
• Function strncmp is equivalent to strcmp, except that strncmp compares a specified number of
characters. If one of the strings is shorter than the number of characters specified, strncmp compares characters until the null character in the shorter string is encountered.
Terminology
369
Section 8.8 Search Functions of the String-Handling Library
• Function strchr searches for the first occurrence of a character in a string. If the character is
found, strchr returns a pointer to the character in the string; otherwise, strchr returns NULL.
• Function strcspn determines the length of the initial part of the string in its first argument that
does not contain any characters from the string in its second argument. The function returns the
length of the segment.
• Function strpbrk searches for the first occurrence in its first argument of any character in its second argument. If a character from the second argument is found, strpbrk returns a pointer to
the character; otherwise, strpbrk returns NULL.
• Function strrchr searches for the last occurrence of a character in a string. If the character is
found, strrchr returns a pointer to the character in the string; otherwise, strrchr returns NULL.
• Function strspn determines the length of the initial part of the string in its first argument that
contains only characters from the string in its second argument. The function returns the length
of the segment.
• Function strstr searches for the first occurrence of its second string argument in its first string
argument. If the second string is found in the first string, a pointer to the location of the string
in the first argument is returned.
• A sequence of calls to strtok breaks the first string s1 into tokens that are separated by characters
contained in the second string s2. The first call contains s1 as the first argument, and subsequent
calls to continue tokenizing the same string contain NULL as the first argument. A pointer to the
current token is returned by each call. If there are no more tokens when the function is called, a
NULL pointer is returned.
Section 8.9 Memory Functions of the String-Handling Library
• Function memcpy copies a specified number of characters from the object to which its second argument points into the object to which its first argument points. The function can receive a
pointer to any type of object. Function memcpy manipulates the bytes of the object as characters.
• Function memmove copies a specified number of bytes from the object pointed to by its second
argument to the object pointed to by its first argument. Copying is accomplished as if the bytes
were copied from the second argument to a temporary character array and then copied from the
temporary array to the first argument.
• Function memcmp compares the specified number of characters of its first and second arguments.
• Function memchr searches for the first occurrence of a byte, represented as unsigned char, in the
specified number of bytes of an object. If the byte is found, a pointer to the byte is returned; otherwise, a NULL pointer is returned.
• Function memset copies its second argument, treated as an unsigned char, to a specified number
of bytes of the object pointed to by the first argument.
Section 8.10 Other Functions of the String-Handling Library
• Function strerror maps an integer error number into a full text string in a locale specific manner. A pointer to the string is returned.
• Function strlen takes a string as an argument and returns the number of characters in the
string—the terminating null character is not included in the length of the string.
Terminology
character constant 335
character-handling library 337
character set 335
comparing strings 350
370
Chapter 8 C Characters and Strings
concatenating strings 350
control characters 341
copying strings 350
delimiter 359
determining the length of a string 350
fgets function 346
general utilities library (<stdlib.h>) 342
getchar function 348
hexadecimal digit 338
isalnum function 338
isalpha function 338
iscntrl function 341
isdigit function 338
isgraph function 341
islower function 340
isprint function 341
ispunct function 341
isspace function 341
isupper function 340
isxdigit function 338
memchr function 363
memcmp function 363
memcpy function 361
memmove function 362
memset function 364
null character (’\0’) 336
numeric code 354
printing character 341
putchar function 346
special character 336
function 349
function 349
<stdio.h> header 346
<stdlib.h> header 342
strcat function 352
strchr function 355
strcmp function 353
strcspn function 356
strerror function 365
string 336
string comparison function 353
string constant 336
string-conversion function 342
string is a pointer 336
string literal 336
strlen function 365
strncat function 351
strncmp function 353
strncpy function 351
strpbrk function 357
strrchr function 357
strstr function 358
strtod function 343
strtok function 359
strtol function 344
strtoul function 345
token 359
tokenizing strings 350
tolower function 340
toupper function 340
sprintf
sscanf
Self-Review Exercises
8.1
Write a single statement to accomplish each of the following. Assume that variables c
(which stores a character), x, y and z are of type int, variables d, e and f are of type double, variable
ptr is of type char * and arrays s1[100] and s2[100] are of type char.
a) Convert the character stored in variable c to an uppercase letter. Assign the result to
variable c.
b) Determine whether the value of variable c is a digit. Use the conditional operator as
shown in Figs. 8.2–8.4 to print " is a " or " is not a " when the result is displayed.
c) Determine whether the value of variable c is a control character. Use the conditional
operator to print " is a " or " is not a " when the result is displayed.
d) Read a line of text into array s1 from the keyboard. Do not use scanf.
e) Print the line of text stored in array s1. Do not use printf.
f) Assign ptr the location of the last occurrence of c in s1.
g) Print the value of variable c. Do not use printf.
h) Determine whether the value of c is a letter. Use the conditional operator to print
" is a " or " is not a " when the result is displayed.
i) Read a character from the keyboard and store the character in variable c.
j) Assign ptr the location of the first occurrence of s2 in s1.
Answers to Self-Review Exercises
371
k) Determine whether the value of variable c is a printing character. Use the conditional
operator to print " is a " or " is not a " when the result is displayed.
l) Read three double values into variables d, e and f from the string "1.27 10.3 9.432".
m) Copy the string stored in array s2 into array s1.
n) Assign ptr the location of the first occurrence in s1 of any character from s2.
o) Compare the string in s1 with the string in s2. Print the result.
p) Assign ptr the location of the first occurrence of c in s1.
q) Use sprintf to print the values of integer variables x, y and z into array s1. Each value
should be printed with a field width of 7.
r) Append 10 characters from the string in s2 to the string in s1.
s) Determine the length of the string in s1. Print the result.
t) Assign ptr to the location of the first token in s2. Tokens in the string s2 are separated
by commas (,).
8.2
Show two different methods of initializing character array vowel with the string of vowels
"AEIOU".
8.3
What, if anything, prints when each of the following C statements is performed? If the
statement contains an error, describe the error and indicate how to correct it. Assume the following
variable definitions:
char s1[ 50 ] = "jack", s2[ 50 ] = "jill", s3[ 50 ];
a)
b)
c)
d)
e)
8.4
printf( "%c%s", toupper( s1[ 0 ] ), &s1[ 1 ] );
printf( "%s", strcpy( s3, s2 ) );
printf( "%s", strcat( strcat( strcpy( s3, s1 ), " and " ), s2 ) );
printf( "%u", strlen( s1 ) + strlen( s2 ) );
printf( "%u", strlen( s3 ) ); // using s3 after part (c) executes
Find the error in each of the following program segments and explain how to correct it:
a) char s[ 10 ];
strncpy( s, "hello", 5 );
printf( "%s\n", s );
b)
c)
printf( "%s", 'a' );
char s[ 12 ];
strcpy( s, "Welcome Home" );
d)
if ( strcmp( string1, string2 ) ) {
puts( "The strings are equal" );
}
Answers to Self-Review Exercises
8.1
a)
b)
c)
c = toupper( c );
d)
e)
f)
g)
h)
i)
j)
fgets( s1, 100, stdin );
printf( "'%c'%sdigit\n", c, isdigit( c ) ? " is a " : " is not a " );
printf( "'%c'%scontrol character\n",
c, iscntrl( c ) ? " is a " : " is not a " );
puts( s1 );
ptr = strrchr( s1, c );
putchar( c );
printf( "'%c'%sletter\n", c, isalpha( c ) ? " is a " : " is not a " );
c = getchar();
ptr = strstr( s1, s2 );
372
Chapter 8 C Characters and Strings
k)
printf( "'%c'%sprinting character\n",
l)
m)
n)
o)
p)
q)
r)
s)
t)
sscanf( "1.27 10.3 9.432", "%f%f%f", &d, &e, &f );
c, isprint( c ) ? " is a " : " is not a " );
8.2
strcpy( s1, s2 );
ptr = strpbrk( s1, s2 );
printf( "strcmp( s1, s2 ) = %d\n", strcmp( s1, s2 ) );
ptr = strchr( s1, c );
sprintf( s1, "%7d%7d%7d", x, y, z );
strncat( s1, s2, 10 );
printf( "strlen(s1) = %u\n", strlen( s1 ) );
ptr = strtok( s2, "," );
char vowel[] = "AEIOU";
char vowel[] = { 'A', 'E', 'I', 'O', 'U', '\0' };
8.3
8.4
a)
b)
c)
d)
e)
Jack
jill
jack and jill
8
13
a) Error: Function strncpy does not write a terminating null character to array s, because
its third argument is equal to the length of the string "hello".
Correction: Make the third argument of strncpy 6, or assign '\0' to s[ 5 ].
b) Error: Attempting to print a character constant as a string.
Correction: Use %c to output the character, or replace 'a' with "a".
c) Error: Character array s is not large enough to store the terminating null character.
Correction: Declare the array with more elements.
d) Error: Function strcmp returns 0 if the strings are equal; therefore, the condition in the
if statement is false, and the printf will not be executed.
Correction: Compare the result of strcmp with 0 in the condition.
Exercises
8.5
(Character Testing) Write a program that inputs a character from the keyboard and tests it
with each of the functions in the character-handling library. The program should print the value
returned by each function.
8.6
(Displaying Strings in Uppercase and Lowercase) Write a program that inputs a line of text
into char array s[100]. Output the line in uppercase letters and in lowercase letters.
8.7
(Converting Strings to Integers for Calculations) Write a program that inputs four strings
that represent integers, converts the strings to integers, sums the values and prints the total of the
four values.
8.8
(Converting Strings to Floating Point for Calculations) Write a program that inputs four
strings that represent floating-point values, converts the strings to double values, sums the values
and prints the total of the four values.
8.9
(Comparing Strings) Write a program that uses function strcmp to compare two strings input by the user. The program should state whether the first string is less than, equal to or greater
than the second string.
