Legacy Document

rinttol

You can use the rinttol function to round a real number to the nearest integer in the current rounding direction.

long int rinttol (double_t x);

x

Any floating-point number.

DESCRIPTION

The rinttol function rounds its argument to the nearest integer in the current rounding direction and places the result in a long int type. The available rounding directions are upward, downward, to nearest, and toward zero.

The rinttol function provides the floating-point to integer conversion as described in the IEEE standard. It differs from rint (described on page 6-13) in that it returns the value in an integer type; rint returns the value in a floating-point type.

EXCEPTIONS

When x is finite and nonzero, either the result of $rinttol(x)$ is exact or it raises one of the following exceptions:

• inexact (if x is not an integer)
• invalid (if the integer result is outside the range of the long int type)

SPECIAL CASES

Table 9-1 shows the results when the argument to the rinttol function is a zero, a NaN, or an Infinity.

Special cases for the rinttol function

Operation	Result	Exceptions raised
$rinttol(+0)$	+0	None
$rinttol(-0)$	$-0$	None
$rinttol(NaN)$	Undefined	None[12]
$rinttol(+$ )	Undefined	Invalid
$rinttol(-$ )	Undefined	Invalid

EXAMPLES

z = rinttol(+INFINITY);/* z = unspecified value for all rounding 
                        directions because +INFINITY exceeds the 
                        range of long int. The invalid exception 
                        is raised. */
z = rinttol(300.1);    /* z = 301 if rounding direction is upward
                        else z = 300. The inexact exception is 
                        raised.*/
z = rinttol(-300.1);   /* z = -301 if rounding direction is 
                        downward else z = -300. The inexact 
                        exception is raised. */

---

[12] If the NaN is a signaling NaN, the invalid exception is raised.