Introduction to C Language

UNIT 2 INTRODUCTION TO C LANGUAGE

Structure
2.0 Introduction 2.1 Unit Objectives 2.2 History of C Language
2.2.1 2.2.2 2.2.3 2.2.4 2.3.1 2.3.2 2.3.3 2.3.4 Developing a C Program Source Code Object Code Linking and Loading Executing a C Program in the UNIX System Entering Program Compilation Execution

NOTES

2.3 Program Execution

2.4 Sample C Program

2.4.1 Variation in the Main Function

2.5 Tokens
2.5.1 2.5.2 2.5.3 2.5.4 The C Character Set Identifiers Keywords Data Types

2.6 Variables
2.6.1 Size of Variables

2.7 Constants
2.7.1 2.7.2 2.7.3 2.7.4 2.7.5 2.7.6 Integer Constants Character Constants Floating Point or Real Numbers Enumeration Constant String Constants Symbolic Constants

2.8 Type Modifiers 2.9 Escape Sequences 2.10 Arrays

2.10.1 Array Declaration

2.11 2.12 2.13 2.14 2.15 2.16

Expressions and Statements Summary Key Terms Answers to Check Your Progress Questions and Exercises Further Reading

2.0 INTRODUCTION
In the previous unit, you were introduced to programming. In this unit, you will get indepth understanding of the C language. Dennis Ritchie developed the C language in 1972 under the influence of BCPL and B. The language is, however, rich in data types unlike the other two languages. It is universal. It can be developed and executed in any platform that has a development environment for C. It can run under UNIX, LINUX and even in parallel computers. In this unit, you will also gain knowledge of tokens. There are six classes of tokens in the C programming language, namely keyword, identifier, constant, string literal, operator and punctuators. The unit also discusses arrays, which is

Self-Instructional Material

25

Introduction to C Language

another form of data type. There can be arrays of integers, arrays of characters and arrays of floating-point numbers. An array means a collection of a number of integers or floats or items of the same data type. An array contains data of the same type.

NOTES

2.1 UNIT OBJECTIVES

After going through this unit, you will be able to: Elucidate the history of the C language Develop a C program Explain different types of tokens Describe variables and constants and their various types Define arrays and declare them Understand expressions, statements and symbolic constants

2.2 HISTORY OF C LANGUAGE

Martin Richards developed a language called BCPL in the mid-1960s. It had many good features. In the year 1970, Ken Thompson, also working at AT&T Bell Labs USA, developed a language called B, which was influenced by BCPL. He developed B as a compact language for system programming. Subsequently, in 1972, Dennis Ritchie, also working at Bell Labs, designed and implemented C language on the UNIX operating system on a Digital Equipment Corporation (DEC) PDP-11 computer. The C language was, therefore, developed under the influence of BCPL and B. It is, however, rich in data types unlike the other two languages. The B language can be considered as the C language without types. C added data types such as char and float to B. By 1973, the C language was ready. It further added new features between 1973 and 1980 and was used to write the UNIX kernel for the PDP-11 computer. The languages BCPL, B and C are procedure-oriented languages similar to FORTRAN. However, they are more compact having few keywords. Simple compilers translate them. They use programs in the standard library for input/output and other interactions with the operating system. During the 1980s, the C language compilers became available for most computer architectures and operating systems. It became a programming tool in PC, which increased its popularity. Ritchie along with Kernighan published a book The C Programming Language in 1978. The Kernighan and Ritchie (K&R) description of the language became a kind of industry standard, popularly called K&R C. Since then, as many C compilers came into existence and there were minor compatibility problems due to variations in implementation. The American National Standards Institute (ANSI) formed a committee (X3JII) for bringing out a standard for the C language. This committee brought out a standardized definition of the C language in 1989. The international standard on the C language, namely ISO/IEC 9899 was released in 1990 by adopting the ANSI standard on the C language. This standard is called C90. Subsequently, it was revised in 1999. The revised standard is called C99. Thus, there are three forms of C, namely K&R C, C90 and C99. Even thereafter, minor amendments to the standard are taking place. You will learn the C language that conforms to the ISO/IEC 9899: 1990 standard, hereinafter called the Standard.
26 Self-Instructional Material

Note: ISOInternational Organization for Standardization IECInternational Electro-technical Commission

Introduction to C Language

2.2.1 Developing a C Program

A number of integrated development environment (IDE) for developing and executing C programs are available in the market. Freeware compilers for C and a host of other languages are available on the Internet from an organization called The Free Software Foundation. This is called gcc C of GNU. It can be downloaded from the following URLs: http://www.gnu.ai.mit.edu/software/gec to work with UNIX operating system http://www.delorie.com/djgpp/ to work under DOS/ Windows environment A typical methodology for the development of a C program using GNU C in the Windows environment in the DOS prompt is given in Figure 2.1.
START

NOTES

Design algorithm

Enter source code in text editor Source Code

Save file

Compile Yes Errors No Object code of the relevant C library functions Get object code

Edit source code

eg. <stdio.h>
Operating systemrelated code Other object files Linker

Yes Errors No Get executable code

Execute run or

Check result

END

Fig. 2.1 Steps involved in developing a C Program

Self-Instructional Material

27

Introduction to C Language

The various steps involved in program development as given in Figure 2.1 are briefly discussed below.