8.10 (Comparing Portions of Strings) Write a program that uses function strncmp to compare
two strings input by the user. The program should input the number of characters to be compared,
then display whether the first string is less than, equal to or greater than the second string.
Exercises
373
8.11 (Random Sentences) Write a program that uses random number generation to create sentences. The program should use four arrays of pointers to char called article, noun, verb and preposition. The program should create a sentence by selecting a word at random from each array in
the following order: article, noun, verb, preposition, article and noun. As each word is picked,
it should be concatenated to the previous words in an array large enough to hold the entire sentence.
The words should be separated by spaces. When the final sentence is output, it should start with a
capital letter and end with a period. The program should generate 20 such sentences. The arrays
should be filled as follows: The article array should contain the articles "the", "a", "one", "some"
and "any"; the noun array should contain the nouns "boy", "girl", "dog", "town" and "car"; the
verb array should contain the verbs "drove", "jumped", "ran", "walked" and "skipped"; the preposition array should contain the prepositions "to", "from", "over", "under" and "on".
After the preceding program is written and working, modify it to produce a short story consisting of several of these sentences. (How about the possibility of a random term paper writer?)
8.12 (Limericks) A limerick is a humorous five-line verse in which the first and second lines
rhyme with the fifth, and the third line rhymes with the fourth. Using techniques similar to those
developed in Exercise 8.11, write a program that produces random limericks. Polishing this program to produce good limericks is a challenging problem, but the result will be worth the effort!
8.13 (Pig Latin) Write a program that encodes English-language phrases into pig Latin. Pig Latin is a form of coded language often used for amusement. Many variations exist in the methods used
to form pig-Latin phrases. For simplicity, use the following algorithm:
To form a pig-Latin phrase from an English-language phrase, tokenize the phrase into words
with function strtok. To translate each English word into a pig-Latin word, place the first letter of
the English word at the end of the English word and add the letters "ay". Thus the word "jump"
becomes "umpjay", the word "the" becomes "hetay" and the word "computer" becomes "omputercay". Blanks between words remain as blanks. Assume the following: The English phrase consists
of words separated by blanks, there are no punctuation marks, and all words have two or more letters. Function printLatinWord should display each word. [Hint: Each time a token is found in a
call to strtok, pass the token pointer to function printLatinWord, and print the pig-Latin word.
Note: We’ve provided simplified rules for converting words to pig Latin here. For more detailed
rules and variations, visit en.wikipedia.org/wiki/Pig_latin.]
8.14 (Tokenizing Telephone Numbers) Write a program that inputs a telephone number as a
string in the form (555) 555-5555. The program should use function strtok to extract the area code
as a token, the first three digits of the phone number as a token and the last four digits of the phone
number as a token. The seven digits of the phone number should be concatenated into one string.
The program should convert the area-code string to int and convert the phone-number string to
long. Both the area code and the phone number should be printed.
8.15 (Displaying a Sentence with Its Words Reversed) Write a program that inputs a line of text,
tokenizes the line with function strtok and outputs the tokens in reverse order.
8.16 (Searching for Substrings) Write a program that inputs a line of text and a search string
from the keyboard. Using function strstr, locate the first occurrence of the search string in the line
of text, and assign the location to variable searchPtr of type char *. If the search string is found,
print the remainder of the line of text beginning with the search string. Then, use strstr again to
locate the next occurrence of the search string in the line of text. If a second occurrence is found,
print the remainder of the line of text beginning with the second occurrence. [Hint: The second call
to strstr should contain searchPtr + 1 as its first argument.]
8.17 (Counting the Occurrences of a Substring) Write a program based on the program of
Exercise 8.16 that inputs several lines of text and a search string and uses function strstr to determine the total occurrences of the string in the lines of text. Print the result.
374
Chapter 8 C Characters and Strings
8.18 (Counting the Occurrences of a Character) Write a program that inputs several lines of text
and a search character and uses function strchr to determine the total occurrences of the character
in the lines of text.
8.19 (Counting the Letters of the Alphabet in a String) Write a program based on the program
of Exercise 8.18 that inputs several lines of text and uses function strchr to determine the total occurrences of each letter of the alphabet in the lines of text. Uppercase and lowercase letters should
be counted together. Store the totals for each letter in an array and print the values in tabular format
after the totals have been determined.
8.20 (Counting the Number of Words in a String) Write a program that inputs several lines of
text and uses strtok to count the total number of words. Assume that the words are separated by
either spaces or newline characters.
8.21 (Alphabetizing a List of Strings) Use the string-comparison functions and the techniques
for sorting arrays to write a program that alphabetizes a list of strings. Use the names of 10 or 15
towns in your area as data for your program.
8.22 The chart in Appendix B shows the numeric code representations for the characters in the
ASCII character set. Study this chart and then state whether each of the following is true or false.
a) The letter "A" comes before the letter "B".
b) The digit “9” comes before the digit "0".
c) The commonly used symbols for addition, subtraction, multiplication and division all
come before any of the digits.
d) The digits come before the letters.
e) If a sort program sorts strings into ascending sequence, then the program will place the
symbol for a right parenthesis before the symbol for a left parenthesis.
8.23 (Strings Starting with "b") Write a program that reads a series of strings and prints only
those beginning with the letter "b".
8.24 (Strings Ending with "ed") Write a program that reads a series of strings and prints only
those that end with the letters "ed".
8.25 (Printing Letters for Various ASCII Codes) Write a program that inputs an ASCII code and
prints the corresponding character.
8.26 (Write Your Own Character-Handling Functions) Using the ASCII character chart in
Appendix B as a guide, write your own versions of the character-handling functions in Fig. 8.1.
8.27 (Write Your String Conversion Functions) Write your own versions of the functions in
Fig. 8.5 for converting strings to numbers.
8.28 (Write Your Own String Copy and Concatenation Functions) Write two versions of each
of the string-copy and string-concatenation functions in Fig. 8.14. The first version should use array
subscripting, and the second should use pointers and pointer arithmetic.
8.29 (Write Your Own String Comparison Functions) Write two versions of each string-comparison function in Fig. 8.17. The first version should use array subscripting, and the second should
use pointers and pointer arithmetic.
8.30 (Write Your Own String Length Function) Write two versions of function strlen in
Fig. 8.33. The first version should use array subscripting, and the second should use pointers and
pointer arithmetic.
Special Section: Advanced String-Manipulation Exercises
375
Special Section: Advanced String-Manipulation Exercises
The preceding exercises are keyed to the text and designed to test the reader’s understanding of fundamental string-manipulation concepts. This section contains intermediate and advanced problems that you should find challenging yet enjoyable. They vary considerably in difficulty. Some
require an hour or two of programming. Others are useful for lab assignments that might require
two or three weeks of study and implementation. Some are challenging term projects.
8.31 (Text Analysis) The availability of computers with string-manipulation capabilities has resulted in some rather interesting approaches to analyzing the writings of great authors. Much attention has been focused on whether William Shakespeare ever lived. Some scholars find substantial
evidence that Christopher Marlowe actually penned the masterpieces attributed to Shakespeare. Researchers have used computers to find similarities in the writings of these two authors. This exercise
examines three methods for analyzing texts with a computer.
a) Write a program that reads several lines of text and prints a table indicating the number
of occurrences of each letter of the alphabet in the text. For example, the phrase
To be, or not to be: that is the question:
contains one “a,” two “b’s,” no “c’s,” and so on.
b) Write a program that reads several lines of text and prints a table indicating the number
of one-letter words, two-letter words, three-letter words, and so on, appearing in the
text. For example, the phrase
Whether 'tis nobler in the mind to suffer
contains
Word length
Occurrences
1
2
3
4
5
6
7
0
2
1
2 (including ’tis)
0
2
1
c) Write a program that reads several lines of text and prints a table indicating the number
of occurrences of each different word in the text. The program should include the words
in the table in the same order in which they appear in the text. For example, the lines
To be, or not to be: that is the question:
Whether 'tis nobler in the mind to suffer
contain the words "to" three times, "be" two times, "or" once, and so on.
8.32 (Printing Dates in Various Formats) Dates are commonly printed in several different formats in business correspondence. Two of the more common formats are
07/21/2003
and
July 21, 2003
Write a program that reads a date in the first format and prints it in the second format.
8.33 (Check Protection) Computers are frequently used in check-writing systems, such as payroll
and accounts payable applications. Many stories circulate regarding weekly paychecks being printed
(by mistake) for amounts in excess of $1 million. Weird amounts are printed by computerized
376
Chapter 8 C Characters and Strings
check-writing systems because of human error and/or machine failure. Systems designers, of course,
make every effort to build controls into their systems to prevent erroneous checks from being issued.
Another serious problem is the intentional alteration of a check amount by someone who
intends to cash it fraudulently. To prevent a dollar amount from being altered, most computerized
check-writing systems employ a technique called check protection.
Checks designed for imprinting by computer contain a fixed number of spaces in which the
computer may print an amount. Suppose a paycheck contains nine blank spaces in which the computer is supposed to print the amount of a weekly paycheck. If the amount is large, then all nine of
those spaces will be filled—for example:
11,230.60
--------123456789
(check amount)
(position numbers)
On the other hand, if the amount is less than $1000, then several of the spaces will ordinarily
be left blank—for example,
99.87
--------123456789
contains four blank spaces. If a check is printed with blank spaces, it’s easier for someone to alter
the amount of the check. To prevent such alteration, many check-writing systems insert leading
asterisks to protect the amount as follows:
****99.87
--------123456789
Write a program that inputs a dollar amount to be printed on a check and then prints the
amount in check-protected format with leading asterisks if necessary. Assume that nine spaces are
available for printing an amount.
