Skip to main content Skip to docs navigation

Time Spans

TimeSpan, MonthSpan, DateTimeSpan, DateTimeRange...

Document not reviewed yet, might be outdated. Please, let us know if you find something invalid here.
On this page

Klock has utilities for representing spans of time, dates, and months.


Klock offers a TimeSpan inline class using a Double to be allocation-free on all targets, and it serves to represent durations without start references. It has millisecond precision up to 2 ** 52, which means that it can represent up to 142808 years with millisecond precision. It has a special TimeSpan.NULL value (internally represented as NaN) to represent an absence of time without having to use nullable types that are not allocation-free.

Constructing instances

There are extension properties for Number to generate TimeSpan instances. The extensions use Number, but are inline, so no allocation is done.

val time = 1_000_000_000.nanoseconds
val time = 1_000_000.microseconds
val time = 1_000.milliseconds
val time = 1.seconds
val time = 0.5.seconds
val time = 60.minutes
val time = 24.hours
val time = 1.days
val time = 1.weeks

You can represent from nanoseconds to weeks. Months and years are not included here but included as part of MonthSpan since months and years work different because leap years.


val time = 4.seconds
val doubleTheTime = time * 2
val negatingTime = -time
val twoHundredMillisecondsMore = time + 200.milliseconds

External Arithmetic

Adding or subtracting time to a date

val now =
val inTenSeconds = now + 10.seconds


TimeSpan implements Comparable<TimeSpan> so you can compare times independently to the unit used to instantiate them:

val isTrue = 4001.milliseconds > 4.seconds

Converting between units

TimeSpan has several properties to get the instance time interpreted in different units of measure:

val value: Double = 1.seconds.nanoseconds // 1.seconds as nanoseconds (1_000_000_000)
val value: Double = 1.seconds.microseconds // 1.seconds as microseconds (1_000_000)
val value: Double = 1.seconds.milliseconds // 1.seconds as milliseconds (1000)
val value: Double = 1.seconds.seconds // 1.seconds as seconds (1)
val value: Double = 1.seconds.minutes // 1.seconds as minutes (1.0/60)
val value: Double = 1.seconds.hours // 1.seconds as hours (1.0/3_600)
val value: Double = 1.seconds.days // 1.seconds as days (1.0/86_4000)

For milliseconds there are a couple of additional properties to get it as Long and Int:

val value: Long = 1.seconds.millisecondsLong // 1.seconds as milliseconds (1000L)
val value: Int  = 1.seconds.millisecondsInt  // 1.seconds as milliseconds (1000)


MonthSpan allows to represent month and year durations (with month precission) where TimeSpan simply can’t work because month distance depends on specific moments to have into account leap years.

Constructing instances

val time = 1.months
val time = 5.years


Adding or subtracting month-based spans

val time: MonthSpan = 5.years + 2.months
val time: MonthSpan = 5.years * 2
val time: DateTimeSpan = 5.years + 5.days

External Arithmetic

Adding or subtracting months to a date

val now =
val inTwoMonths = now + 2.months


val time = 5.years + 2.months + 4.months

val years : Int = time.years  // 5
val months: Int = time.months // 6

val totalYears : Double = time.totalYears  // 5.0
val totalMonths: Int    = time.totalMonths // 5 * 12 + 6 = 66


DateTimeSpan is a combination of MonthSpan and TimeSpan.

This class is not inline, so whenever it is possible use MonthSpan or TimeSpan to alter DateTime directly.


DateTimeRange is a range between two DateTime.

Constructing Instances

val today =
val tomorrow = + 1.days

val rangeOpen = today until tomorrow
val rangeClosed = today .. tomorrow


val inTenMinutes = now + 10.minutes
val contains: Boolean = inTenMinutes in rangeOpen

Span and Duration

val duration: TimeSpan     = rangeOpen.duration
val span    : DateTimeSpan = rangeOpen.span

Days Between two DateTime

val inFourMonths = today + 4.month
val days = (today until inFourMonths).span.days


The DateTimeRangeSet represents a non-overlapping sets of DateTimeRange. It can be used to measure availability ranges among other things.

Constructing Instances

val set: DateTimeRangeSet = DateTimeRangeSet(dateTimeRange1, dateTimeRange2, dateTimeRange3)

Combining two instances (addition/union)

It will combine all the ranges and generate a non-overlapping instance

val set = dateTimeRangeSet1 + dateTimeRangeSet2

Substracting two time ranges

val set = dateTimeRangeSet1 - dateTimeRangeSet2

Intersection between two ranges

val set = dateTimeRangeSet1.intersection(dateTimeRangeSet2)
Was this article useful?