2.2.2 Source Code

NOTES
The program statements written in a high-level language are referred to as source code. The source code is entered in a text editor. Before attempting to enter the source code, you should formulate the algorithm and document it in pseudo code. The pseudo code may be converted into the C language code and typed in the text editor available in the system. After entering the code, it may be saved in a file with file name .c extension. You can save the file as Example 2.1.c

2.2.3 Object Code

Now you are ready to compile the source code. You have to call the compiler and submit your source code for compilation. If you are using a GNU compiler, you type it in the DOS Command for compilation as follows:
gcc2.1.c

If you are using Turbo C++, you select compile and click. After the compilation process is over, there are two possible outcomes: There are errors in the source code. There are no errors. If there are no errors, you get an object code file with the same filename with .o extension automatically. In case of the above-mentioned example, you will get Ex 2.1.c. If there are errors, the compiler will give error messages. The error messages will give the line numbers in the program along with the type of error. The programmer should note the line numbers and the error messages and then edit the source file. The statements in the line numbers should be checked for errors. Therefore, the programmer should carefully examine the program and correct all the errors and recompile. When there are no errors, the compiler will automatically generate the object code and save it with a .o extension with the file name same as that of the corresponding source code. The object code is in the machine language. However, it is not yet ready for execution. If your program contains more than one source code file, all the source files are to be compiled separately to get the corresponding object code. Assume that you have another source file called px.c. You will get px.o when the compilation is successful.

2.2.4 Linking and Loading

During this process, you combine all the object files. In this case, Ex 2.1.c and px.o. Additionally, you add the object code corresponding to the header files included in the source code or program under execution. For instance, you will combine the object code corresponding to the printf() function in the <stdio.h> file. Another important code is that corresponding to the particular operating system of the computer system. This is essential for executing the program in the given hardware and the operating system. The linker to get an executable file for the source file will combine all these. The executable file will be in the machine code. It will be generated automatically on successful linking and loaded for execution. You get Ex 2.1.c.exe in this case after successful linking.

28 Self-Instructional Material

2.3 PROGRAM EXECUTION

In the DOS Prompt, you can execute the program by typing Ex 2.1.c.exe. If you are using Turbo C++, you can select RUN and click. The result will be displayed by the computer system on the monitor. The programmer may check the specific methodology for all the above steps in the development environment used by him. However, it will not be difficult to write a program and start executing the program. The programs may also be executed under other operating systems such as UNIX, DOS and in any other computer hardware such as Sun Workstations and IBM computer systems ranging from PCs to mainframes, etc.

Introduction to C Language

NOTES

2.3.1 Executing a C Program in the UNIX System

The C language is universal. The examples given in the book were developed and executed in a PC running the Windows operating system. However, it can be developed and executed in any platform that has a development environment for C. It can run under UNIX, LINUX and even in parallel computers. UNIX was the system in which Dennis Ritchie executed the first C program. The following are the steps to be carried out for program development and execution in a system with a typical UNIX operating system:

2.3.2 Entering Program

The C program can be entered in one of the UNIX editors such as ed or vi. After entering the program, you have to save it with a lower case .c extension. For instance, you can save it as Example 2.1.c

2.3.3 Compilation
The compiler in the UNIX system is called cc. You can compile the above program by typing the following:
ccEx 2.1.C

If there are errors, the program will not compile. The errors will be listed. The errors are to be corrected. On clean compilation, the system generates an executable file called a.out.

2.3.4 Execution
You execute the above-mentioned file a.out and the result will appear on the screen. Whenever a file is compiled as mentioned above, and if there are no errors, the system will produce an executable file with the name a.out. Therefore, the old executable files will be erased. If you want to retain all the executable codes for future use, you have to rename it. From the above discussions, it will be clear that the object file is not available. On clean compilation, the object code produced is converted into an executable code and placed in a.out. Hence, the object file is not available. Suppose, you want to retain the executable code in Ex 2.1.c, you can compile the program as follows:
cc2.1.C

This will result in having a copy of the executable file in Ex 2.1.c

Self-Instructional Material

29

Introduction to C Language

Thus, you have to understand the applicable commands for editing, compiling, linking and executing the program in a computer system. Once it is learnt, it will hold well forever in the system.

NOTES

2.4 SAMPLE C PROGRAM

A program in the C language is given below, which displays a text. Take a close look at the program:
/* Example 2.1*/ /* program for displaying a text */ #include <stdio.h> int main() { printf (Om Vinayaga) ; }

Create a new file in the text editor. Then type in the program exactly as given above. Save this program as Example 2.1. Next, compile the program. The compiler after compilation will give a message. Look at the message box. If the C program was compiled in a C++ compiler, there may be warnings, but you can safely ignore them. However, the errors, if any, should not be ignored. The compiler will give the errors and line numbers. Sometimes, even an experienced programmer will find it difficult to understand the error messages. If you encounter the same difficulty in understanding the error messages, you need not worry. Open the program file again and check whether you have typed it exactly as in the book including the semicolons, quotation marks and brackets and the program is a working program. Keep checking and compiling till the compiler says success, meaning there is no error in the program after compilation. Now, you have to link the object code to get the executable code. When there are no errors, even after linking, you have to execute or run the program. Use the right command for run. On execution, you will get the result as follows: Result of the program
Om Vinayaga

You have succeeded in establishing a communication link with the computer. You can now talk to it regularly by learning new commands and using better communication methods. You have now become familiar with the methodology for learning a programming language. The steps involved are summarized as follows: Type the program in a text editor. Save and give a name to the program. Compile the program. Look for errors and correct them. Link the object code to produce an executable code. Execute/Run the program and analyse the result. The methodology for the preparation of the source code, compilation, linking and execution may vary from system to system. Learn the correct operations of the IDE/system being used for the various steps mentioned above.
30 Self-Instructional Material

2.4.1 Variation in the Main Function

Most compilers have implemented K&R C. Some have implemented C90. Not many have implemented the new features introduced in C99. Hence, it is important that the reader checks what standard his compiler supports. In the Example 2.1, you did not explicitly return anything. On successful execution, the program returns zero. Return of any other integer means that the program execution was not successful. Hence, you can add a statement return 0 at the end of the program. The main function did not receive any value, which was indicated by the empty parentheses. You can indicate nothing by the keyword void. The modified program including the above two features is as follows:
/* Example 2.2*/ /* program for displaying a text */ #include <stdio.h> int main(void) { printf (Om Vinayaga) ; return 0; }

Introduction to C Language

NOTES

Result of the program

Om Vinayaga

Thus, both the above programs have carried out the task successfully. However, in the rest of the book, the style given in the previous program will be used. The reader may adopt the style that works in his compiler.