8.34 (Writing the Word Equivalent of a Check Amount) Continuing the discussion of the previous exercise, we reiterate the importance of designing check-writing systems to prevent alteration of
check amounts. One common security method requires that the check amount be both written in
numbers and “spelled out” in words. Even if someone is able to alter the numerical amount of the
check, it’s extremely difficult to change the amount in words. Write a program that inputs a numeric check amount and writes the word equivalent of the amount. For example, the amount 52.43
should be written as
FIFTY TWO and 43/100
8.35 (Project: A Metric Conversion Program) Write a program that will assist the user with metric conversions. Your program should allow the user to specify the names of the units as strings (i.e.,
centimeters, liters, grams, and so on for the metric system and inches, quarts, pounds, and so on for
the English system) and should respond to simple questions such as
"How many inches are in 2 meters?"
"How many liters are in 10 quarts?"
Your program should recognize invalid conversions. For example, the question
"How many feet are in 5 kilograms?"
is not meaningful, because "feet" are units of length while "kilograms" are units of mass.
A Challenging String-Manipulation Project
8.36 (Project: A Crossword-Puzzle Generator) Most people have worked a crossword puzzle at
one time or another, but few have ever attempted to generate one. Generating a crossword puzzle is
Making a Difference
377
a difficult problem. It’s suggested here as a string-manipulation project requiring substantial
sophistication and effort. There are many issues you must resolve to get even the simplest crosswordpuzzle generator program working. For example, how does one represent the grid of a crossword
puzzle inside the computer? Should one use a series of strings, or perhaps double-subscripted arrays?
You need a source of words (i.e., a computerized dictionary) that can be directly referenced by the
program. In what form should these words be stored to facilitate the complex manipulations required by the program? The really ambitious reader will want to generate the “clues” portion of the
puzzle in which the brief hints for each “across” word and each “down” word are printed for the
puzzle worker. Merely printing a version of the blank puzzle itself is not a simple problem.
Making a Difference
8.37 (Cooking with Healthier Ingredients) Obesity in America is increasing at an alarming rate.
Check the map from the Centers for Disease Control and Prevention (CDC) at www.cdc.gov/
nccdphp/dnpa/Obesity/trend/maps/index.htm, which shows obesity trends in the United States
over the last 20 years. As obesity increases, so do occurrences of related problems (e.g., heart disease,
high blood pressure, high cholesterol, type 2 diabetes). Write a program that helps users choose
healthier ingredients when cooking, and helps those allergic to certain foods (e.g., nuts, gluten) find
substitutes. The program should read a recipe from the user and suggest healthier replacements for
some of the ingredients. For simplicity, your program should assume the recipe has no abbreviations
for measures such as teaspoons, cups, and tablespoons, and uses numerical digits for quantities (e.g.,
1 egg, 2 cups) rather than spelling them out (one egg, two cups). Some common substitutions are
shown in Fig. 8.36. Your program should display a warning such as, “Always consult your physician
before making significant changes to your diet.”
Your program should take into consideration that replacements are not always one-for-one.
For example, if a cake recipe calls for three eggs, it might reasonably use six egg whites instead.
Conversion data for measurements and substitutes can be obtained at websites such as:
chinesefood.about.com/od/recipeconversionfaqs/f/usmetricrecipes.htm
www.pioneerthinking.com/eggsub.html
www.gourmetsleuth.com/conversions.htm
Your program should consider the user’s health concerns, such as high cholesterol, high blood pressure, weight loss, gluten allergy, and so on. For high cholesterol, the program should suggest substitutes for eggs and dairy products; if the user wishes to lose weight, low-calorie substitutes for
ingredients such as sugar should be suggested.
Ingredient
Substitution
1 cup sour cream
1 cup milk
1 teaspoon lemon juice
1 cup sugar
1 cup yogurt
1/2 cup evaporated milk and 1/2 cup water
1/2 teaspoon vinegar
1/2 cup honey, 1 cup molasses
or 1/4 cup agave nectar
1 cup margarine or yogurt
1 cup rye or rice flour
1 cup cottage cheese
or 1/8 cup mayonnaise and 7/8 cup yogurt
1 cup butter
1 cup flour
1 cup mayonnaise
Fig. 8.36 | Common ingredient substitutions. (Part 1 of 2.)
378
Chapter 8 C Characters and Strings
Ingredient
Substitution
1 egg
2 tablespoons cornstarch, arrowroot flour
or potato starch or 2 egg whites
or 1/2 of a large banana (mashed)
1 cup soy milk
1/4 cup applesauce
whole-grain bread
1 cup milk
1/4 cup oil
white bread
Fig. 8.36 | Common ingredient substitutions. (Part 2 of 2.)
8.38 (Spam Scanner) Spam (or junk e-mail) costs U.S. organizations billions of dollars a year in
spam-prevention software, equipment, network resources, bandwidth, and lost productivity.
Research online some of the most common spam e-mail messages and words, and check your own
junk e-mail folder. Create a list of 30 words and phrases commonly found in spam messages. Write
a program in which the user enters an e-mail message. Read the message into a large character array
and ensure that the program does not attempt to insert characters past the end of the array. Then
scan the message for each of the 30 keywords or phrases. For each occurrence of one of these within
the message, add a point to the message’s “spam score.” Next, rate the likelihood that the message
is spam, based on the number of points it received.
8.39 (SMS Language) Short Message Service (SMS) is a communications service that allows
sending text messages of 160 or fewer characters between mobile phones. With the proliferation of
mobile phone use worldwide, SMS is being used in many developing nations for political purposes
(e.g., voicing opinions and opposition), reporting news about natural disasters, and so on. For example, check out comunica.org/radio2.0/archives/87. Because the length of SMS messages is limited, SMS Language—abbreviations of common words and phrases in mobile text messages, e-mails,
instant messages, etc.—is often used. For example, “in my opinion” is “IMO” in SMS Language.
Research SMS Language online. Write a program that lets the user enter a message using SMS Language, then the program translates it into English (or your own language). Also provide a mechanism
to translate text written in English (or your own language) into SMS Language. One potential problem is that one SMS abbreviation could expand into a variety of phrases. For example, IMO (as used
above) could also stand for “International Maritime Organization,” “in memory of,” etc.
8.40 (Gender Neutrality) In Exercise 1.15, you researched eliminating sexism in all forms of
communication. You then described the algorithm you’d use to read through a paragraph of text
and replace gender-specific words with gender-neutral equivalents. Create a program that reads a
paragraph of text, then replaces gender-specific words with gender-neutral ones. Display the resulting gender-neutral text.
9
C Formatted Input/Output
All the news that’s fit to print.
—Adolph S. Ochs
What mad pursuit? What
struggle to escape?
—John Keats
Objectives
In this chapter, you’ll:
■
Use input and output
streams.
■
Use all print formatting
capabilities.
■
Use all input formatting
capabilities.
■
Print with field widths and
precisions.
■
Use formatting flags in the
printf format control
string.
■
Output literals and escape
sequences.
■
Format input using scanf.
380
Chapter 9 C Formatted Input/Output
9.1 Introduction
9.2 Streams
9.3 Formatting Output with printf
9.4 Printing Integers
9.5 Printing Floating-Point Numbers
9.6 Printing Strings and Characters
9.7 Other Conversion Specifiers
9.8 Printing with Field Widths and
Precision
9.9 Using Flags in the printf Format
Control String
9.10 Printing Literals and Escape
Sequences
9.11 Reading Formatted Input with scanf
9.12 Secure C Programming
Summary | Terminology | Self-Review Exercises | Answers to Self-Review Exercises | Exercises
9.1 Introduction
An important part of the solution to any problem is the presentation of the results. In this
chapter, we discuss in depth the formatting features of scanf and printf. These functions
input data from the standard input stream and output data to the standard output stream.
Include the header <stdio.h> in programs that call these functions. Chapter 11 discusses
several additional functions included in the standard input/output (<stdio.h>) library.
9.2 Streams
All input and output is performed with streams, which are sequences of bytes. In input
operations, the bytes flow from a device (e.g., a keyboard, a disk drive, a network connection) to main memory. In output operations, bytes flow from main memory to a device (e.g.,
a display screen, a printer, a disk drive, a network connection, and so on).
When program execution begins, three streams are connected to the program automatically. Normally, the standard input stream is connected to the keyboard and the standard output stream is connected to the screen. Operating systems often allow these streams
to be redirected to other devices. A third stream, the standard error stream, is connected
to the screen. We’ll show how to output error messages to the standard error stream in
Chapter 11, C File Processing. Streams are discussed in detail in Chapter 11.
9.3 Formatting Output with printf
Precise output formatting is accomplished with printf. Every printf call contains a format control string that describes the output format. The format control string consists of
conversion specifiers, flags, field widths, precisions and literal characters. Together with
the percent sign (%), these form conversion specifications. Function printf can perform
the following formatting capabilities, each of which is discussed in this chapter:
1. Rounding floating-point values to an indicated number of decimal places.
2. Aligning a column of numbers with decimal points appearing one above the other.
3. Right justification and left justification of outputs.
4. Inserting literal characters at precise locations in a line of output.
5. Representing floating-point numbers in exponential format.
9.4 Printing Integers
381
6. Representing unsigned integers in octal and hexadecimal format. See Appendix C
for more information on octal and hexadecimal values.
7. Displaying all types of data with fixed-size field widths and precisions.
The printf function has the form
printf( format-control-string, other-arguments );
format-control-string describes the output format, and other-arguments (which are optional)
correspond to each conversion specification in format-control-string. Each conversion specification begins with a percent sign and ends with a conversion specifier. There can be
many conversion specifications in one format control string.
Common Programming Error 9.1
Forgetting to enclose a format-control-string in quotation marks is a syntax error.
9.4 Printing Integers
An integer is a whole number, such as 776, 0 or –52. Integer values are displayed in one
of several formats. Figure 9.1 describes the integer conversion specifiers.
Conversion specifier
Description
d
Display as a signed decimal integer.
Display as a signed decimal integer. [Note: The i and d specifiers are
different when used with scanf.]
Display as an unsigned octal integer.
Display as an unsigned decimal integer.
