Library: Floating point number support (OMFLOAT) Import: omfloat.xmd Library: Binary coded decimal (BCD) numbers (OMBCD) Import: ombcd.xmd

Return type: BCD or float Returns: A BCD or floating point number rounded to the nearest value. If a precision is specified, the value is rounded to the nearest multiple of the precision specifier. If bankers-nearest is specified, the value is rounded to the nearest even multiple of the precision specifier.

Declaration

export overloaded external float function
   round         value float x
         nearest value float base optional

export overloaded bcd function
   round                 value bcd b
         nearest         value bcd base         optional
         bankers-nearest value bcd bankers-base optional

Use this function to round any of the following to the nearest digit or decimal place specified:

• a BCD number
• a floating point number
• an integer coerced into a BCD or floating point number

If you do not specify the precision, you will round to the nearest integer.

Import either ombcd.xmd or omfloat.xmd in your program after deciding whether you want BCD mathematics (excellent for financial calculations) or floating point mathematics (excellent for extremely large numbers).

BCD Example:

Calculate the area of a circle:

  import "ombcd.xmd" unprefixed
  process
      local bcd pi  initial {"3.14179"}
      local bcd radius  initial {"74"}
      local bcd area
  
      set area to round (pi * radius * radius)
      output "area of circle with radius " || "d" % radius || " = " || "d" % area || "%n"

Floating Point Example:

Calculate the area of a circle:

  import "omfloat.xmd" unprefixed
  process
      local float radius initial {"74"}
      local float area
  
      set area to round (pi * radius * radius)
      output "Area of circle with radius " || "d" % radius || " = " || "d" % area || "%n"

You can specify the precision of the rounding. The following program uses a precision of 0.01 to round to the nearest cent:

  
  import "ombcd.xmd" unprefixed
  process
      local bcd exchange-rate initial {1.4356}
      local bcd us-price  initial {129.95}
      local bcd canadian-price
  
      set canadian-price to round (us-price * exchange-rate) nearest bcd 0.01
      output "US " || "<$,NNZ.ZZ>" % us-price || " = " || "<$,NNZ.ZZ>" % canadian-price || " CDN %n"

You can use bankers rounding (available only with the BCD library) to resolve ties to the nearest even multiple of the precision specifier:

  ; Bankers-nearest rounds to the nearest even digit
  import "ombcd.xmd" unprefixed
  process
     local bcd x initial {9.95}
     local bcd y initial {9.85}
     local bcd rounded-x
     local bcd rounded-y
     set rounded-x to round (x) bankers-nearest 0.1
     set rounded-y to round (y) bankers-nearest 0.1
     output "Rounded-x = "
         || "d" % rounded-x
         || "%n"
         || "Rounded-y = "
         || "d" % rounded-y
         || "%n"
  ; Output: "Rounded-x = 10
  ;          Rounded-y = 9.8"

The precision specifier must be a power of 10, or an error occurs.

Related Topics BCD data type Floating point data type ceiling floor truncate

Other Library Functions ** (power) abs ceiling exp floor ln log10 round sqrt truncate

[ INDEX ] [ CONCEPTS ] [ TASKS ] [ SYNTAX ] [ LIBRARIES ] [ LEGACY LIBRARIES ] [ ERRORS ]