2.5 TOKENS
The Standard defines six classes of tokens in the C programming language as follows: Keyword Identifier Constant String literal Operator Punctuators

Tokens are similar to the atomic elements or building blocks of a program. A C program is constructed using tokens. There are certain other building blocks of a program that do not form part of any of the above. They are as follows: Blanks Horizontal tabs Vertical tabs New line characters Form feed Comments
Self-Instructional Material

31

Introduction to C Language

2.5.1 The C Character Set

The C language supports and implements the American Standard Code for Information Interchange (ASCII) for representing characters. The ASCII uses 7 bits for representing each character or digit. The characters are coded from 0000000 (decimal 0) to 1111111 (decimal 127). Therefore, the ASCII consists of a code for 128 characters in all. The ASCII values (decimal equivalent of the 7 bits) of some alphabets and digits are given in Table 2.1.
Table 2.1 ASCII Values of Selected Alphabets ASCII Value 48 49 57 65 66 67 89 90 97 98 121 122 Character or Digit 0 1 9 A B C Y Z a b y z

NOTES

The digits and alphabets are organized sequentially and hence, it is easy to get the ASCII value; for instance, the ASCII value of D is 68, E is 69, 8 is 56, x is 120 and so on. The ASCII table is given in Annexure 1.

2.5.2 Identifiers
Any name is an identifier. Just as the name of a person, street or city helps in the identification of a person or a street or a city, the identifier in the C language assigns names to files, functions, constants, variables, etc. An identifier in the C language is defined as a sequence of alphanumeric characters, i.e., alphabets or digits. The first character of an identifier has to be an alphabet. In the C language, lowercase alphabets and uppercase alphabets are considered to be different. For instance, VAR and var represent different names in the C language. VALID IDENTIFIERS
C1 PROC1 P34 VAR_1 EX1 a bc Ual1 Aa

INVALID IDENTIFIERS
1PROGA 4.3 A-B

32 Self-Instructional Material

Any function name is also an identifier. For instance, printf is the name of the function available with the C language system. The function helps in printing. Therefore, identifiers can be constructed with alphabets (A...Z), (a...z), (0...9). In addition, underscore can also be used in identifiers. Unless otherwise specified, small letters are usually used for identifiers.

Introduction to C Language

NOTES

2.5.3 Keywords
These are also known as reserved words in C. In the first program, int is a reserved word or keyword. They have a specific meaning to the compiler. They should be used for giving specific instructions to the computer. These words cannot be used for any other purpose such as naming a variable. C is a very concise language containing only 32 reserved words and this is one of its strengths. Common statements, such as print, input, etc., are implemented through library functions in C, giving relief to programmers and reducing the size of code as compared to other programming languages. This makes the task of programming simple. Now take a look at the keywords given in Table 2.2. You will use most of them in the book. Their meaning will become clear as you read the book.
Table 2.2 C Keywords auto const double float int short struct unsigned break continue else for long signed switch void case default enum goto register sizeof typedef volatile char do extern if return static union while

2.5.4 Data Types

Data is used in a program to get information. In a program used to find out the greater of two numbers, the numbers are data and the output, which says which number is greater, is information. C is a versatile language and handles many different types of data in an elegant manner. Bit stands for binary digit, i.e., 0 or 1. Each byte is a collection of 8 bits, i.e., 8 consecutive bits of 0 or 1. Data is handled in a computer generally in terms of bytes and therefore, will be in the form of multiples of 8 bits. Each ASCII character is represented by one byte.

Fundamental data types

An item that holds data is also called an object. An object has a name or identifier associated with it. Each object can hold a specific type of data. There are five basic data types in C, as follows: Character Integer Real numbers Void (comprising an empty set of values) Enum (which will be introduced later)
Self-Instructional Material

33

Introduction to C Language

NOTES

You have to understand how a computer works. Assume that two numbers a and b are to be multiplied. First of all, the two numbers have to be stored in the memory. Then the required calculation has to be performed. The result has also to be stored in the memory. Each number is of a specific data type; for instance, all three of them can be declared to be integers. Each data type occupies a specific size in the memory. What does one mean by size? It is the amount of storage space required; each bit needs one storage space. One byte needs eight storage spaces. If a number is of type integer declared as int, it is stored in 2 bytes. The number depending on its type gets stored in different forms. If a number is of float type, it takes 4 bytes to store it. All characters can be represented according to the ASCII table and hence, 1 byte, i.e., 8 bits are good enough to store a character, which is represented as char. These sizes may vary from computer to computer. The header files <limits.h> and <float.h> contain information about the sizes of the data types. Real numbers can be expressed with single precision or double precision. Double precision means that real numbers can be expressed more precisely. Double precision also means more digits in mantissa. The type float means single precision and double means a double precision real number. Table 2.3 indicates the size of various data types.
Table 2.3 Size of Data Types Data Type char int float double Size 1 byte 2 bytes 4 bytes 8 bytes

Maximum and minimum magnitudes

The maximum and minimum values of data types are not limitless. For example, <limits.h> specifies the minimum and maximum magnitudes for integers and characters. Since char is stored in a byte, it is as good as a short integer. char can be stored as an unsigned character, which means all the 8 bits can be used to store it. The maximum value of an 8-bit number is 255 when all bits are 1. If it is a signed char, the first bit will be reserved for storing the sign. The sign bit will be 0 for a positive number and 1 for a negative number. The integer values of signed and unsigned chars are as follows: CHAR-BIT SCHAR SCHAR UCHAR MAX + MIN MAX 8 127 127 255 bits in a char maximum value of signed char minimum value of signed char maximum value of unsigned char

Now, let us look at the maximum and minimum magnitudes for the integer data type. They are: INT INT MAX + MIN 32767 32767 maximum value of int minimum value of int

34 Self-Instructional Material

Integer means signed integer. The data type integer occupies 2 bytes or 16 bits. The most significant bit is reserved for sign. It will be 0 for a positive number and 1 for a negative number. Therefore, you can easily calculate how the limits for the integer data types have been arrived at.