Display as an unsigned hexadecimal integer. X causes the digits 0-9
and the uppercase letters A-F to be displayed and x causes the digits
0-9 and the lowercase letters a-f to be displayed.
Place before any integer conversion specifier to indicate that a short,
long or long long integer is displayed, respectively. These are called
length modifiers.
i
o
u
x
or X
h, l
or ll (letter “ell”)
Fig. 9.1 | Integer conversion specifiers.
Figure 9.2 prints an integer using each of the integer conversion specifiers. Only the
minus sign prints; plus signs are normally suppressed. Later in this chapter we’ll see how
to force plus signs to print. Also, the value -455, when read by %u (line 15), is interpreted
as an unsigned value 4294966841.
Common Programming Error 9.2
Printing a negative value with a conversion specifier that expects an unsigned value.
382
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
Chapter 9 C Formatted Input/Output
// Fig. 9.2: fig09_02.c
// Using the integer conversion specifiers
#include <stdio.h>
int main( void )
{
printf( "%d\n", 455 );
printf( "%i\n", 455 ); // i same as d in printf
printf( "%d\n", +455 ); // plus sign does not print
printf( "%d\n", -455 ); // minus sign prints
printf( "%hd\n", 32000 );
printf( "%ld\n", 2000000000L ); // L suffix makes literal a long int
printf( "%o\n", 455 ); // octal
printf( "%u\n", 455 );
printf( "%u\n", -455 );
printf( "%x\n", 455 ); // hexadecimal with lowercase letters
printf( "%X\n", 455 ); // hexidecimal with uppercase letters
} // end main
455
455
455
-455
32000
2000000000
707
455
4294966841
1c7
1C7
Fig. 9.2 | Using the integer conversion specifiers.
9.5 Printing Floating-Point Numbers
A floating-point value contains a decimal point as in 33.5, 0.0 or -657.983. Floatingpoint values are displayed in one of several formats. Figure 9.3 describes the floating-point
conversion specifiers. The conversion specifiers e and E display floating-point values in
exponential notation—the computer equivalent of scientific notation used in mathematics. For example, the value 150.4582 is represented in scientific notation as
1.504582
× 102
and in exponential notation as
1.504582E+02
by the computer. This notation indicates that 1.504582 is multiplied by 10 raised to the
second power (E+02). The E stands for “exponent.”
Values displayed with the conversion specifiers e, E and f show six digits of precision
to the right of the decimal point by default (e.g., 1.04592); other precisions can be specified explicitly. Conversion specifier f always prints at least one digit to the left of the decimal point. Conversion specifiers e and E print lowercase e and uppercase E, respectively,
preceding the exponent, and print exactly one digit to the left of the decimal point.
9.5 Printing Floating-Point Numbers
Conversion specifier
or
or
E
f
g
or
G
e
L
F
383
Description
Display a floating-point value in exponential notation.
Display floating-point values in fixed-point notation (F is not supported in the Visual C++ compiler).
Display a floating-point value in either the floating-point form f or
the exponential form e (or E), based on the magnitude of the value.
Place before any floating-point conversion specifier to indicate that
a long double floating-point value is displayed.
Fig. 9.3 | Floating-point conversion specifiers.
Conversion specifier g (or G) prints in either e (E) or f format with no trailing zeros
(1.234000 is printed as 1.234). Values are printed with e (E) if, after conversion to exponential notation, the value’s exponent is less than -4, or the exponent is greater than or
equal to the specified precision (six significant digits by default for g and G). Otherwise,
conversion specifier f is used to print the value. Trailing zeros are not printed in the fractional part of a value output with g or G. At least one decimal digit is required for the decimal point to be output. The values 0.0000875, 8750000.0, 8.75 and 87.50 are printed
as 8.75e-05, 8.75e+06, 8.75 and 87.5 with the conversion specifier g. The value
0.0000875 uses e notation because, when it’s converted to exponential notation, its exponent (-5) is less than -4. The value 8750000.0 uses e notation because its exponent (6) is
equal to the default precision.
The precision for conversion specifiers g and G indicates the maximum number of significant digits printed, including the digit to the left of the decimal point. The value
1234567.0 is printed as 1.23457e+06, using conversion specifier %g (remember that all
floating-point conversion specifiers have a default precision of 6). There are six significant
digits in the result. The difference between g and G is identical to the difference between
e and E when the value is printed in exponential notation—lowercase g causes a lowercase
e to be output, and uppercase G causes an uppercase E to be output.
Error-Prevention Tip 9.1
When outputting data, be sure that the user is aware of situations in which data may be
imprecise due to formatting (e.g., rounding errors from specifying precisions).
Figure 9.4 demonstrates each of the floating-point conversion specifiers. The %E, %e
and %g conversion specifiers cause the value to be rounded in the output and the conversion
specifier %f does not. [Note: With some compilers, the exponent in the outputs will be
shown with two digits to the right of the + sign.]
1
2
3
4
// Fig. 9.4: fig09_04.c
// Using the floating-point conversion specifiers
#include <stdio.h>
Fig. 9.4 | Using the floating-point conversion specifiers. (Part 1 of 2.)
384
5
6
7
8
9
10
11
12
13
14
Chapter 9 C Formatted Input/Output
int main( void )
{
printf( "%e\n",
printf( "%e\n",
printf( "%e\n",
printf( "%E\n",
printf( "%f\n",
printf( "%g\n",
printf( "%G\n",
} // end main
1234567.89 );
+1234567.89 ); // plus does not print
-1234567.89 ); // minus prints
1234567.89 );
1234567.89 );
1234567.89 );
1234567.89 );
1.234568e+006
1.234568e+006
-1.234568e+006
1.234568E+006
1234567.890000
1.23457e+006
1.23457E+006
Fig. 9.4 | Using the floating-point conversion specifiers. (Part 2 of 2.)
9.6 Printing Strings and Characters
The c and s conversion specifiers are used to print individual characters and strings, respectively. Conversion specifier c requires a char argument. Conversion specifier s requires a pointer to char as an argument. Conversion specifier s causes characters to be
printed until a terminating null ('\0') character is encountered. The program shown in
Fig. 9.5 displays characters and strings with conversion specifiers c and s.
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
// Fig. 9.5: fig09_05c
// Using the character and string conversion specifiers
#include <stdio.h>
int main( void )
{
char character = 'A'; // initialize char
char string[] = "This is a string"; // initialize char array
const char *stringPtr = "This is also a string"; // char pointer
printf( "%c\n",
printf( "%s\n",
printf( "%s\n",
printf( "%s\n",
} // end main
character );
"This is a string" );
string );
stringPtr );
A
This is a string
This is a string
This is also a string
Fig. 9.5 | Using the character and string conversion specifiers.
9.7 Other Conversion Specifiers
385
Common Programming Error 9.3
Using %c to print a string is an error. The conversion specifier %c expects a char argument.
A string is a pointer to char (i.e., a char *).
Common Programming Error 9.4
Using %s to print a char argument often causes a fatal execution-time error called an access violation. The conversion specifier %s expects an argument of type pointer to char.
Common Programming Error 9.5
Using single quotes around character strings is a syntax error. Character strings must be
enclosed in double quotes.
Common Programming Error 9.6
Using double quotes around a character constant creates a pointer to a string consisting of
two characters, the second of which is the terminating null.
9.7 Other Conversion Specifiers
Figure 9.6 shows the p and % conversion specifiers. Figure 9.7’s %p prints the value of ptr
and the address of x; these values are identical because ptr is assigned the address of x. The
last printf statement uses %% to print the % character in a character string.
Portability Tip 9.1
The conversion specifier p displays an address in an implementation-defined manner (on
many systems, hexadecimal notation is used rather than decimal notation).
Common Programming Error 9.7
Trying to print a literal percent character using % rather than %% in the format control string.
When % appears in a format control string, it must be followed by a conversion specifier.
Conversion specifier
Description
p
Display a pointer value in an implementation-defined manner.
Display the percent character.
%
Fig. 9.6 | Other conversion specifiers.
1
2
3
4
5
6
7
// Fig. 9.7: fig09_07.c
// Using the p and % conversion specifiers
#include <stdio.h>
int main( void )
{
int *ptr; // define pointer to int
Fig. 9.7 | Using the p and % conversion specifiers. (Part 1 of 2.)
386
8
9
10
11
12
13
14
15
Chapter 9 C Formatted Input/Output
int x = 12345; // initialize int x
ptr = &x; // assign address of x to ptr
printf( "The value of ptr is %p\n", ptr );
printf( "The address of x is %p\n\n", &x );
puts( "Printing a %% in a format control string" );
} // end main
The value of ptr is 002EF778
The address of x is 002EF778
Printing a % in a format control string
Fig. 9.7 | Using the p and % conversion specifiers. (Part 2 of 2.)
9.8 Printing with Field Widths and Precision
The exact size of a field in which data is printed is specified by a field width. If the field
width is larger than the data being printed, the data will normally be right justified within
that field. An integer representing the field width is inserted between the percent sign (%)
and the conversion specifier (e.g., %4d). Figure 9.8 prints two groups of five numbers each,
right justifying those numbers that contain fewer digits than the field width. The field
width is increased to print values wider than the field. Note that the minus sign for a negative value uses one character position in the field width. Field widths can be used with all
conversion specifiers.
Common Programming Error 9.8
Not providing a sufficiently large field width to handle a value to be printed can offset
other data being printed and can produce confusing outputs. Know your data!
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
// Fig. 9.8: fig09_08.c
// Right justifying integers in a field
#include <stdio.h>
int main( void )
{
printf( "%4d\n", 1 );
printf( "%4d\n", 12 );
printf( "%4d\n", 123 );
printf( "%4d\n", 1234 );
printf( "%4d\n\n", 12345 );
printf( "%4d\n",
printf( "%4d\n",
printf( "%4d\n",
printf( "%4d\n",
printf( "%4d\n",
} // end main
-1 );
-12 );
-123 );
-1234 );
-12345 );
Fig. 9.8 | Right justifying integers in a field. (Part 1 of 2.)
