1		INTRODUCTION Two types of data can be used in C programs: constants and variables. This paper discusses variables. A variable is a named area of storage (also called object), and must have: type, the meanings to be attached to the values; a set of allowable operations; a range of values; predetermined memory requirements; scope, that is, the region of program in which the variable is recognised. There are others, but they are beyond the scope of this course. For example, the value stored in a variable of type int will be regarded as a whole number (no fractional part). Its allowable range of operations will include addition, subtraction, division and multiplication. Its memory requirements might be four bytes.
2		BASIC VARIABLE TYPES The three basic data types in C of interest to us in this course are characters (char), integers (int) and floating-point (float).
2.1		Char A character variable stores character codes. It can hold any member of the execution character set. The value actually held depends on the character set being used. Don't assume it's ASCII. A character variable occupies a byte of storage. The number of bits in a byte is defined by the Standard C macro CHAR_BIT, declared in <limits.h>. It's usually 8, but don't count on it. The range of values of a character variable, and whether it's signed or unsigned is implementation-dependent. If plain char is signed then its range of values is CHAR_MIN to CHAR_MAX inclusive. If it's unsigned then its range is 0 to UCHAR_MAX. These macros are defined in the header file <limits.h>.
2.2		Int An integer variable holds numbers with no fractional part. Printable character values stored in a variable of type char can be safely stored in an int: C guarantees that any printable character will have a positive value. Be careful when copying a char whose leftmost bit is 1 - it may be converted to a negative integer, or become positive. The memory requirements of an integer variable reflects the size suggested by the host architecture. The range of values of an integer variable is implementation-dependent, but at a minimium must be -32767 to +32767. The actual limits for your implementation will be specified by the macros INT_MIN and INT_MAX, in <limits.h>.
2.3		Float A floating-point variable holds numbers with a fractional part. The range of values is implementation-dependent and specified by the macros FLT_MIN and FLT_MAX defined in <float.h>.
3		ARRAYS A collection of elements of the same type is called an array. An array is commonly used in C to implement a string, that is, an array of characters.
4		DECLARATION OF VARIABLES A declaration specifies the properties associated with the variables listed. Declarations of variables, at present, should be made after the opening brace of the function main.
4.1		Basic Variable Types Basic variable types are created by declarations of the form: type-specifier identifier ; or type-specifier identifier, identifier, ...; where type-specifier is one of the keywords char, int, or float. Examples: int i, count, nflag; char ch; float radius, area; Variables can be initialised at the point of declaration using the assignment operator =, which copies the value or constant on the right side to the variable on the left side; for example: int i, count = 0, nflag; float f = 3.7; declares i, count and nflag as integer variables, with count initalised to zero and the others containing garbage. A floating-point variable f is declared, initialised to 3.7. Variables initialised at the point of declaration can be difficult to see, so the above sequence could be re-written: int i, count, nflag; /* declare the variables */ float f; count = 0; /* initialise the variables */ f = 3.7;
4.2		Arrays The format for the declaration of a one-dimensional array is: type-specifier identifier[constant-expression] ; or type-specifier identifier[constant-expression], identifier[constant-expression], ...; where type-specifier is one of the keywords char, int, or float and constant-expression evaluates to an integer. Examples: int arr[4]; /* creates array arr[0]...arr[3] */ float f1[9], f2[5], f3[7]; char s[5]; /* holds a string of four characters + '\0' */ To access an element use the form identifier[index] where index is any integer expression; for example: arr[0] = 15; f3[i] = f2[i*2]; f3[i+3] = f1[arr[1]]; s[0] = 'A'; where all variables used on the right side of the assignment operator have been initialised prior to the actual assignment.
4.2.1		Notes (a) In C, array subscripts always begin at zero, so an array declared as: arr[10] has subscripts in the range 0...9; more generally, an array declared with n elements, has its subscripts in the range 0...n-1. (b) There is no string data type in C, but a string is implemented as an array of char (see array s in the Examples in Section 4.2). The end of the C string is marked by a null character; therefore, when declaring an array to hold a string of length n, it should be declared with n+1 elements. (b) An array subscript that is out of range will give neither a compile error or a run-time error. It is up to the programmer to defend against an out of range access.

Back to C Home Page