The standard has also provided for another type of integer called short int with the same maximum and minimum values.

Introduction to C Language

2.6 VARIABLES
The names of variables and constants are identifiers. The names are made up of alphabets, digits and underscore, but the first character of an identifier must be an alphabet. C allows up to 31 characters for the identifier (names) and therefore, the naming of the variables should be carried out in an easily understandable manner. For example, in the program for the calculation of, Simple interest I = pnr/100, You can declare them with actual names, p = principal, r = rate_of_interest, n = number_of_ years Naturally, programmers may not like typing long names for fear of making mistakes elsewhere in the program apart from being reluctant to increase their typing workload. Meaningful names can, however, be assigned to data types even with few letters. For example,
p = princ; r = intrate; n = years

NOTES

Some compilers may allow names with upto 31 (thirty-one) characters, but may consider the first eight characters for all purposes. Hence, a programmer could coin shorter yet meaningful names, instead of using single alphabets for variable names. One should, however, be careful not to use the reserved words, such as 32 keywords, for variable names as they have a specific meaning to the compiler. If they are used as variable names, then the compiler will get confused. Be careful not to use the reserved words as identifiers. A program to find out the square of integer 5 is given below.
/*Example 2.3*/ /*program to find square of 5*/ #include <stdio.h> int main() { printf(square of %d= %d, 5, 5*5); }

Result of the program

square of 5= 25

Check Your Progress

You have now achieved the objective of finding the square of 5. Later on, you may want to find out the square of another number, say 7, for example. You would have to write the same program again replacing 5 by 7 and then compile and run it. This would waste a lot of time. To save time, you can, therefore, write a general-purpose program as follows:
/*Example 2.4*/ /*program to find square of any given number*/ #include <stdio.h>

1. Who developed the C language and when? 2. What are tokens? How many classes of tokens are there? 3. What is an identifier? 4. What do you mean by keywords?
Self-Instructional Material

35

Introduction to C Language

int main() { int num;

NOTES

printf(Enter the integer whose square is to be found\n); scanf(%d, &num); printf(square of %d= %d, num, num*num); }

Here you define num as an integer variable. When & precedes num, it indicates the memory address of num. At the first printf, the message appears as it is and the cursor goes to the next line because of the new line character \n at the end of the message, but before the closing quotation mark. The next statement is new to you. It is used to receive an integer typed on the console. You can type in any integer number, and the number typed will be stored in the memory at the memory location named num. The purpose of the statement is, therefore, to get the integer (because of the appearance of %d within quotes) and it is stored at the memory address num. The next statement prints the number typed and its square. When you execute the program, the following message appears on the monitor: Enter the integer whose square is to be found. Since you want to find out the square of 25, type:
25

Promptly the reply will be as shown:

square of 25 = 625

The next time you may want to find out the square of another number, say 121. Then simply run the program and when prompted, type 121 to get the answer. Here, the number whose square has to be found out has been declared as a variable. The variable has a name and is stored at the location pointed to by the variable name. Therefore, the execution of the program for finding out the square of any number is easy with the above modification as in Example 2.4. Variables and constants are fundamental data types. A variable can be assigned only one value at a time, but can change value during program execution. A constant, as the name indicates, cannot be assigned a different value during program execution. For example, PI, if declared as a constant, cannot have its value varied in a given program. If PI has been declared as a constant = 3.14, it cannot be reassigned any other value in the program. Programs may declare a number of constants. Variables are similarly useful for any programming language. If PI has been declared as a variable, then it can be changed in the program to any value. This is one difference between a variable and a constant. Whether an identifier is constant or variable depends on how it is declared. Both variables and constants belong to one of the data types like int, float, etc. The convention in C is to indicate the names of constants by the upper case letters.
PI SIGMA
36 Self-Instructional Material

Variable names are, on the other hand, indicated by the lower case letters:
int a float xy

Introduction to C Language

2.6.1 Size of Variables

The C programmer should understand how much memory storage each variable type occupies in the IDE used by him. The following example will help you to find the size of each variable type. The result will be in terms of bytes occupied by the variable type. A new keyword sizeof is used to find out the size. The syntax for using the keyword is as follows:
sizeof (<data type>)

NOTES

or
sizeof (<expression>)

Consider the following example:

/*Example 2.5*/ /*program to find out the sizes of various types of integers*/ #include<stdio.h> int main() { printf(size of char =%d\n, sizeof(char)); printf(size of short=%d\n, sizeof(short)); printf(size of int =%d\n, sizeof(int)); printf(size of unsigned int=%d\n, sizeof(unsigned)); printf(size of long int=%d\n, sizeof(long)); printf(size of unsigned long int=%d\n, sizeof(unsigned long)); printf(size of float =%d\n, sizeof(float)); printf(size of double=%d\n, sizeof(double)); printf(size of long double%d\n, sizeof(long double)); }

Result of the program

size size size size size size size size size of of of of of of of of of char = 1 short = 2 int = 2 unsigned int = 2 long int = 4 unsigned long int = 4 float = 4 double = 8 long double = 10

Therefore, it is obvious that a long double occupies 10 bytes and stores long floatingpoint numbers with double precision. Note that the size of short int will be either equal to or less than the size of an integer variable.
Self-Instructional Material

37

Introduction to C Language

Variables, which require more than 1 byte for storage, will be stored consecutively in the memory.

