vetr (version 0.2.20)

abstract: Turn S3 Objects Into Templates

Description

Create templates for use by alike. Currently somewhat experimental; behavior may change in future.

Usage

abstract(x, ...)
# S3 method for data.frame
abstract(x, ...)
# S3 method for default
abstract(x, ...)
# S3 method for array
abstract(x, ...)
# S3 method for matrix
abstract(x, ...)
# S3 method for list
abstract(x, ...)
# S3 method for lm
abstract(x, ...)
# S3 method for environment
abstract(x, ...)
# S3 method for ts
abstract(x, what = c("start", "end", "frequency"), ...)

Value

abstracted object

Arguments

x

the object to abstract

...

arguments for methods that require further arguments

what,

for time series which portion of the ts attribute to abstract, by default all three are abstracted, but you can select, any one, two, or all

Time Series

abstract replaces the "tsp" attribute with a "tsp_vetr" attribute with the components specified in what set to zero. alike will treat it as a "tsp" attribute except that the zero components become wildcards. If you manually create a template object with both "tsp" and "tsp_vetr" attributes, "tsp_vetr" is treated as a regular attribute. vetr does not consider whether the "ts" class is also set when attributing special semantics to the "tsp". The "tsp_vetr" attribute is required because R does not permit 0 values in "tsp". Prior to R4.6.0 and vetr 2.20, vetr constructed a "tsp" attribute with zeroes, but it no longer does that.

Details

abstract is intended to create templates for use by alike. The result of abstraction is often a partially specified object. This type of object may not be suited for use in typical R computations and may cause errors (or worse) if you try to use them as normal R objects.

There is no guarantee that the abstracted object is suitable for use as a template to alike as is. You may need to modify it further so that it suits your purposes.

abstract is an S3 generic. The default method will dispatch on implicit classes, so if you attempt to abstract an object without an explicit abstract method, it will get abstracted based on its implicit class. If you define your own abstract method and do not wish further abstraction based on implicit classes do not use NextMethod.

S4 and RC objects are returned unchanged.

Examples

iris.tpl <- abstract(iris)
alike(iris.tpl, iris[1:10, ])
alike(iris.tpl, transform(iris, Species=as.character(Species)))

abstract(1:10)
abstract(matrix(1:9, nrow=3))
abstract(list(1:9, runif(10)))