9.8 Printing with Field Widths and Precision
387
1
12
123
1234
12345
-1
-12
-123
-1234
-12345
Fig. 9.8 | Right justifying integers in a field. (Part 2 of 2.)
Function printf also enables you to specify the precision with which data is printed. Precision has different meanings for different data types. When used with integer conversion
specifiers, precision indicates the minimum number of digits to be printed. If the printed value
contains fewer digits than the specified precision and the precision value has a leading zero
or decimal point, zeros are prefixed to the printed value until the total number of digits is
equivalent to the precision. If neither a zero nor a decimal point is present in the precision
value, spaces are inserted instead. The default precision for integers is 1. When used with
floating-point conversion specifiers e, E and f, the precision is the number of digits to appear
after the decimal point. When used with conversion specifiers g and G, the precision is the
maximum number of significant digits to be printed. When used with conversion specifier s,
the precision is the maximum number of characters to be written from the string.
To use precision, place a decimal point (.), followed by an integer representing the precision between the percent sign and the conversion specifier. Figure 9.9 demonstrates the use
of precision in format control strings. When a floating-point value is printed with a precision
smaller than the original number of decimal places in the value, the value is rounded.
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
// Fig. 9.9: fig09_09.c
// Printing integers, floating-point numbers and strings with precisions
#include <stdio.h>
int main( void )
{
int i = 873; // initialize int i
double f = 123.94536; // initialize double f
char s[] = "Happy Birthday"; // initialize char array s
puts( "Using precision for integers" );
printf( "\t%.4d\n\t%.9d\n\n", i, i );
puts( "Using precision for floating-point numbers" );
printf( "\t%.3f\n\t%.3e\n\t%.3g\n\n", f, f, f );
puts( "Using precision for strings" );
printf( "\t%.11s\n", s );
} // end main
Fig. 9.9 | Printing integers, floating-point numbers and strings with precisions. (Part 1 of 2.)
388
Chapter 9 C Formatted Input/Output
Using precision for integers
0873
000000873
Using precision for floating-point numbers
123.945
1.239e+002
124
Using precision for strings
Happy Birth
Fig. 9.9 | Printing integers, floating-point numbers and strings with precisions. (Part 2 of 2.)
The field width and the precision can be combined by placing the field width, followed by a decimal point, followed by a precision between the percent sign and the conversion specifier, as in the statement
printf( "%9.3f", 123.456789 );
which displays 123.457 with three digits to the right of the decimal point right justified in
a nine-digit field.
It’s possible to specify the field width and the precision using integer expressions in
the argument list following the format control string. To use this feature, insert an asterisk
(*) in place of the field width or precision (or both). The matching int argument in the
argument list is evaluated and used in place of the asterisk. A field width’s value may be
either positive or negative (which causes the output to be left justified in the field, as
described in the next section). The statement
printf( "%*.*f", 7, 2, 98.736 );
uses 7 for the field width, 2 for the precision and outputs the value 98.74 right justified.
9.9 Using Flags in the printf Format Control String
Function printf also provides flags to supplement its output formatting capabilities. Five
flags are available for use in format control strings (Fig. 9.10). To use a flag in a format
control string, place the flag immediately to the right of the percent sign. Several flags may
be combined in one conversion specifier.
Figure 9.11 demonstrates right justification and left justification of a string, an
integer, a character and a floating-point number.
Flag
+
(minus sign)
(plus sign)
space
Description
Left justify the output within the specified field.
Display a plus sign preceding positive values and a minus sign preceding negative values.
Print a space before a positive value not printed with the + flag.
Fig. 9.10 | Format control string flags. (Part 1 of 2.)
9.9 Using Flags in the printf Format Control String
389
Flag
Description
#
Prefix 0 to the output value when used with the octal conversion specifier o.
Prefix 0x or 0X to the output value when used with the hexadecimal conversion specifiers x or X.
Force a decimal point for a floating-point number printed with e, E, f, g or G
that does not contain a fractional part. (Normally the decimal point is
printed only if a digit follows it.) For g and G specifiers, trailing zeros are not
eliminated.
Pad a field with leading zeros.
0
(zero)
Fig. 9.10 | Format control string flags. (Part 2 of 2.)
1
2
3
4
5
6
7
8
9
// Fig. 9.11: fig09_11.c
// Right justifying and left justifying values
#include <stdio.h>
int main( void )
{
printf( "%10s%10d%10c%10f\n\n", "hello", 7, 'a', 1.23 );
printf( "%-10s%-10d%-10c%-10f\n", "hello", 7, 'a', 1.23 );
} // end main
hello
hello
7
7
a
a
1.230000
1.230000
Fig. 9.11 | Right justifying and left justifying values.
Figure 9.12 prints a positive number and a negative number, each with and without
the + flag. The minus sign is displayed in both cases, but the plus sign is displayed only
when the + flag is used.
1
2
3
4
5
6
7
8
9
// Fig. 9.12: fig09_12.c
// Printing positive and negative numbers with and without the + flag
#include <stdio.h>
int main( void )
{
printf( "%d\n%d\n", 786, -786 );
printf( "%+d\n%+d\n", 786, -786 );
} // end main
786
-786
+786
-786
Fig. 9.12 | Printing positive and negative numbers with and without the + flag.
390
Chapter 9 C Formatted Input/Output
Figure 9.13 prefixes a space to the positive number with the space flag. This is useful for
aligning positive and negative numbers with the same number of digits. The value -547 is
not preceded by a space in the output because of its minus sign.
1
2
3
4
5
6
7
8
9
// Fig. 9.13: fig09_13.c
// Using the space flag
// not preceded by + or #include <stdio.h>
int main( void )
{
printf( "% d\n% d\n", 547, -547 );
} // end main
547
-547
Fig. 9.13 | Using the space flag.
Figure 9.14 uses the # flag to prefix 0 to the octal value and 0x and 0X to the hexadecimal values, and to force the decimal point on a value printed with g.
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
// Fig. 9.14: fig09_14.c
// Using the # flag with conversion specifiers
// o, x, X and any floating-point specifier
#include <stdio.h>
int main( void )
{
int c = 1427; // initialize c
double p = 1427.0; // initialize p
printf( "%#o\n", c );
printf( "%#x\n", c );
printf( "%#X\n", c );
printf( "\n%g\n", p );
printf( "%#g\n", p );
} // end main
02623
0x593
0X593
1427
1427.00
Fig. 9.14 | Using the # flag with conversion specifiers.
Figure 9.15 combines the + flag and the 0 (zero) flag to print 452 in a 9-space field with
a + sign and leading zeros, then prints 452 again using only the 0 flag and a 9-space field.
9.10 Printing Literals and Escape Sequences
1
2
3
4
5
6
7
8
9
391
// Fig. 9.15: fig09_15.c
// Using the 0( zero ) flag
#include <stdio.h>
int main( void )
{
printf( "%+09d\n", 452 );
printf( "%09d\n", 452 );
} // end main
+00000452
000000452
Fig. 9.15 | Using the 0 (zero) flag.
9.10 Printing Literals and Escape Sequences
Most literal characters to be printed in a printf statement can simply be included in the
format control string. However, there are several “problem” characters, such as the quotation mark (") that delimits the format control string itself. Various control characters, such
as newline and tab, must be represented by escape sequences. An escape sequence is represented by a backslash (\), followed by a particular escape character. Figure 9.16 lists the
escape sequences and the actions they cause.
Escape sequence
\'
\"
\?
\\
\a
\b
\f
\n
\r
\t
\v
(single quote)
(double quote)
(question mark)
(backslash)
(alert or bell)
(backspace)
(new page or form feed)
(newline)
(carriage return)
(horizontal tab)
(vertical tab)
Description
Output the single quote (') character.
Output the double quote (") character.
Output the question mark (?) character.
Output the backslash (\) character.
Cause an audible (bell) or visual alert.
Move the cursor back one position on the current line.
Move the cursor to the start of the next logical page.
Move the cursor to the beginning of the next line.
Move the cursor to the beginning of the current line.
Move the cursor to the next horizontal tab position.
Move the cursor to the next vertical tab position.
Fig. 9.16 | Escape sequences.
9.11 Reading Formatted Input with scanf
Precise input formatting can be accomplished with scanf. Every scanf statement contains
a format control string that describes the format of the data to be input. The format control string consists of conversion specifiers and literal characters. Function scanf has the
following input formatting capabilities:
392
Chapter 9 C Formatted Input/Output
1. Inputting all types of data.
2. Inputting specific characters from an input stream.
3. Skipping specific characters in the input stream.
Function scanf is written in the following form:
scanf( format-control-string, other-arguments );
format-control-string describes the formats of the input, and other-arguments are pointers
to variables in which the input will be stored.
Good Programming Practice 9.1
When inputting data, prompt the user for one data item or a few data items at a time.
Avoid asking the user to enter many data items in response to a single prompt.
Good Programming Practice 9.2
Always consider what the user and your program will do when (not if) incorrect data is
entered—for example, a value for an integer that’s nonsensical in a program’s context, or
a string with missing punctuation or spaces.
Figure 9.17 summarizes the conversion specifiers used to input all types of data. The
remainder of this section provides programs that demonstrate reading data with the
various scanf conversion specifiers.
Conversion specifier
Description
Integers
d
i
o
u
x
or X
h, l
and ll
Read an optionally signed decimal integer. The corresponding argument
is a pointer to an int.
Read an optionally signed decimal, octal or hexadecimal integer. The
corresponding argument is a pointer to an int.
Read an octal integer. The corresponding argument is a pointer to an
unsigned int.
Read an unsigned decimal integer. The corresponding argument is a
pointer to an unsigned int.
Read a hexadecimal integer. The corresponding argument is a pointer
to an unsigned int.
Place before any of the integer conversion specifiers to indicate that a
short, long or long long integer is to be input, respectively.