NOTES

2.7 CONSTANTS
The following are the types of constants: Integer constant Character constant Float constant Enumeration constant String constant Symbolic constant All these types will now be explained:

2.7.1 Integer Constants

The following are the types of integers:
int unsigned int long unsigned long

Variations in Integer Types

You can use the sign bit also for holding the value. In such cases, the variable will be called unsigned int. The maximum value of an unsigned int will be equal to 65535 because you are using the Most Significant Bit (MSB) also for storing the value. The minimum value will obviously be 0. A long integer is represented as long int or simply long. The maximum and minimum values of long are as follows: LONG LONG MAX MIN + 2147483647 2147483647

Unless otherwise specified, integers or long integers will be signed, i.e., the first bit will be reserved for the sign. The long int obviously uses 4 bytes or 32 bits. The magnitudes of long can also be doubled by using an unsigned long integer denoted as unsigned long. However, integers are not suitable for very low values and very large values. This can be overcome by floating point or real numbers. An integer constant may be suffixed by the letter u or U to specify that it is an unsigned integer. Similarly, if the integer is suffixed with l or L, it signifies a long integer. If you specify unsigned long integer you suffix the constant with ul or UL.

38 Self-Instructional Material

The following are the examples of valid and invalid integers: Valid integers
+345 345 345 729u 729U 112345L 112345UL +112345l 112345l positive /* integer */ /* integer */ /* integer */ /* unsigned integer */ /* unsigned integer */ /* Long integer */ /* Unsigned Long integer */ /* Long integer */ /* Long integer - if no sign precedes, it is a number */ /* /* /* /* /* /* /* decimal point not allowed */ no comma allowed */ = blank not allowed */ exceeds the maximum */ unsigned cannot have + */ ( not allowed */ illegal characters */

Introduction to C Language

NOTES

Invalid integers
345.0 112, 345L 112 345UL 112890345L +112 345UL (345l 345s

You have so far considered only decimal numbers. C, however, entertains other types of numbers as well. The octal numbers will be preceded by 0 (zero). The following are the examples of valid and invalid octal numbers: Valid octal number
0346 0547 0120

Invalid octal number

0394 0 x 345 /* 8 or 9 are not allowed in an octal number */ /* prefix has to be zero only */

The C language also supports hexadecimal numbers. Here, since the base is 16, we use alphabets also in the numbers as given in Table 2.4.
Table 2.4

a b c d e f

or or or or or or

A B C D E F

for for for for for for

10 11 12 13 14 15

Additionally, hexadecimal numbers will be preceded by 0X or ox, i.e., zero followed by x.

Self-Instructional Material

39

Introduction to C Language

The following are the examples of valid and invalid hexadecimal numbers: Valid hexadecimal numbers
0x345 0xA132 0x100 0x20B
Invalid hexadecimal numbers

NOTES

0x, 123 0x 0A00

/* no comma */ /* cannot be empty */ /* x is missing */

2.7.2 Character Constants

A character constant is a single character enclosed in single quotes as in x. Characters can be alphabets, digits or special symbols. The following are the examples of valid and invalid character constants: Valid character constants
A Z C c

Invalid character constants

\n \t \u \b AA AA AA 1a

A character constant represents its integer value as defined in the character set of the machine. Therefore, you can add 2 characters. For example, the ASCII values of digit 1 = 49 and C = 67. When you add these values you get code 116 whose equivalent character is t. Now, verify this with the following example:
/* Example 2.6 demonstrates that chars can be treated like integers*/ #include <stdio.h> int main() { const char ALPHA1=1'; char alpha2=C; char alpha3; alpha3=ALPHA1+alpha2;
40 Self-Instructional Material

putchar(alpha3); }

Introduction to C Language

Result of the program

t

NOTES

Therefore, characters can be treated like integers as well, although they are declared as character variables and constants. Since characters are of type int, you could add them. Characters can also be defined as integers as in the following example:
/* Example 2.7 Demonstrates that a char can also be declared as int*/ #include <stdio.h> int main() { int x; x=1'+C; printf(x as integer=%d\n, x);/*x printed as integer*/ printf(x as character=%c\n, x);/*x printed as character*/ }

Result of the program

x as integer=116 x as character=t

2.7.3 Floating Point or Real Numbers

The difference between floating point and integer numbers are as follows: Integers are whole numbers without decimal points but a float has always a decimal point. Even if the number is a whole number, it is written with a decimal point. For instance, 42 is an integer, while 42.0 is floating-point number. Floating-point numbers occupy more space for storage as we have already seen. A real number in the simple form consists of a whole number followed by the decimal point and also one or more decimal numbers following the decimal point, which makes the fractional part. This form of representation is known as the fractional form. It must have a decimal point. It could be either positive or negative. As usual, the default sign is positive. No commas or blanks or special characters are allowed in between. The following are the examples of valid and invalid float types: Valid floats
144.00 226.012

Invalid floats
+144 /* no decimal point */ 1,44.0 /* comma not allowed */

Self-Instructional Material

41

Introduction to C Language

Scientific notation: Floating-point numbers can also be expressed in scientific notation. For example, 3.0 E2 is a floating-point number. The value of the number will be equal to 3.0 102 = 300.0 Instead of the upper case E, the lower case e can be used as in
0.453 e + 05, which will be equal to 0.453 105 = 45300

NOTES

There are two parts in the scientific notation of a real number, which are as follows: Mantissa (before E) Exponent (after E) In the scientific form, the following rules are to be observed: The mantissa part can be positive or negative. The exponent must have at least one digit, which can be a positive or negative integer. Remember the exponent cannot be a floating-point number. type float is a single precision number occupying a storage space of 4 bytes. type double represents floating-point numbers of double precision and hence occupies 8 bytes. If you look at the file <float.h> you will find the maximum and minimum floating-point numbers as shown: FLT MAX FLT MIN F or f no suffix L or l 1E 1E + 37 37 float double long double maximum floating point number minimum floating point number

Floating-point constants The constants are suffixed as shown:

If an integer is suffixed with L or l, then it is a long integer. If a float is suffixed with L or l, then it is a long double floating-point number. Examples Valid floating-point constants
1.0 e 5 123.0 f /* float 11123.05 /* double 23467.34 e 5 l /* long double */ */ */ */ */ */

Invalid real constants

245.0 456 1.0 e 5.0 /* invalid float, but valid double /* It is an integer /* exponent cannot be a real number

