Functions to operate on nanosecond time resolution using integer64 bit representation. Conversion functions for several standard R types are provided, and more will be added as needed.
nanotime(from, ...)as.nanotime(from, ...)
# S4 method for character
nanotime(from, format = "", tz = "")
# S4 method for character
as.nanotime(from, format = "", tz = "")
nanotime.matrix(x)
# S4 method for POSIXct
nanotime(from, accurate = TRUE)
# S4 method for POSIXct
as.nanotime(from, accurate = TRUE)
# S4 method for POSIXlt
nanotime(from)
# S4 method for POSIXlt
as.nanotime(from)
# S4 method for Date
nanotime(from)
# S4 method for Date
as.nanotime(from)
# S4 method for nanotime
print(x, format = "", tz = "", quote = FALSE, ...)
# S4 method for nanotime
show(object)
# S3 method for nanotime
format(x, format = "", tz = "", ...)
# S3 method for nanotime
index2char(x, ...)
# S3 method for nanotime
as.POSIXct(x, tz = "", ...)
# S3 method for nanotime
as.POSIXlt(x, tz = "", ...)
# S3 method for nanotime
as.Date(x, ...)
# S3 method for nanotime
as.data.frame(x, ...)
# S3 method for nanotime
as.integer64(x, ...)
# S4 method for nanotime,character
-(e1, e2)
# S4 method for nanotime,nanotime
-(e1, e2)
# S4 method for nanotime,integer64
-(e1, e2)
# S4 method for nanotime,numeric
-(e1, e2)
# S4 method for ANY,nanotime
-(e1, e2)
# S4 method for nanotime,ANY
-(e1, e2)
# S4 method for nanotime,ANY
+(e1, e2)
# S4 method for nanotime,integer64
+(e1, e2)
# S4 method for nanotime,numeric
+(e1, e2)
# S4 method for ANY,nanotime
+(e1, e2)
# S4 method for integer64,nanotime
+(e1, e2)
# S4 method for numeric,nanotime
+(e1, e2)
# S4 method for nanotime,nanotime
+(e1, e2)
# S4 method for nanotime,nanotime
Arith(e1, e2)
# S4 method for nanotime,ANY
Arith(e1, e2)
# S4 method for ANY,nanotime
Arith(e1, e2)
# S4 method for nanotime,character
Compare(e1, e2)
# S4 method for character,nanotime
Compare(e1, e2)
# S4 method for nanotime,POSIXt
Compare(e1, e2)
# S4 method for POSIXt,nanotime
Compare(e1, e2)
# S4 method for nanotime,ANY
Compare(e1, e2)
# S4 method for nanotime,ANY
Logic(e1, e2)
# S4 method for ANY,nanotime
Logic(e1, e2)
# S4 method for nanotime
Math(x)
# S4 method for nanotime
Math2(x, digits)
# S4 method for nanotime
Summary(x, ..., na.rm = FALSE)
# S4 method for nanotime
min(x, ..., na.rm = FALSE)
# S4 method for nanotime
max(x, ..., na.rm = FALSE)
# S4 method for nanotime
range(x, ..., na.rm = FALSE)
# S4 method for nanotime
Complex(z)
# S4 method for nanotime
[[(x, i, j, ..., drop = FALSE)
# S4 method for nanotime,numeric
[(x, i, j, ..., drop = FALSE)
# S4 method for nanotime,logical
[(x, i, j, ..., drop = FALSE)
# S4 method for nanotime,character
[(x, i, j, ..., drop = FALSE)
# S4 method for nanotime,ANY
[(x, i, j, ..., drop = FALSE)
# S4 method for nanotime,ANY,ANY,ANY
[(x, i, j, ...) <- value
# S3 method for nanotime
c(...)
# S4 method for nanotime
names(x) <- value
# S4 method for nanotime
is.na(x)
NA_nanotime_
# S3 method for nanotime
as.character(x, ...)
