Use the as.* family of functions to make a derived copy of a variable that has been converted into a new type.
as.Text(x, ...)
as.Numeric(x)
as.Categorical(x, ...)
as.Datetime(x, format = "%Y-%m-%d %H:%M:%S", resolution, offset)
# S4 method for CrunchVariable
as.Numeric(x)
# S4 method for CrunchVariable
as.Text(x, format)
# S4 method for CrunchVariable
as.Categorical(x, format)
# S4 method for CrunchVariable
as.Datetime(x, format = "%Y-%m-%d %H:%M:%S", resolution, offset)
# S3 method for CrunchVariable
as.double(x, ...)
# S3 method for CrunchVariable
as.character(x, ...)
# S4 method for CrunchExpr
as.Numeric(x)
# S4 method for CrunchExpr
as.Text(x, format)
# S4 method for CrunchExpr
as.Categorical(x, format)
# S4 method for CrunchExpr
as.Datetime(x, format = "%Y-%m-%d %H:%M:%S", resolution, offset)
# S3 method for CrunchExpr
as.double(x, ...)
# S3 method for CrunchExpr
as.character(x, ...)a Crunch variable to derive and convert to a new type
additional arguments for as.character and as.numeric, ignored when used with
Crunch variables
for as.Datetime, when the variable in x is a text or
categorical variable, format is the typographical format that the datetime
is already formatted in that needs to be parse from (default:
"%Y-%m-%d %H:%M:%S"); for as.Text and as.Categorical, is
the typographical format that the datetime is to be formatted as (e.g.
"%Y-%m-%d %H:%M:%S" for "2018-01-08 12:39:57", the default if
no rollup resolution is specified on the source variable. If a rollup
resolution is specified, a reasonable default will be used.).
for as.Datetime, when the variable in x is a numeric
variable, the resolution of the number (e.g. "ms" for milliseconds, "s"
for seconds, etc. see expressions for more information about valid values.)
for as.Datetime, when the variable in x is a numeric the, a
character of the offset to count from in the shape "2018-01-08 12:39:57". If
not supplied, Crunch's default of 1970-01-01 00:00:00 will be used.
a CrunchExpr to be used as the derivation
Each type of Crunch variable (text, numeric, categorical, etc.) has an as.*
function (as.Text, as.Numeric, and as.Categorical respectively) that
takes the input given as x, and makes a new derived variable that is now of
the type specified. See below for detailed examples.
For as.Text and as.Numeric, aliases to the R-native functions
as.character and as.numeric are provided for convenience.
if (FALSE) {
# ds$v1 is of type Text
is.Text(ds$v1)
# [1] TRUE
# that has strings of numbers
as.vector(ds$v1)
# [1] "32" "8" "4096" "1024"
# convert this to a numeric variable with the alias `v1_numeric`
ds$v1_numeric <- as.Numeric(ds$v1)
# the values are the same, but are now numerics and the type is Numeric
as.vector(ds$v1_numeric)
# [1] 32 8 4096 1024
is.Numeric(ds$v1_numeric)
# [1] TRUE
# this new variable is derived, so if new data is appended or streamed, the
# new rows of data will be updated.
is.derived(ds$v1_numeric)
# [1] TRUE
}