When they are declared as variables, they can be declared as follows:

float a = 3.12; float a, b, c; float val1;
42 Self-Instructional Material

float val2; long double val3;

Introduction to C Language

The values of constants cannot be altered in programs. They can be defined as follows:
const int PRINC = 1000; const float INT_RATE = 0.12 f;

NOTES

The values of PRINC and INT_RATE cannot be changed in the program even by introducing another assignment statement when they are declared as constants using const. The following example verifies this statement:
/*Example 2.8 Demonstrates that constants cannot be changed even with assignment statements. To verify, include statements 7, 8 & 9 in the program by removing the comment specifiers at the beginning of the program statement 7 and the end of statement 9*/ #include<stdio.h> main() { const int PRINC =1000; const float INTST=0.12f; printf(PRINCIPAL=%d INTEREST=%f\n, PRINC, INTST); /*PRINC =2000; INTST=0.24f; printf(PRINCIPAL=%d INTEREST=%f\n, PRINC, INTST);*/ }

Key in the example and execute the program. After successful compilation, you will get the result as given below. Result of the program
PRINCIPAL = 1000; INTEREST = 0.1200

Now include the second part of the program by removing /* and */ at statements 7 and 9, respectively. Earlier this was treated as a comment. Now this part will get included in the program. Now compile it. You will get a compilation error. This is due to your attempt to redefine the constants PRINC and INTST, which is not allowed. Incidentally, the technique of including or excluding a program segment at will using /* and */ is a convenient method for program development.

2.7.4 Enumeration Constant

The keyword for this data type is enum. You can define guardian as follows:
enum guardian { father, husband, guardian };
Self-Instructional Material

43

Introduction to C Language

NOTES

Note that there are no semicolons, but only a comma after the members except the last member. There is not even a comma after the last member. There is no conflict between both guardians. The top one is enum and the bottom one is a member of enum guardian. See the similarity between structure, union and enum. The enum variables can be declared as follows:
enum guardian emp1, emp2, emp3;

This is similar to structures and unions. The first part is the declaration of the data type. The second part is the declaration of the variable. The initial values can be assigned in a simpler manner as given below:
emp1 = husband; emp2 = guardian; emp3 = father;

You have to assign only those declared as part of the enum declaration. Assigning constants not declared will cause error. The compiler treats the enumerators given within the braces as constants. The first enumerator father will be treated as 0, the husband as 1 and the guardian as 2. Therefore, it is strictly as per the natural order starting from 0. The enumerated data type is never used alone. It is used in conjunction with other data types. You can write a program using enum and struct. It is shown as follows:
/*Example 2.9 enum within structure*/ #include <stdio.h> main() { enum guardian { father, husband, relative }; struct employee { char *name; float basic; char *birthdate; enum guardian guard; }emp[2]; int i; emp[0].name=RAM; emp[0].basic= 20000.00; emp[0].birthdate= 19/11/1948; emp[0].guard= father; emp[1].name=SITA; emp[1].basic= 12000.00; emp[1].birthdate= 19/11/1958;
44 Self-Instructional Material

emp[1].guard=husband; for(i=0;i<2;i++) { if( emp[i].basic ==12000) { printf(Name:%s\nbirthdate:%s\nguardian: %d\n, emp[i].name, emp[i]. birthdate, emp[i].guard); } } }

Introduction to C Language

NOTES

Result of the program

Name:SITA birthdate:19/11/1958 guardian: 1

The program clearly assigns the relationships between the employee and the guardian. enum guardian is the data type, and guard is a variable of this type. However, when you are printing emp [i]. guard, you are printing an integer. Hence 0, 1 or 2 will only be printed for the status, and this is a limitation. This can be overcome by modifying the program. The program modified with a switch statement is given below:
/*Example 2.10 expanding enum*/ #include <stdio.h> main() { enum guardian { father, husband, relative }; struct employee { char *name; float basic; char *birthdate; enum guardian guard; }emp[2]; int i; emp[0].name=RAM; emp[0].basic= 20000.00; emp[0].birthdate= 19/11/1948; emp[0].guard= father; emp[1].name=SITA; emp[1].basic= 12000.00; emp[1].birthdate= 19/11/1958; emp[1].guard=husband;
Self-Instructional Material

45

Introduction to C Language

NOTES

for(i=0;i<2;i++) { if( emp[i].basic ==12000) {printf(Name:%s\nbirthdate:%s\nguardian:, emp[i].name, emp[i]. birthdate); switch(emp[i].guard) { case 0:printf(father\n); break; case 1:printf(husband\n); break; case 2:printf(relative\n); break; } } } }

Result of the program

Name:SITA birthdate:19/11/1958 guardian:husband

Even though the conversion of an integer to actual name is additional work, enum is a useful construct since it improves readability in addition to the number of other advantages. For example, you can define boolean as follows:
enum boolean false, true }; {

Here false will contain an integer value 0 and true 1. This can be used to assign values for found in our sorting programs. You can define found as follows after declaring boolean:
enum boolean found;

You have allowed the system to assign integer values to the members of enum, but we can also assign specific values to the various members of enum. When values are assigned, then it takes precedence over what the system assigns. You can define boolean as:
enum boolean { yes = 1, no = 0 };

This is similar to defining using #define. The #define equivalent for this will be:
#define yes 1 #define no 0
46 Self-Instructional Material

Although #define and enum provide a way to associate symbolic names such as boolean with constants, there are differences between them. The differences are: (i) enum can generate values itself unlike #define where you have to specify the replacement constant. (ii) The compilers need not check the validity of what is stored in the enum variable, but the #define replacement constant will be checked for valildity. (iii) It is possible to print out the values of enum variables in symbolic form, but this is not possible with # define. Anyway, either of them can be used depending on the context.