Floating-point numbers
e, E, f, g or G
Read a floating-point value. The corresponding argument is a pointer
to a floating-point variable.
l or L
Place before any of the floating-point conversion specifiers to indicate
that a double or long double value is to be input. The corresponding
argument is a pointer to a double or long double variable.
Fig. 9.17 | Conversion specifiers for scanf. (Part 1 of 2.)
9.11 Reading Formatted Input with scanf
Conversion specifier
393
Description
Characters and strings
Read a character. The corresponding argument is a pointer to a char;
no null ('\0') is added.
Read a string. The corresponding argument is a pointer to an array of
type char that’s large enough to hold the string and a terminating null
('\0') character—which is automatically added.
c
s
Scan set
[scan characters]
Scan a string for a set of characters that are stored in an array.
Miscellaneous
Read an address of the same form produced when an address is output
with %p in a printf statement.
Store the number of characters input so far in this call to scanf. The
corresponding argument is a pointer to an int.
Skip a percent sign (%) in the input.
p
n
%
Fig. 9.17 | Conversion specifiers for scanf. (Part 2 of 2.)
Figure 9.18 reads integers with the various integer conversion specifiers and displays
the integers as decimal numbers. Conversion specifier %i can input decimal, octal and
hexadecimal integers.
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
// Fig. 9.18: fig09_18.c
// Reading input with integer conversion specifiers
#include <stdio.h>
int main( void )
{
int a;
int b;
int c;
int d;
int e;
int f;
int g;
puts( "Enter seven integers: " );
scanf( "%d%i%i%i%o%u%x", &a, &b, &c, &d, &e, &f, &g );
puts( "\nThe input displayed as decimal integers is:" );
printf( "%d %d %d %d %d %d %d\n", a, b, c, d, e, f, g );
} // end main
Fig. 9.18 | Reading input with integer conversion specifiers. (Part 1 of 2.)
394
Chapter 9 C Formatted Input/Output
Enter seven integers:
-70 -70 070 0x70 70 70 70
The input displayed as decimal integers is:
-70 -70 56 112 56 70 112
Fig. 9.18 | Reading input with integer conversion specifiers. (Part 2 of 2.)
When inputting floating-point numbers, any of the floating-point conversion specifiers e, E, f, g or G can be used. Figure 9.19 reads three floating-point numbers, one with
each of the three types of floating conversion specifiers, and displays all three numbers with
conversion specifier f. The program output confirms the fact that floating-point values are
imprecise—this is highlighted by the third value printed.
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
// Fig. 9.19: fig09_19.c
// Reading input with floating-point conversion specifiers
#include <stdio.h>
// function main begins program execution
int main( void )
{
double a;
double b;
double c;
puts( "Enter three floating-point numbers:" );
scanf( "%le%lf%lg", &a, &b, &c );
puts( "\nHere are the numbers entered in plain:" );
puts( "floating-point notation:\n" );
printf( "%f\n%f\n%f\n", a, b, c );
} // end main
Enter three floating-point numbers:
1.27987 1.27987e+03 3.38476e-06
Here are the numbers entered in plain
floating-point notation:
1.279870
1279.870000
0.000003
Fig. 9.19 | Reading input with floating-point conversion specifiers.
Characters and strings are input using the conversion specifiers c and s, respectively.
Figure 9.20 prompts the user to enter a string. The program inputs the first character of
the string with %c and stores it in the character variable x, then inputs the remainder of the
string with %s and stores it in character array y.
9.11 Reading Formatted Input with scanf
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
395
// Fig. 9.20: fig09_20.c
// Reading characters and strings
#include <stdio.h>
int main( void )
{
char x;
char y[ 9 ];
printf( "%s", "Enter a string: " );
scanf( "%c%8s", &x, y );
puts( "The input was:\n" );
printf( "the character \"%c\" and the string \"%s\"\n", x, y );
} // end main
Enter a string: Sunday
The input was:
the character "S" and the string "unday"
Fig. 9.20 | Reading characters and strings.
A sequence of characters can be input using a scan set. A scan set is a set of characters
enclosed in square brackets, [], and preceded by a percent sign in the format control
string. A scan set scans the characters in the input stream, looking only for those characters
that match characters contained in the scan set. Each time a character is matched, it’s
stored in the scan set’s corresponding argument—a pointer to a character array. The scan
set stops inputting characters when a character that’s not contained in the scan set is
encountered. If the first character in the input stream does not match a character in the
scan set, the array is not modified. Figure 9.21 uses the scan set [aeiou] to scan the input
stream for vowels. Notice that the first seven letters of the input are read. The eighth letter
(h) is not in the scan set and therefore the scanning is terminated.
1
2
3
4
5
6
7
8
9
10
11
12
13
14
// Fig. 9.21: fig09_21.c
// Using a scan set
#include <stdio.h>
// function main begins program execution
int main( void )
{
char z[ 9 ]; // define array z
printf( "%s", "Enter string: " );
scanf( "%8[aeiou]", z ); // search for set of characters
printf( "The input was \"%s\"\n", z );
} // end main
Fig. 9.21 | Using a scan set. (Part 1 of 2.)
396
Chapter 9 C Formatted Input/Output
Enter string: ooeeooahah
The input was "ooeeooa"
Fig. 9.21 | Using a scan set. (Part 2 of 2.)
The scan set can also be used to scan for characters not contained in the scan set by
using an inverted scan set. To create an inverted scan set, place a caret (^) in the square
brackets before the scan characters. This causes characters not appearing in the scan set to
be stored. When a character contained in the inverted scan set is encountered, input terminates. Figure 9.22 uses the inverted scan set [^aeiou] to search for consonants—more
properly to search for “nonvowels.”
1
2
3
4
5
6
7
8
9
10
11
12
13
// Fig. 9.22: fig09_22.c
// Using an inverted scan set
#include <stdio.h>
int main( void )
{
char z[ 9 ];
printf( "%s", "Enter a string: " );
scanf( "%8[^aeiou]", z ); // inverted scan set
printf( "The input was \"%s\"\n", z );
} // end main
Enter a string: String
The input was "Str"
Fig. 9.22 | Using an inverted scan set.
A field width can be used in a scanf conversion specifier to read a specific number of
characters from the input stream. Figure 9.23 inputs a series of consecutive digits as a twodigit integer and an integer consisting of the remaining digits in the input stream.
1
2
3
4
5
6
7
8
9
10
11
12
13
14
// Fig. 9.23: fig09_23.c
// inputting data with a field width
#include <stdio.h>
int main( void )
{
int x;
int y;
printf( "%s", "Enter a six digit integer: " );
scanf( "%2d%d", &x, &y );
printf( "The integers input were %d and %d\n", x, y );
} // end main
Fig. 9.23 | Inputting data with a field width. (Part 1 of 2.)
9.11 Reading Formatted Input with scanf
397
Enter a six digit integer: 123456
The integers input were 12 and 3456
Fig. 9.23 | Inputting data with a field width. (Part 2 of 2.)
Often it’s necessary to skip certain characters in the input stream. For example, a date
could be entered as
11-10-1999
Each number in the date needs to be stored, but the dashes that separate the numbers
can be discarded. To eliminate unnecessary characters, include them in the format control
string of scanf (whitespace characters—such as space, newline and tab—skip all leading
whitespace). For example, to skip the dashes in the input, use the statement
scanf( "%d-%d-%d", &month, &day, &year );
Although this scanf does eliminate the dashes in the preceding input, it’s possible that
the date could be entered as
10/11/1999
In this case, the preceding scanf would not eliminate the unnecessary characters. For this
reason, scanf provides the assignment suppression character *. This character enables
scanf to read any type of data from the input and discard it without assigning it to a variable. Figure 9.24 uses the assignment suppression character in the %c conversion specifier
to indicate that a character appearing in the input stream should be read and discarded.
Only the month, day and year are stored. The values of the variables are printed to demonstrate that they’re in fact input correctly. The argument lists for each scanf call do not
contain variables for the conversion specifiers that use the assignment suppression character. The corresponding characters are simply discarded.
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
// Fig. 9.24: fig09_24.c
// Reading and discarding characters from the input stream
#include <stdio.h>
int main( void )
{
int month1;
int day1;
int year1;
int month2;
int day2;
int year2;
printf( "%s", "Enter a date in the form mm-dd-yyyy: " );
scanf( "%d%*c%d%*c%d", &month1, &day1, &year1 );
printf( "month = %d
day = %d
year = %d\n\n", month1, day1, year1 );
Fig. 9.24 | Reading and discarding characters from the input stream. (Part 1 of 2.)
398
19
20
21
22
23
Chapter 9 C Formatted Input/Output
printf( "%s", "Enter a date in the form mm/dd/yyyy: " );
scanf( "%d%*c%d%*c%d", &month2, &day2, &year2 );
printf( "month = %d
} // end main
day = %d
year = %d\n", month2, day2, year2 );
Enter a date in the form mm-dd-yyyy: 11-18-2012
month = 11 day = 18 year = 2012
Enter a date in the form mm/dd/yyyy: 11/18/2012
month = 11 day = 18 year = 2012
Fig. 9.24 | Reading and discarding characters from the input stream. (Part 2 of 2.)
9.12 Secure C Programming
The C standard lists many cases in which using incorrect library-function arguments can result in undefined behaviors. These can cause security vulnerabilities, so they should be avoided. Such problems can occur when using printf (or any of its variants, such as sprintf,
fprintf, printf_s, etc.) with improperly formed conversion specifications. CERT rule
FIO00-C (www.securecoding.cert.org) discusses these issues and presents a table showing
the valid combinations of formatting flags, length modifiers and conversion-specifier characters that can be used to form conversion specifications. The table also shows the proper
argument type for each valid conversion specification. In general, as you study any programming language, if the language specification says that doing something can lead to undefined behavior, avoid doing it to prevent security vulnerabilities.
Summary
Section 9.2 Streams
• All input and output is performed with streams—which are sequences of bytes.
