Object-oriented programming (OOP) is a programming paradigm that allows programmers to define data structures (called objects) that have attributes and behaviors. In R, there are two systems for implementing OOP: the S4 system and the R6 system.
S4 System
The S4 system is a formal, class-based system that provides explicit control over inheritance and dispatch. It is more complex than the S3 system but more powerful, allowing for more control over the behavior of objects. The following is an example of defining an S4 class in R:
setClass("Person",
representation(
name = "character",
age = "numeric"
)
)
person <- new("Person", name = "John", age = 30)In this example, we define a new class called "Person" using the setClass() function. We define the class to have two attributes: name and age, which are of type character and numeric, respectively. We then create a new object of class Person called person using the new() function.
We can then define methods for our Person class using the setMethod() function. For example, we can define a method for printing out a summary of a Person object:
setMethod("summary", "Person",
function(object) {
cat("Name:", object@name, "n")
cat("Age:", object@age, "n")
}
)summary(person)
In this example, we define a method for summary() that applies to objects of class Person. The method takes a single argument, object, which is an instance of the Person class. The method then prints out a summary of the object’s name and age attributes using the cat() function.
R6 System
The R6 system is a simpler, reference-based system that is easier to learn and use. It provides a simpler syntax for defining classes and objects than the S4 system. The following is an example of defining an R6 class in R:
library(R6)
Person <- R6Class(
"Person",
public = list(
name = NULL,
age = NULL,
initialize = function(name, age) {
self$name <- name
self$age <- age
},
summary = function() {
cat("Name:", self$name, "n")
cat("Age:", self$age, "n")
}
)
)
person <- Person$new(name = "John", age = 30)In this example, we define a new class called Person using the R6Class() function from the R6 package. We define the class to have two attributes: name and age. We also define an initialize() method that takes name and age arguments and sets the name and age attributes of the object. We also define a summary() method that prints out a summary of the object’s name and age attributes using the cat() function.
We then create a new object of class Person called person using the $new() method.
In conclusion, both the S4 and R6 systems provide powerful tools for implementing OOP in R. The S4 system is more complex and provides more control over inheritance and dispatch, while the R6 system is simpler and easier to learn and use.