Introduction to C Language

NOTES

2.7.5 String Constants

A character constant is a single character enclosed within single quotes. A string constant is a number of characters, arranged consecutively and enclosed within double quotes. Examples of valid strings:
God is within me

You may be surprised about the third string constant, which has no characters. This is called a NULL or empty string and is allowed in C. The string constant can contain blanks, special characters, quotation marks, etc. within the string. In order to distinguish the end of a string constant, the compiler places a null character \0 (back slash followed by zero) at the end of each string before the quotation mark. The null character is not visible when the string appears on the screen. The programmer does not include the null character either. It is the compiler which automatically inserts the null character at the end of every string. Invalid string:
Yoga /* should be enclosed in double quotes */

2.7.6 Symbolic Constants

The format for symbolic constant is as follows:
# define name constant

For example, we can define:

# define INITIAL 1

Which defines INITIAL as 1. The INITIAL type of definition is called symbolic constants. They are not variables and hence, they are not defined as part of the declarations of variables. They are specified on top of the program before the main function. The symbolic constants are to be written in capital or upper case letters. Wherever the symbolic constant names appear in the program, the compiler will replace them with the corresponding replacement constants defined in the # define statement. In this case, 1 will be substituted wherever INITIAL appears in the program. Note that there is no semicolon at the end of the # define statement.

Self-Instructional Material

47

Introduction to C Language

2.8 TYPE MODIFIERS

To summarize, there are four basic data types as given below:
char int float void

NOTES

The basic data types could be modified and a summary of all modified data types is given below: Basic data type char int Modified signed char unsigned char short unsigned int long long unsigned float double long double No. of bytes occupied 1 1 <=2 2 4 4 8 10

You have read about these modifiers and how their range gets affected by modifying the basic data types.

2.9 ESCAPE SEQUENCES

Escape sequences are quite useful in formatting input and output. You have already used \n, popularly called line feed, in the print statement to move to the next line. When the printer or a console monitor comes across \n, the cursor automatically goes to the first position in the next line. All such escape sequences start with a back slash (\) followed by a character. Some of the escape sequences are given below with their purposes. Escape sequence \a \b \t \v \n \f \r \0 Purpose bell or audible alert back spacego back by 1 space horizontal tab or go to next tab vertical tabs line feed or go to new line form feedused while printing carriage return null character

Although all escape sequences appear to be 2 characters wide, they are represented as a single character with the unique ASCII value.
48 Self-Instructional Material

2.10 ARRAYS
An array is another form of data type. There can be arrays of integers, arrays of characters and arrays of floating-point numbers. What does an array mean? It means a collection of a number of integers or floats or items of the same data type. An array contains data of the same type. For example, A = { 2, 3, 5, 7} Here A is an array of prime numbers. B = { Red, Green, Yellow} B is an array of colours. Thus, an array has a name, which is A in the former case and B in the latter case. How do you give a name to each element? You can use an index with the name of an array to indicate the elements. While in mathematics the first element can be called with an index [1], in C the first element is called with the index [0]. Index and subscript are used interchangeably to indicate the position of the element in an array. Thus,
A A A & [0] = [1] = [2] = A [3] 2 3 5 = 7

Introduction to C Language

NOTES

Similarly,
B [0] = Red B [1] = Green B [2] = Yellow

Array A has four elements and hence, the size of A is 4. Similarly, B has three elements and hence, its size is 3. Therefore, the first element of an array will have a subscript of 0 and the final element will have a subscript of n 1, where n is the size of the array.

2.10.1 Array Declaration

Assuming that there are 40 employees in an office and you want to store their ages, you would have to create 40 variables of type integer or float and store their ages. While such a definition seems alright, it would cause complications while programming. Instead, this can be declared as an array of size 40. For example,
int emp_age [40];

Here, you have declared an integer variable known as emp_age with a dimension of 40. A single variable has been declared to store the age of 40 employees. But for this feature, you would have to write 40 lines to declare the same with 40 different names. An array is a variable and hence, must be declared like any other variable. The naming convention is the same as in the case of other variables. The array name cannot be a reserved word and must be unique in the program. An array variable name and another ordinary variable name cannot be identical. Since there is no limit to variable names, do not use similar names for a variable and an array. What distinguishes an array variable from a single variable is the declaration of the dimension within the square brackets.

Self-Instructional Material

49

Introduction to C Language

NOTES

What does variable declaration do? It allots memory space for the variable. If you declare an array variable of type integer and dimension 40, then the computer will allot a memory size of 40 words for the variable contiguously. The last word is important. The elements in an array will be stored in consecutive memory locations. Since an integer needs 2 bytes for storage and if the memory is organized and addressed in terms of bytes, the following allocation would be carried out for Here, an integer variable has been declared, known as emp elements:
emp_age [0] emp_age [1] emp_age [2] 1000 1002 1004

If emp_age [0] is stored at the location with address 1000, emp_age [1] will be at 1002. It is enough for you to indicate the starting address to find the address of any of the elements of the array. The emp_age [2] will be stored at 1004. The formula for finding the address of element is emp_age [n] = starting address + n * 2. If emp_age had been defined as a float variable and if emp_age [0] starts at the location 1000, then emp_age [10] would be found at location 1040. The formula is: address of nth element = starting address + n * (size of variable). The important point to be noted is that if the starting address is known and the type of variable is known, the exact location where an element of the array is stored can be computed. An array has to contain elements of the same type. A float array cannot have integers or characters. A character array is nothing but a string. You must always specify the array size. You would normally expect to get an error message if the array size is exceeded. But, this does not happen in the C language. In the above example, if you try to read the value of emp_age [50] nothing will happen. No error message will be printed. Therefore, it is the responsibility of the programmer not to exceed the array size. Since we have a single subscript, the arrays declared so far are known as one-dimensional arrays. Examples of one-dimensional arrays are as follows:
int emp_age[100]; float mark[100]; char name[25]; /* no space between variable name & dimension */ /* name contains 25 characters */

2.11 EXPRESSIONS AND STATEMENTS

i + 2 is an expression. This expression consists of variable i, constant 2 and the operator +. Thus, expressions can be formed by a combination of the following: Data types (Variables and constants) Operators (Arithmetic operators, relational operators, logical operators) Such expressions should be meaningful in order to perform useful operations. In the example, 2 is added to i. a*a + 2*a*b + b*b is another example of an expression.

