R: Creates an S3 method

setMethodS3 {R.methodsS3}

R Documentation

Creates an S3 method

Description

Creates an S3 method. A function with name <name>.<class> will be set to definition. The method will get the modifiers specified by modifiers. If there exists no generic function for this method, it will be created automatically.

Usage

## Default S3 method:
setMethodS3(name, class="default", definition, private=FALSE, protected=FALSE, static=FALSE, abstract=FALSE, trial=FALSE, deprecated=FALSE, envir=parent.frame(), overwrite=TRUE, conflict=c("warning", "error", "quiet"), createGeneric=TRUE, appendVarArgs=TRUE, validators=getOption("R.methodsS3:validators:setMethodS3"), ...)

Arguments

`name`	The name of the method.
`class`	The class for which the method should be defined. If `class == "default"` a function with name `<name>.default` will be created.
`definition`	The method defintion.
`private, protected`	If `private=TRUE`, the method is declared private. If `protected=TRUE`, the method is declared protected. In all other cases the method is declared public.
`static`	If `TRUE` this method is defined to be static, otherwise not. Currently this has no effect expect as an indicator.
`abstract`	If `TRUE` this method is defined to be abstract, otherwise not. Currently this has no effect expect as an indicator.
`trial`	If `TRUE` this method is defined to be a trial method, otherwise not. A trial method is a method that is introduced to be tried out and it might be modified, replaced or even removed in a future release. Some people prefer to call trial versions, beta version. Currently this has no effect expect as an indicator.
`deprecated`	If `TRUE` this method is defined to be deprecated, otherwise not. Currently this has no effect expect as an indicator.
`envir`	The environment for where this method should be stored.
`overwrite`	If `TRUE` an already existing method with the same name (and of the same class) will be overwritten, otherwise not.
`conflict`	If a method already exists with the same name (and of the same class), different actions can be taken. If `"error"`, an exception will be thrown and the method will not be created. If `"warning"`, a `warning` will be given and the method will be created, otherwise the conflict will be passed unnotice.
`createGeneric`	If `TRUE`, a generic S3/UseMethod function is defined for this method.
`appendVarArgs`	If `TRUE`, argument `...` is added with a warning, if missing. For special methods such as `$` and `[[`, this is never done. This will increase the chances that the method is consistent with a generic function with many arguments and/or argument `...`.
`validators`	An optional `list` of `function`s that can be used to assert that the generated method meets certain criteria.
`...`	Not used.

Author(s)

Henrik Bengtsson (http://www.braju.com/R/)

Examples

 ######################################################################
# Example 1
######################################################################
setMethodS3("foo", "default", function(x, ...) {
  cat("In default foo():\n");
  print(x, ...);
})

setMethodS3("foo", "character", function(s, ...) {
  cat("In foo() for class 'character':\n");
  print(s, ...);
})

# The generic function is automatically created!
print(foo)

foo(123)
foo("123")

######################################################################
# Example 2
#
# Assume that in a loaded package there is already a function bar(),
# but you also want to use the name 'bar' for the character string.
# It may even be the case that you do not know of the other package,
# but your users do!
######################################################################
# bar() in other package
bar <- function(x, y, ...) {
  cat("In bar() of 'other' package.\n");
}

# Your defintion; will redefine bar() above to bar.default().
setMethodS3("bar", "character", function(object, ...) {
  cat("In bar() for class 'character':\n");
  print(object, ...);
})

bar(123)
bar("123")