Data Types

data RoundingMode

Rounding modes.

RoundingUp

Round away from zero.

RoundingDown

Round towards zero.

RoundingCeiling

Round towards positive infinity.

RoundingFloor

Round towards negative infinity.

RoundingHalfUp

Round towards the nearest neighbor unless both neighbors are equidistant, in which case round away from zero.

RoundingHalfDown

Round towards the nearest neighbor unless both neighbors are equidistant, in which case round towards zero.

RoundingHalfEven

Round towards the nearest neighbor unless both neighbors are equidistant, in which case round towards the even neighbor.

RoundingUnnecessary

Do not round. Raises an error if the result cannot be represented without rounding at the targeted scale.

Functions

: NumericScale n3 => Numeric n1 -> Numeric n2 -> Numeric n3

Multiply two numerics. Both inputs and the output may have different scales, unlike (*) which forces all numeric scales to be the same. Raises an error on overflow, rounds to chosen scale otherwise.

: NumericScale n3 => Numeric n1 -> Numeric n2 -> Numeric n3

Divide two numerics. Both inputs and the output may have different scales, unlike (/) which forces all numeric scales to be the same. Raises an error on overflow, rounds to chosen scale otherwise.

: NumericScale n2 => Numeric n1 -> Numeric n2

Cast a Numeric. Raises an error on overflow or loss of precision.

: NumericScale n2 => Numeric n1 -> Numeric n2

Cast a Numeric. Raises an error on overflow, rounds to chosen scale otherwise.

: NumericScale n2 => Numeric n1 -> Numeric n2

Move the decimal point left or right by multiplying the numeric value by 10^(n1 - n2). Does not overflow or underflow.

: NumericScale n => Numeric n

The number pi.

: NumericScale n => Numeric n

The minimum strictly positive value that can be represented by a numeric of scale n.

: NumericScale n => Int -> RoundingMode -> Numeric n -> Numeric n

Round a Numeric number. The value of round n r a is the value of a rounded to n decimal places (i.e. scale), according to the rounding mode r.

This will fail when using the RoundingUnnecessary mode for a number that cannot be represented exactly with at most n decimal places.