Check Your Progress

5. What is the difference between a character constant and a string constant? 6. Define array.

50 Self-Instructional Material

Statements
A statement is a sentence, which terminates with a semicolon in the C program. You are familiar with the assignment operator which is =. For example, in the statement:
int i = 3; int i is the declaration of the variable; i = 3 is an assignment statement and = is the assignment operator.

Introduction to C Language

NOTES

2.12 SUMMARY
In this unit, you have learned about the C language. C was preceded by two languages called BCPL and B. Though all the three languages BCPL, B and C are procedureoriented languages similar to FORTRAN, the C language removed the lacuna faced by the other two languages, namely BCPL and B. C is a more compact language having few keywords. Simple compilers translate them. They use programs in the standard library for input/output and other interactions with the operating system. During the 1980s, the C language compilers became available for most computer architectures and operating systems. It became a programming tool in PC, which increased its popularity. The C language is rich in data types and universal. It can be developed and executed in any platform that has a development environment for C. It can run under UNIX, LINUX and even in parallel computers. You also gained knowledge of tokens of which there are six classes, namely keyword, identifier, constant, string literal, operator and punctuators. The unit also discussed array, which is another form of data type. Arrays comprise integers, characters and floating-point numbers. An array is a collection of a number of integers or floats or items of the same data type. Simply put, an array contains data of the same type.

2.13 KEY TERMS

Source code: Program statements written in a high-level language are referred to as source code. Identifier: Any name is an identifier. An identifier in the C language assigns names to files, functions, constants, variables, etc. An identifier is defined as a sequence of alphanumeric characters, i.e., alphabets or digits. The first character of an identifier has to be an alphabet. Keywords: These are also known as reserved words in C. They have a specific meaning to the compiler and are used for giving specific instructions to the computer. These words cannot be used for any other purpose such as naming a variable. Statement: A statement is a sentence which terminates with a semicolon in the C program. Array: An array is another form of data type. Array means a collection of a number of integers or floats or items of the same data type. An array contains data of the same type.
Self-Instructional Material

51

Introduction to C Language

2.14 ANSWERS TO CHECK YOUR PROGRESS

1. Dennis Ritchie designed and implemented C language on the UNIX operating system on a Digital Equipment Corporation (DEC) PDP-11 computer in 1972. 2. Tokens are similar to atomic elements or building blocks of a program. A C program is constructed using tokens. There are six classes of tokens, which are keyword, identifier, constant, string literal, operator and punctuators. 3. Any name is an identifier. Just as the name of a person, street or city helps in the identification of a person or a street or a city, the identifier assigns names to files, functions, constants, variables, etc. An identifier in the C language is defined as a sequence of alphanumeric characters, i.e., alphabets or digits. The first character of an identifier has to be an alphabet. 4. These are also known as reserved words in C. They have a specific meaning to the compiler. They should be used for giving specific instructions to the computer. These words cannot be used for any other purpose such as naming a variable. 5. A character constant is a single character enclosed within single quotes. A string constant is a number of characters, arranged consecutively and enclosed within double quotes. 6. An array is another form of data type. There can be arrays of integers, arrays of characters and arrays of floating-point numbers. Array means a collection of a number of integers or floats or items of the same data type. An array contains data of the same type.

NOTES

2.15 QUESTIONS AND EXERCISES

Short-Answer Questions
1. Does C language have a reserved word for the print command? 2. How many classes of tokens are there in C? Name them. 3. In what ways are unsigned integer variables better than signed integers? 4. What are the features of octal numbers? 5. What is a floating-point number? 6. What is an escape sequence? 7. Check which of the following are valid integer constants: (a) 4.13 (b) z (c) 240UL (d) + 240 (e) ( 245 8. Check which of the following are valid character constants: (a) c (b) A (c) 35
52 Self-Instructional Material

(d) character (e) 0x123 4. State which of the following are valid floating-point constants: (a) 123 E 2 (b) 123.0 e 2 (c) 3350.0001L (d) 267.1 E 2 f (e) 225.12 e 2.5

Introduction to C Language

NOTES

Long-Answer Questions
1. Define data types. 2. What is the difference between floating number and integers? 3. Explain constants. 4. Explain Tokens with the help of examples. 5. What are variables? How are they declared in a C program? 6. Define enumeration constant with the help of C program. 7. What are arrays? How are they declared? Explain with the help of C program.

2.16 FURTHER READING

Gottfried, Byron S. Programming with C, 2nd edition. Schaums Outline Series. New York: McGraw-Hill, 1996. Forouzan, Behrouz A. and Richard F. Gilberg. Computer Science: A Structured Programming Approach Using C, 2nd edition. California: Thomson, Brooks Cole, 2000. Dey, Pradip and Manas Ghosh. Computer Fundamentals and Programming in C. New Delhi: Oxford Higher Education, 2006. Bronson, Gary J. A First Book of ANSI C, 3rd edition. California: Thomson, Brooks Cole, 2000. Kanetkar, Yashwant. Understanding Pointers in C. New Delhi: BPB Publication, 2001. Kanetkar, Yashwant. Let us C. New Delhi: BPB Publication, 1999. Kernighan, Brian W. and Dennis Ritchie. C Programming Language, 2nd edition. New Jersey: Prentice Hall, 1988. Foster, W. D. and L. S. Foster. C by Discovery. Boston: Addison-Wesley, 2005. Kanetkar, Yashwant. Working with C. New Delhi: BPB Publication, 2003. Horton, Ivor. Instant C Programming. New Jersey: Wrox Press (John Willey & Sons), 1995. Lawlor, Steven C. The Art of Programming Computer Science with C. New Jersey: West Publishing Company, 1996.

Self-Instructional Material

53