Classes in R

In R, the concept of "classes" is not as rigorous as in traditional object-oriented languages like Java or Python. Instead, R uses a more flexible system based on generic functions and method dispatch, known as the S3 system. The S4 system is another system that's more formal than S3 but less commonly used.

In this tutorial, we will go over the basics of the S3 system, which is the more prevalent method of creating and using classes in R.

1. Creating an S3 Class

In the S3 system, an object's class is simply a string that gets attached to it. Here's how you can create a new S3 class:

# Creating an object of class 'person'
person <- list(name = "John", age = 30)
class(person) <- "person"

2. Creating S3 Methods

Methods in the S3 system are just regular functions with a special naming convention: [generic].[class].

For example, let's create a print method for our person class:

print.person <- function(x) {
  cat(paste("Name:", x$name, "\n"))
  cat(paste("Age:", x$age, "\n"))
}

# Now, using the print function on a 'person' object will use our custom method
print(person)

3. Creating Generic Functions

A generic function is a function that calls the class-specific method based on the class of its argument.

# The `print` function is already a generic, but let's see how we'd define one ourselves:
greet <- function(x) {
  UseMethod("greet")
}

greet.person <- function(x) {
  paste("Hello, my name is", x$name, "and I'm", x$age, "years old.")
}

greet(person)

4. Inheritance

S3 classes can have inheritance. If a method for a specific class is not found, R looks for methods of the parent class.

class(person) <- c("employee", "person")
greet.employee <- function(x) {
  paste("Hello, I work here and my name is", x$name)
}

greet(person)  # This will use the `greet.employee` method because "employee" is the first class in the class vector

5. Checking and setting classes

Use class(object) to get the class of an object and is(object, "class") to check if an object is of a certain class.

class(person)           # Returns "employee" "person"
is(person, "employee")  # Returns TRUE

Summary:

Classes in R, particularly using the S3 system, provide a flexible way to implement object-oriented-like behavior without the strictness seen in languages like Java. With S3, you can create classes, define methods for these classes, and make use of inheritance. The simplicity and flexibility of S3 make it popular, but for more formal class definitions, the S4 system or the R6 package can be used.

How to Create and Define Classes in R:

# Define a simple S3 class
setClass("Person", slots = c(name = "character", age = "numeric"))
person <- new("Person", name = "John", age = 30)

Object-Oriented Programming in R:

# Create methods for the Person class
setMethod("show", "Person", function(object) {
  cat("Name:", object@name, "\n")
  cat("Age:", object@age, "\n")
})

# Call the method
show(person)

Inheritance and Polymorphism in R Classes:

# Define a Student class inheriting from Person
setClass("Student", contains = "Person", slots = c(grade = "numeric"))
student <- new("Student", name = "Alice", age = 25, grade = 95)

# Polymorphism - show method works for both Person and Student
show(student)

S3 Classes vs S4 Classes in R:

# S3 class example
my_s3_object <- structure(list(name = "Bob", age = 40), class = "Person")

# S4 class example
setClass("S4Person", slots = c(name = "character", age = "numeric"))
my_s4_object <- new("S4Person", name = "Bob", age = 40)

Class Methods and Functions in R:

# Define a method for the Person class
setMethod("introduce", "Person", function(object) {
  cat("Hi, I'm", object@name, "and I'm", object@age, "years old.\n")
})

# Call the method
introduce(person)

Defining Constructors in R Classes:

# Define a constructor for the Person class
Person <- setClass(
  "Person",
  representation(name = "character", age = "numeric"),
  prototype = list(name = character(), age = numeric())
)

# Create an instance using the constructor
person_instance <- new("Person", name = "Sam", age = 35)

Accessing and Modifying Class Attributes in R:

# Access and modify attributes
person_name <- person@name
person@age <- 31