• Normally, the standard input stream is connected to the keyboard, and the standard output and
error streams are connected to the computer screen.
• Operating systems often allow the standard input and standard output streams to be redirected
to other devices.
Section 9.3 Formatting Output with printf
• A format control string describes the formats in which the output values appear. The format control string consists of conversion specifiers, flags, field widths, precisions and literal characters.
Section 9.4 Printing Integers
• Integers are printed with the following conversion specifiers: d or i for optionally signed integers,
o for unsigned integers in octal form, u for unsigned integers in decimal form and x or X for unsigned integers in hexadecimal form. The modifier h or l is prefixed to the preceding conversion
specifiers to indicate a short or long integer, respectively.
Section 9.5 Printing Floating-Point Numbers
• Floating-point values are printed with the following conversion specifiers: e or E for exponential
notation, f for regular floating-point notation, and g or G for either e (or E) notation or f nota-
Summary
399
tion. When the g (or G) conversion specifier is indicated, the e (or E) conversion specifier is used
if the value’s exponent is less than -4 or greater than or equal to the precision with which the
value is printed.
• The precision for the g and G conversion specifiers indicates the maximum number of significant
digits printed.
Section 9.6 Printing Strings and Characters
• The conversion specifier c prints a character.
• The conversion specifier s prints a string of characters ending in the null character.
Section 9.7 Other Conversion Specifiers
• The conversion specifier p displays an address in an implementation-defined manner (on many
systems, hexadecimal notation is used).
• The conversion specifier %% causes a literal % to be output.
Section 9.8 Printing with Field Widths and Precision
• If the field width is larger than the object being printed, the object is right justified by default.
• Field widths can be used with all conversion specifiers.
• Precision used with integer conversion specifiers indicates the minimum number of digits printed.
Zeros are prefixed to the printed value until the number of digits is equivalent to the precision.
• Precision used with floating-point conversion specifiers e, E and f indicates the number of digits
that appear after the decimal point. Precision used with floating-point conversion specifiers g and
G indicates the number of significant digits to appear.
• Precision used with conversion specifier s indicates the number of characters to be printed.
• The field width and the precision can be combined by placing the field width, followed by a decimal point, followed by the precision between the percent sign and the conversion specifier.
• It’s possible to specify the field width and the precision through integer expressions in the argument list following the format control string. To do so, use an asterisk (*) for the field width or
precision. The matching argument in the argument list is used in place of the asterisk.
Section 9.9 Using Flags in the printf Format Control String
• The - flag left justifies its argument in a field.
• The + flag prints a plus sign for positive values and a minus sign for negative values. The space
flag prints a space preceding a positive value that’s not displayed with the + flag.
• The # flag prefixes 0 to octal values and 0x or 0X to hexadecimal values, and forces the decimal
point to be printed for floating-point values printed with e, E, f, g or G.
• The 0 flag prints leading zeros for a value that does not occupy its entire field width.
Section 9.10 Printing Literals and Escape Sequences
• Most literal characters to be printed in a printf statement can simply be included in the format
control string. However, there are several “problem” characters, such as the quotation mark (")
that delimits the format control string itself. Various control characters, such as newline and tab,
must be represented by escape sequences. An escape sequence is represented by a backslash (\),
followed by a particular escape character.
Section 9.11 Reading Formatted Input with scanf
• Precise input formatting is accomplished with the scanf library function.
400
Chapter 9 C Formatted Input/Output
• Integers are input with scanf with the conversion specifiers d and i for optionally signed integers
and o, u, x or X for unsigned integers. The modifiers h and l are placed before an integer conversion specifier to input a short or long integer, respectively.
• Floating-point values are input with scanf with the conversion specifiers e, E, f, g or G. The modifiers l and L are placed before any of the floating-point conversion specifiers to indicate that the
input value is a double or long double value, respectively.
• Characters are input with scanf with the conversion specifier c.
• Strings are input with scanf with the conversion specifier s.
• A scan set scans the characters in the input, looking only for those characters that match characters contained in the scan set. When a character is matched, it’s stored in a character array. The
scan set stops inputting characters when a character not contained in the scan set is encountered.
• To create an inverted scan set, place a caret (^) in the square brackets before the scan characters.
This causes characters input with scanf and not appearing in the scan set to be stored until a
character contained in the inverted scan set is encountered.
• Address values are input with scanf with the conversion specifier p.
• Conversion specifier n stores the number of characters input previously in the current scanf. The
corresponding argument is a pointer to int.
• The conversion specifier %% with scanf matches a single % character in the input.
• The assignment suppression character reads data from the input stream and discards the data.
• A field width is used in scanf to read a specific number of characters from the input stream.
Terminology
(quotation mark) 391
assignment suppression character 397
# flag 390
% character in a conversion specifier 380
%% conversion specifier 385
%c conversion specifier 384, 394
%E conversion specifier 382
%e conversion specifier 382
%f conversion specifier 382
%g (or %G) conversion specifier 383
%i conversion specifier 393
%p conversion specifier 385
%s conversion specifier 384, 394
%u conversion specifier 381
+ flag 389
0 (zero) flag 390
caret (^) 396
conversion specification 380
conversion specifier 380
exponential notation 382
field width 380
"
*
flag 380
format control string 380
integer conversion specifier 381
inverted scan set 396
left justification 380
length modifier 381
literal character 380
precision 380
printf function 380
right justification 380
rounding 380
scan set 395
scanf function 380
scientific notation 382
space flag 390
standard error stream (stderr) 380
standard input stream (stdin) 380
standard output stream (stdout) 380
<stdio.h> header file 380
stream 380
whitespace characters 397
Self-Review Exercises
9.1
Fill in the blanks in each of the following:
a) All input and output is dealt with in the form of
.
Self-Review Exercises
b)
c)
d)
e)
f)
g)
h)
i)
j)
k)
l)
m)
n)
o)
p)
q)
r)
s)
t)
u)
v)
9.2
401
The
stream is normally connected to the keyboard.
stream is normally connected to the computer screen.
The
Precise output formatting is accomplished with the
function.
The format control string may contain
,
,
,
and
.
The conversion specifier
or
may be used to output a signed decimal
integer.
,
and
are used to display unThe conversion specifiers
signed integers in octal, decimal and hexadecimal form, respectively.
The modifiers
and
are placed before the integer conversion specifiers to indicate that short or long integer values are to be displayed.
The conversion specifier
is used to display a floating-point value in exponential notation.
is placed before any floating-point conversion specifier to indiThe modifier
cate that a long double value is to be displayed.
The conversion specifiers e, E and f are displayed with
digits of precision to
the right of the decimal point if no precision is specified.
and
are used to print strings and characThe conversion specifiers
ters, respectively.
character.
All strings end in the
The field width and precision in a printf conversion specifier can be controlled with
for the field width or for the precision
integer expressions by substituting a(n)
and placing an integer expression in the corresponding argument of the argument list.
The
flag causes output to be left justified in a field.
flag causes values to be displayed with either a plus sign or a minus sign.
The
Precise input formatting is accomplished with the
function.
A(n)
is used to scan a string for specific characters and store the characters in
an array.
can be used to input optionally signed octal, decimal
The conversion specifier
and hexadecimal integers.
can be used to input a double value.
The conversion specifiers
The
is used to read data from the input stream and discard it without assigning it to a variable.
can be used in a scanf conversion specifier to indicate that a specific
A(n)
number of characters or digits should be read from the input stream.
Find the error in each of the following and explain how it can be corrected.
a) The following statement should print the character 'c'.
printf( "%s\n", 'c' );
b) The following statement should print 9.375%.
printf( "%.3f%", 9.375 );
c) The following statement should print the first character of the string "Monday".
printf( "%c\n", "Monday" );
d)
e)
f)
g)
9.3
puts( ""A string in quotes"" );
printf( %d%d, 12, 20 );
printf( "%c", "x" );
printf( "%s\n", 'Richard' );
Write a statement for each of the following:
a) Print 1234 right justified in a 10-digit field.
b) Print 123.456789 in exponential notation with a sign (+ or -) and 3 digits of precision.
c) Read a double value into variable number.
402
Chapter 9 C Formatted Input/Output
d)
e)
f)
g)
Print 100 in octal form preceded by 0.
Read a string into character array string.
Read characters into array n until a nondigit character is encountered.
Use integer variables x and y to specify the field width and precision used to display the
double value 87.4573.
h) Read a value of the form 3.5%. Store the percentage in float variable percent and eliminate the % from the input stream. Do not use the assignment suppression character.
i) Print 3.333333 as a long double value with a sign (+ or -) in a field of 20 characters with
a precision of 3.
Answers to Self-Review Exercises
9.1
a) streams. b) standard input. c) standard output. d) printf. e) conversion specifiers,
flags, field widths, precisions, literal characters. f) d, i. g) o, u, x (or X). h) h, l. i) e (or E). j) L.
k) 6. l) s, c. m) NULL ('\0'). n) asterisk (*). o) - (minus). p) + (plus). q) scanf. r) scan set. s) i.
t) le, lE, lf, lg or lG. u) assignment suppression character (*). v) field width.
9.2
a) Error: Conversion specifier s expects an argument of type pointer to char.
Correction: To print the character 'c', use the conversion specifier %c or change
'c' to "c".
b) Error: Trying to print the literal character % without using the conversion specifier %%.
Correction: Use %% to print a literal % character.
c) Error: Conversion specifier c expects an argument of type char.
Correction: To print the first character of "Monday" use the conversion specifier %1s.
d) Error: Trying to print the literal character " without using the \" escape sequence.
Correction: Replace each quote in the inner set of quotes with \".
e) Error: The format control string is not enclosed in double quotes.
Correction: Enclose %d%d in double quotes.
f) Error: The character x is enclosed in double quotes.
Correction: Character constants to be printed with %c must be enclosed in single
quotes.
g) Error: The string to be printed is enclosed in single quotes.
