Decimals is a small library of functions for rounding and formatting base ten numbers in Go. These are functions that are either missing from the standard libraries or are more convenient for presenting numbers in a human-readable format.

In the original version of this package, integer rounding was performed with mathematical operations. This worked in most cases but produced incorrect results when rounding at or near the minimum and maximum int64 values. The latest version of the library provides the same public interface, but internally uses string representations of decimal numbers in order to produce consistent rounding behaviour across the full range of integer values. If you attempt to round to a number beyond the minimum or maximum int64 values the minimum or maximum is returned instead.

Install with go get.

go get github.com/olihawkins/decimals

Use go test to run the tests.

See the GoDoc for the full documentation.

Round int64 and float64 numbers.

decimals.RoundInt(x int64, precision int) int64
decimals.RoundFloat(x float64, precision int) float64

Round an integer to the nearest power of ten using a negative value for precision.

i := decimals.RoundInt(555, 0)  // i = 555
i := decimals.RoundInt(555, -1) // i = 560
i := decimals.RoundInt(555, -2) // i = 600
i := decimals.RoundInt(555, -3) // i = 1000 

Round a float to the given number of decimal places using a positive value for precision, or to the nearest power of ten using a negative value for precision.

f := decimals.RoundFloat(5.5555, 3)  // f = 5.556
f := decimals.RoundFloat(5.5555, 2)  // f = 5.56
f := decimals.RoundFloat(5.5555, 1)  // f = 5.6
f := decimals.RoundFloat(5.5555, 0)  // f = 6
f := decimals.RoundFloat(5.5555, -1) // f = 10

Convert integers and floats to formatted strings with the given decimal precision, using a comma separator for thousands.

decimals.FormatInt(x int64, precision int) string
decimals.FormatFloat(x float64, precision int) string

Format an integer rounded to the nearest power of ten using a negative value for precision.

i := decimals.FormatInt(5555555, 0)  // i = "5,555,555"
i := decimals.FormatInt(5555555, -1) // i = "5,555,560"
i := decimals.FormatInt(5555555, -2) // i = "5,555,600"
i := decimals.FormatInt(5555555, -3) // i = "5,556,000" 

Format a float rounded to the given number of decimal places using a positive value for precision, or to the nearest power of ten using a negative value for precision.

f := decimals.FormatFloat(5555.555, 2)  // f = "5,555.56"
f := decimals.FormatFloat(5555.555, 1)  // f = "5,555.6"
f := decimals.FormatFloat(5555.555, 0)  // f = "5,556"
f := decimals.FormatFloat(5555.555, -1) // f = "5,560"
f := decimals.FormatFloat(5555.555, -2) // f = "5,600"