Time­Cowan

Time

(current-posix-millisecond)

Returns a rational number representing the current millisecond of the Posix epoch, which began on 00:00:00 on 1 January 1970, Coordinated Universal Time but excludes all leap seconds. Implementations SHOULD return a value that is:

• As accurate as reasonably achievable.
• Provides at least 1-second precision.
• Provides a range of at least 2³¹ seconds before and after the epoch.

. Implementations MAY return exact or inexact values, and SHOULD return inexact values unless:

• The value is an integer.
• The range of most-efficiently-implemented exact integers in the implementation exceeds the integer range of inexact rationals.

In practice this means that implementations with fixnums greater than 2⁵³ should return fixnums unless they have time-of-day clocks that provide more than 1 ms precision, and all other implementations should return IEEE 64-bit floats.

If the implementation cannot provide the time when current-posix-milliseconds is called (either temporarily or permanently), it MUST return #f.

Note that there is about a 1 in 10^-8 probability that a computation of elapsed time made by calling this procedure twice will be off by 1 second, given the number of leap seconds between 1972 and 2011.

Date and Time Arithmetic (WG2)

See TimeAdvancedCowan.

Time Durations and Periods (WG2)

See TimePeriodsCowan.

Time formatting (WG2)

See TimeFormattingCowan.