Correction: Use double quotes instead of single quotes to represent a string.
9.3
a)
b)
c)
d)
e)
f)
g)
h)
i)
printf( "%10d\n", 1234 );
printf( "%+.3e\n", 123.456789 );
scanf( "%lf", &number );
printf( "%#o\n", 100 );
scanf( "%s", string );
scanf( "%[0123456789]", n );
printf( "%*.*f\n", x, y, 87.4573 );
scanf( "%f%%", &percent );
printf( "%+20.3Lf\n", 3.333333 );
Exercises
9.4
Write a printf or scanf statement for each of the following:
a) Print unsigned integer 40000 left justified in a 15-digit field with 8 digits.
b) Read a hexadecimal value into variable hex.
c) Print 200 with and without a sign.
d) Print 100 in hexadecimal form preceded by 0x.
Exercises
403
e) Read characters into array s until the letter p is encountered.
f) Print 1.234 in a 9-digit field with preceding zeros.
g) Read a time of the form hh:mm:ss, storing the parts of the time in the integer variables
hour, minute and second. Skip the colons (:) in the input stream. Use the assignment
suppression character.
h) Read a string of the form "characters" from the standard input. Store the string in
character array s. Eliminate the quotation marks from the input stream.
i) Read a time of the form hh:mm:ss, storing the parts of the time in the integer variables
hour, minute and second. Skip the colons (:) in the input stream. Do not use the assignment suppression character.
9.5
Show what each of the following statements prints. If a statement is incorrect, indicate why.
a) printf( "%-10d\n", 10000 );
b) printf( "%c\n", "This is a string" );
c) printf( "%*.*lf\n", 8, 3, 1024.987654 );
d) printf( "%#o\n%#X\n%#e\n", 17, 17, 1008.83689 );
e) printf( "% ld\n%+ld\n", 1000000, 1000000 );
f) printf( "%10.2E\n", 444.93738 );
g) printf( "%10.2g\n", 444.93738 );
h) printf( "%d\n", 10.987 );
9.6
Find the error(s) in each of the following program segments. Explain how each error can be
corrected.
a) printf( "%s\n", 'Happy Birthday' );
b) printf( "%c\n", 'Hello' );
c) printf( "%c\n", "This is a string" );
d) The following statement should print "Bon Voyage":
printf( ""%s"", "Bon Voyage" );
e)
char day[] = "Sunday";
printf( "%s\n", day[ 3 ] );
f) puts( 'Enter your name: ' );
g) printf( %f, 123.456 );
h) The following statement should print the characters 'O' and 'K':
printf( "%s%s\n", 'O', 'K' );
i)
char s[ 10 ];
scanf( "%c", s[ 7 ] );
9.7
(Differences Between %d and %i) Write a program to test the difference between the %d and
%i conversion specifiers when used in scanf statements. Ask the user to enter two integers separated
by a space. Use the statements
scanf( "%i%d", &x, &y );
printf( "%d %d\n", x, y );
to input and print the values. Test the program with the following sets of input data:
10
-10
010
0x10
10
-10
010
0x10
9.8
(Printing Numbers in Various Field Widths) Write a program to test the results of printing
the integer value 12345 and the floating-point value 1.2345 in various size fields. What happens
when the values are printed in fields containing fewer digits than the values?
404
Chapter 9 C Formatted Input/Output
9.9
(Rounding Floating-Point Numbers) Write a program that prints the value
rounded to the nearest digit, tenth, hundredth, thousandth and ten-thousandth.
100.453627
9.10 (Temperature Conversions) Write a program that converts integer Fahrenheit temperatures
from 0 to 212 degrees to floating-point Celsius temperatures with 3 digits of precision. Perform the
calculation using the formula
celsius = 5.0 / 9.0 * ( fahrenheit - 32 );
The output should be printed in two right-justified columns of 10 characters each, and the Celsius
temperatures should be preceded by a sign for both positive and negative values.
9.11 (Escape Sequences) Write a program to test the escape sequences \', \", \?, \\, \a, \b, \n,
\r and \t. For the escape sequences that move the cursor, print a character before and after printing
the escape sequence so it’s clear where the cursor has moved.
9.12 (Printing a Question Mark) Write a program that determines whether ? can be printed as
part of a printf format control string as a literal character rather than using the \? escape sequence.
9.13 (Reading an Integer with Each scanf Conversion Specifier) Write a program that inputs the
value 437 using each of the scanf integer conversion specifiers. Print each input value using all the
integer conversion specifiers.
9.14 (Outputting a Number with the Floating-Point Conversion Specifiers) Write a program
that uses each of the conversion specifiers e, f and g to input the value 1.2345. Print the values of
each variable to prove that each conversion specifier can be used to input this same value.
9.15 (Reading Strings in Quotes) In some programming languages, strings are entered surrounded by either single or double quotation marks. Write a program that reads the three strings suzy,
"suzy" and 'suzy'. Are the single and double quotes ignored by C or read as part of the string?
9.16 (Printing a Question Mark as a Character Constant) Write a program that determines
whether ? can be printed as the character constant '?' rather than the character constant escape sequence '\?' using conversion specifier %c in the format control string of a printf statement.
9.17 (Using %g with Various Precisions) Write a program that uses the conversion specifier g to
output the value 9876.12345. Print the value with precisions ranging from 1 to 9.
C Structures, Unions, Bit
Manipulation and
Enumerations
10
But yet an union in partition.
—William Shakespeare
I could never make out what
those damned dots meant.
—Winston Churchill
Objectives
In this chapter, you’ll:
■
Create and use structures,
unions and enumerations.
■
Pass structures to functions
by value and by reference.
■
Use typedefs to create aliases
for existing type names.
■
Manipulate data with the
bitwise operators.
■
Create bit fields for storing
data compactly.
406
Chapter 10 C Structures, Unions, Bit Manipulation and Enumerations
10.1 Introduction
10.2 Structure Definitions
10.2.1
10.2.2
10.2.3
10.2.4
10.3
10.4
10.5
10.6
10.7
Self-Referential Structures
Defining Variables of Structure Types
Structure Tag Names
Operations That Can Be Performed on
Structures
Initializing Structures
Accessing Structure Members
Using Structures with Functions
typedef
Example: High-Performance Card
Shuffling and Dealing Simulation
10.8 Unions
10.8.2 Operations That Can Be Performed on
Unions
10.8.3 Initializing Unions in Declarations
10.8.4 Demonstrating Unions
10.9 Bitwise Operators
10.9.1 Displaying an Unsigned Integer in Bits
10.9.2 Making Function displayBits
More Scalable and Portable
10.9.3 Using the Bitwise AND, Inclusive OR,
Exclusive OR and Complement
Operators
10.9.4 Using the Bitwise Left- and RightShift Operators
10.9.5 Bitwise Assignment Operators
10.10 Bit Fields
10.11 Enumeration Constants
10.12 Secure C Programming
10.8.1 Union Declarations
Summary | Terminology | Self-Review Exercises | Answers to Self-Review Exercises | Exercises |
Making a Difference
10.1 Introduction
Structures—sometimes referred to as aggregates—are collections of related variables under one name. Structures may contain variables of many different data types—in contrast
to arrays, which contain only elements of the same data type. Structures are commonly
used to define records to be stored in files (see Chapter 11, C File Processing). Pointers and
structures facilitate the formation of more complex data structures such as linked lists,
queues, stacks and trees (see Chapter 12, C Data Structures). We’ll also discuss:
•
typedefs—for
creating aliases for previously defined data types
•
unions—derived data types like
structures, but with members that share the same
storage space
•
bitwise operators—for manipulating the bits of integral operands
•
bit fields—unsigned int or int members of structures or unions for which you
specify the number of bits in which the members are stored, helping you pack information tightly
•
enumerations—sets of integer constants represented by identifiers.
10.2 Structure Definitions
Structures are derived data types—they’re constructed using objects of other types. Consider the following structure definition:
struct card {
char *face;
char *suit;
}; // end struct card
10.2 Structure Definitions
407
Keyword struct introduces a structure definition. The identifier card is the structure tag,
which names the structure definition and is used with struct to declare variables of the
structure type—e.g., struct card. Variables declared within the braces of the structure
definition are the structure’s members. Members of the same structure type must have
unique names, but two different structure types may contain members of the same name
without conflict (we’ll soon see why). Each structure definition must end with a semicolon.
Common Programming Error 10.1
Forgetting the semicolon that terminates a structure definition is a syntax error.
The definition of struct card contains members face and suit, each of type char *. Structure members can be variables of the primitive data types (e.g., int, float, etc.), or aggregates, such as arrays and other structures. As we saw in Chapter 6, each element of an array
must be of the same type. Structure members, however, can be of different types. For example, the following struct contains character array members for an employee’s first and last
names, an unsigned int member for the employee’s age, a char member that would contain
'M' or 'F' for the employee’s gender and a double member for the employee’s hourly salary:
struct employee {
char firstName[ 20 ];
char lastName[ 20 ];
unsigned int age;
char gender;
double hourlySalary;
}; // end struct employee
10.2.1 Self-Referential Structures
A structure cannot contain an instance of itself. For example, a variable of type struct employee cannot be declared in the definition for struct employee. A pointer to struct employee,
however, may be included. For example,
struct employee2 {
char firstName[ 20 ];
char lastName[ 20 ];
unsigned int age;
char gender;
double hourlySalary;
struct employee2 person; // ERROR
struct employee2 *ePtr; // pointer
}; // end struct employee2
struct employee2 contains an instance of itself (person), which is an error. Because ePtr is
a pointer (to type struct employee2), it’s permitted in the definition. A structure containing
a member that’s a pointer to the same structure type is referred to as a self-referential structure. Self-referential structures are used in Chapter 12 to build linked data structures.
10.2.2 Defining Variables of Structure Types
Structure definitions do not reserve any space in memory; rather, each definition creates a
new data type that’s used to de