Data.BigDecimal

This module defines the type BigDecimal which provides a representation of arbitrary precision decimal numbers. BigDecimal is a native Haskell implementation based on arbitrary sized Integer values. The implementation was inspired by Java BigDecimals.

BigDecimal instantiates the typeclasses Num, Fractional and Real. It is thus possible to use all common operators like +, -, *, /, ^ on them.

Here are a few examples from an interactive GHCI session:

 λ> a = BigDecimal 144 2
 λ> toString a
 "1.44"
 λ> b = sqrt a
 λ> toString b
 "1.2"
 λ> b * b
 BigDecimal 144 2
 λ> b * b * b
 BigDecimal 1728 3
 λ> b^2
 BigDecimal 144 2
 λ> c = fromString "123.4567890"
 λ> c
 BigDecimal 1234567890 7
 λ> a / c
 BigDecimal 1166400010614240096589584878965222398584 41
 λ> roundBD it (halfUp 10)
 BigDecimal 116640001 10
 λ> divide (a, c) $ halfUp 20
 BigDecimal 1166400010614240097 20

Documentation

data BigDecimal

BigDecimal is represented by an unscaled Integer value plus a second Integer value that defines the scale E.g.: (BigDecimal 1234 2) represents the decimal value 12.34.

Constructors

BigDecimal Integer Integer

creates a BigDecimal value from an unscaled Integer value and a scale, given as a positive Integer. Example: (BigDecimal 1234 2) creates the value 12.34

Instances

Eq BigDecimal

Methods

(==) :: BigDecimal -> BigDecimal -> Bool

(/=) :: BigDecimal -> BigDecimal -> Bool

Fractional BigDecimal

Methods

(/) :: BigDecimal -> BigDecimal -> BigDecimal

recip :: BigDecimal -> BigDecimal

fromRational :: Rational -> BigDecimal

Num BigDecimal

Methods

(+) :: BigDecimal -> BigDecimal -> BigDecimal

(-) :: BigDecimal -> BigDecimal -> BigDecimal

(*) :: BigDecimal -> BigDecimal -> BigDecimal

negate :: BigDecimal -> BigDecimal

abs :: BigDecimal -> BigDecimal

signum :: BigDecimal -> BigDecimal

fromInteger :: Integer -> BigDecimal

Ord BigDecimal

Methods

compare :: BigDecimal -> BigDecimal -> Ordering

(<) :: BigDecimal -> BigDecimal -> Bool

(<=) :: BigDecimal -> BigDecimal -> Bool

(>) :: BigDecimal -> BigDecimal -> Bool

(>=) :: BigDecimal -> BigDecimal -> Bool

max :: BigDecimal -> BigDecimal -> BigDecimal

min :: BigDecimal -> BigDecimal -> BigDecimal

Read BigDecimal

Methods

readsPrec :: Int -> ReadS BigDecimal

readList :: ReadS [BigDecimal]

readPrec :: ReadPrec BigDecimal

readListPrec :: ReadPrec [BigDecimal]

Real BigDecimal

Methods

toRational :: BigDecimal -> Rational

Show BigDecimal

Methods

showsPrec :: Int -> BigDecimal -> ShowS

show :: BigDecimal -> String

showList :: [BigDecimal] -> ShowS

data RoundingMode

RoundingMode defines how to handle loss of precision in divisions or explicit rounding.

Constructors

UP

Rounding mode to round away from zero.

DOWN

Rounding mode to round towards zero.

CEILING

Rounding mode to round towards positive infinity.

FLOOR

Rounding mode to round towards negative infinity.

HALF_UP

Rounding mode to round towards "nearest neighbor" unless both neighbors are equidistant, in which case round up.

HALF_DOWN

Rounding mode to round towards "nearest neighbor" unless both neighbors are equidistant, in which case round down.

HALF_EVEN

Rounding mode to round towards "nearest neighbor" unless both neighbors are equidistant, in which case, round towards the even neighbor.

PRECISE

Rounding mode to assert that the requested operation has an exact result, hence no rounding is applied.

type MathContext = (RoundingMode, Maybe Integer)

A MathContext is interpreted by divisions and rounding operations to specify the expected loss of precision and the rounding behaviour. MathContext is a pair of a RoundingMode and a target precision of type Maybe Integer. The precision defines the number of digits after the decimal point. If Nothing is given as precision all decimal digits are to be preserved, that is precision is not limited.

getScale :: BigDecimal -> Integer

gets the scale part of a BigDecimal

getValue :: BigDecimal -> Integer

get the unscaled value of a BigDecimal

precision :: BigDecimal -> Integer

returns the number of digits of an Integer

trim :: Integer -> BigDecimal -> BigDecimal

removes trailing zeros from a BigDecimals intValue by decreasing the scale

nf :: BigDecimal -> BigDecimal

computes the normal form of a BigDecimal

divide

Arguments

:: (BigDecimal, BigDecimal)

the tuple of dividend and divisor. I.e. (dividend, divisor)

-> MathContext

MathContext (i.e. a tuple of RoundingMode and the specified precision) defines the rounding behaviour. if Nothing if given as precision the maximum possible precision is used.

-> BigDecimal

the resulting BigDecimal

divide two BigDecimals and applies the MathContext (i.e. a tuple of RoundingMode and the specified precision) for rounding.

roundBD :: BigDecimal -> MathContext -> BigDecimal

round a BigDecimal to n digits applying the MathContext mc

fromRatio :: Rational -> MathContext -> BigDecimal

creates a BigDecimal from a Rational value. MathContext defines precision and rounding mode.

halfUp :: Integer -> MathContext

construct a MathContext for rounding HALF_UP with scale decimal digits

fromString :: String -> BigDecimal

read a BigDecimal from a human readable decimal notation. e.g. fromString "3.14" yields 'BigDecimal 314 2'

matchScales :: (BigDecimal, BigDecimal) -> (BigDecimal, BigDecimal)

match the scales of a tuple of BigDecimals

toString :: BigDecimal -> String

returns a readable String representation of a BigDecimal e.g. toString (BigDecimal 314 2) yields "3.14"