FloatingPointRoundingRule

enum FloatingPointRoundingRule

A rule for rounding a floating-point number.

Cases

case awayFromZero Required

Round to the closest allowed value whose magnitude is greater than or equal to that of the source.

The following example shows the results of rounding numbers using this rule:

(5.2).rounded(.awayFromZero)
// 6.0
(5.5).rounded(.awayFromZero)
// 6.0
(-5.2).rounded(.awayFromZero)
// -6.0
(-5.5).rounded(.awayFromZero)
// -6.0

Declaration

case awayFromZero
case down Required

Round to the closest allowed value that is less than or equal to the source.

The following example shows the results of rounding numbers using this rule:

(5.2).rounded(.down)
// 5.0
(5.5).rounded(.down)
// 5.0
(-5.2).rounded(.down)
// -6.0
(-5.5).rounded(.down)
// -6.0

This rule is equivalent to the C floor function and implements the roundToIntegralTowardNegative operation defined by the IEEE 754 specification.

Declaration

case down
case toNearestOrAwayFromZero Required

Round to the closest allowed value; if two values are equally close, the one with greater magnitude is chosen.

This rounding rule is also known as "schoolbook rounding." The following example shows the results of rounding numbers using this rule:

(5.2).rounded(.toNearestOrAwayFromZero)
// 5.0
(5.5).rounded(.toNearestOrAwayFromZero)
// 6.0
(-5.2).rounded(.toNearestOrAwayFromZero)
// -5.0
(-5.5).rounded(.toNearestOrAwayFromZero)
// -6.0

This rule is equivalent to the C round function and implements the roundToIntegralTiesToAway operation defined by the IEEE 754 specification.

Declaration

case toNearestOrAwayFromZero
case toNearestOrEven Required

Round to the closest allowed value; if two values are equally close, the even one is chosen.

This rounding rule is also known as "bankers rounding," and is the default IEEE 754 rounding mode for arithmetic. The following example shows the results of rounding numbers using this rule:

(5.2).rounded(.toNearestOrEven)
// 5.0
(5.5).rounded(.toNearestOrEven)
// 6.0
(4.5).rounded(.toNearestOrEven)
// 4.0

This rule implements the roundToIntegralTiesToEven operation defined by the IEEE 754 specification.

Declaration

case toNearestOrEven
case towardZero Required

Round to the closest allowed value whose magnitude is less than or equal to that of the source.

The following example shows the results of rounding numbers using this rule:

(5.2).rounded(.towardZero)
// 5.0
(5.5).rounded(.towardZero)
// 5.0
(-5.2).rounded(.towardZero)
// -5.0
(-5.5).rounded(.towardZero)
// -5.0

This rule is equivalent to the C trunc function and implements the roundToIntegralTowardZero operation defined by the IEEE 754 specification.

Declaration

case towardZero
case up Required

Round to the closest allowed value that is greater than or equal to the source.

The following example shows the results of rounding numbers using this rule:

(5.2).rounded(.up)
// 6.0
(5.5).rounded(.up)
// 6.0
(-5.2).rounded(.up)
// -5.0
(-5.5).rounded(.up)
// -5.0

This rule is equivalent to the C ceil function and implements the roundToIntegralTowardPositive operation defined by the IEEE 754 specification.

Declaration

case up