# S3 method for nanoduration
as.data.frame(x, ...)
A nanotime object
An object of class nanotime of length 1.
further arguments passed to or from methods.
A character string. Can also be set via
options("nanotimeFormat") and uses
‘%Y-%m-%dT%H:%M:%E9S%Ez’ as a default and
fallback
character specifying a timezone which is required for
as.POSIXct, as.POSIXlt and can be specified for
as.nanotime, format and print; it can
also be set via options("nanotimeTz") and uses
‘UTC’ as a default and fallback
nanotime objects
in the conversion from POSIXct to
nanotime, indicates if one wants to preserve the
maximum precision possible; the default is TRUE, but in
most situations the loss of precision is negligible, and
setting this parameter to TRUE will make the conversion
nearly an order of magnitude faster
indicates if the output of print should be
quoted
argument for method show
Operand of class nanotime
Operand of class nanotime
Required for Math2 signature but ignored here
a logical indicating whether missing values should be removed.
Required for Complex signature but ignored here
index specifying elements to extract or replace.
Required for [ signature but ignored here
Required for [ signature but ignored here
argument for nanotime-class
Working with dates and times is difficult. One needs a
representation of both time points and time
duration. In R, think of Date or POSIXct objects
for the former, and difftime for the later. Here we have
time points nanotime, an interval type nanoival and
two flavors of duration which are a simple count of nanoseconds
nanoduration and a calendar duration that is able to track
concepts such as months and days nanoperiod. Point in time
and intervals are all based on durations relative to the epoch of
January 1, 1970.
Formatting and character conversion for nanotime objects is
done by functions from the RcppCCTZ package relying
on code from its embedded CCTZ library. The default format
is ISO3339 compliant: %Y-%m-%dT%H:%M:%E9S%Ez. It
specifies a standard ISO 8601 part for date and time --- as well
as nine digits of precision for fractional seconds (down to
nanoseconds) and on offset (typically zero as we default to UTC).
It can be overriden by using options() with the key of
nanotimeFormat and a suitable value. Similarly,
nanotimeTz can be used to select a different timezone.
For input, some slack it cut, and various shortened formats are
accepted by default such as 2020-03-10 or 2020-03-10
18:16:00, or 2020-03-10 18:16:00.001 (and the ‘T’
separator is optional.
The tz parameter is allowed only when constructing a
nanotime from a character. This is because any
numeric, Date and POSIXct is de facto
considered an offset since the epoch. On the contrary, a
character is considered interpretable and hence if it does
not contain a timezone in its representation, it is possible to
specify the tz argument to specify in which timezone it
should be interpreted. This is useful in particular if one wants
to convert a Date to be aligned to the beginning of the day
in a specific timezone; in this case one should convert the
Date to a character before calling the
nanotime constructor with the desired timezone.
Dirk Eddelbuettel
Leonardo Silvestri
Notice that the conversion from POSIXct explicitly sets the last three digits to zero. Nanosecond time stored in a 64-bit integer has nineteen digits precision where doubles (which are used internally for POSIXct as well) only have sixteen digits. So rather than showing three more (essentially random) digits it is constructed such that these three additional digits are zeros.
nanoival, nanoduration,
nanoperiod, seq.nanotime as well as
the documentation in package RcppCCTZ.
if (FALSE) {
x <- nanotime(1)
print(x)
as.nanotime("1970-01-01T00:00:00.000000001+00:00")
as.nanotime("2020-03-10 Europe/Berlin")
as.nanotime("2020-03-10 18:31:23.001", tz="America/New_York")
as.nanotime("2020-03-10T040947190301440", format="%Y-%m-%dT%H%M%S%E*f")
x <- x + 1
print(x)
format(x)
x <- x + 10
print(x)
format(x)
nanotime(Sys.time()) + 1:3 # three elements each 1 ns apart
seq(x, by=as.nanoperiod("1d"), length.out=5, tz="Asia/Tokyo")
}
Run the code above in your browser using DataLab