What Does x?y:z Mean?

What Does x?y:z Mean?

In C, ?: is called the conditional operator, which is the only operator that takes three operands. The general form of the conditional operator is

x ? y : z

Here x, y, and z are three operands. Among them, x contains the test condition, and y and z represent the final value of the expression. If x returns nonzero (that is, TRUE), y is chosen; otherwise, z is the result.

For instance, the expression x > 0 ? `T' : `F' returns `T' if the value of x is greater than 0. Otherwise, the expression returns `F'.

Listing 8.7 demonstrates the usage of the conditional operator in the C language.

TYPE
Listing 8.7. Using the conditional operator.

1:  /* 08L07.c: Using the ?: operator */
2:  #include <stdio.h>
3:
4:  main()
5:  {
6:     int   x;
7:
8:     x = sizeof(int);
9:     printf("%s\n",
10:       (x == 2) ? "The int data type has 2 bytes." : "int doesn't have 2
                      Âbytes.");
11:    printf("The maximum value of int is: %d\n",
12:       (x != 2) ? ~(1 << x * 8 - 1) : ~(1 << 15) );
13:    return 0;
14: }

OUTPUT
I get the following output shown on the screen when I run the executable 08L07.exe from a DOS prompt on my machine:

C:\app> 08L07
The int data type has 2 bytes.
The maximum value of int is: 32767
C:\app>

ANALYSIS
In Listing 8.7, the size of the int data type is measured first in line 8, by using the sizeof operator and the number of bytes assigned to the integer variable x.

Lines 9 and 10 contain one statement, in which the conditional operator (?:) is used to test whether the number of bytes saved in x is equal to 2. (Here you see another example that a single statement can span multiple lines.) If the x == 2 expression returns nonzero (that is, TRUE), the string of The int data type has 2 bytes. is printed out by the printf() function in the statement. Otherwise, the second string, int doesn't have 2 bytes., is displayed on the screen.

In addition, the statement in lines 11 and 12 tries to find out the maximum value of the int data type on the current machine. The x != 2 expression is evaluated first in the statement. If the expression returns nonzero (that is, the byte number of the int data type is not equal to 2), the ~( << x * 8 - 1) expression is evaluated, and the result is chosen as the return
value. Here the ~(1 << x * 8 - 1) expression is a general form to calculate the maximum value of the int data type, which is equivalent to 2 ** (x * 8 - 1) - 1. (The complement operator, ~, and the shift operator, <<, were introduced in the previous sections of this hour.)

On the other hand, if the test condition x != 2 in line 12 returns 0, which means the value of x is indeed equal to 2, the result of the ~(1 << 15) expression is chosen. Here you may have already figured out that ~(1 << 15) is equivalent to 2¹⁵_1, which is the maximum value that the 16-bit int data type can have.

The result displayed on the screen shows that the int data type on my machine is 2 bytes
(or 16 bits) long, and the maximum value of the int data type is 32767.