If you are new to the R language, you can find a detailed introduction, language history, and functionality on the official R site (http://www.r-project.org/). When you are ready to download and install R, please connect to the comprehensive R archive network (http://cran.r-project.org/).

Perform the following steps in order to create your first R function:

R functions are a block of organized and reusable statements, which makes programming less repetitive by allowing you to reuse code. Additionally, by modularizing statements within a function, your R code will become more readable and maintainable.

By following these steps, you can now create two addnum and addnum2 R functions, and you can successfully add two input arguments with either function. In R, the function usually takes the following form:

FunctionName<- function (arg1, arg2) {
body
return(expression)
}

FunctionName is the name of the function, and arg1 and arg2 are arguments. Inside the curly braces, we can see the function body, where a body is a collection of a valid statement, expression, or assignment. At the bottom of the function, we can find the return statement, which passes expression back to the caller and exits the function.

The addnum function is in standard function syntax, which contains both body and return statement. However, you do not necessarily need to put a return statement at the end of the function. Similar to the addnum2 function, the function itself will return the last expression back to the caller.

If you want to view the composition of the function, simply type the function name on the interactive shell. You can also examine the body and formal arguments of the function further using the body and formal functions. Alternatively, you can use the args function to obtain the argument list of the function.

If you want to see the documentation of a function in R, you can use the help function or simply type ? in front of the function name. For example, if you want to examine the documentation of the sum function, you would do the following:

>help(sum)
> ?sum

Ensure that you completed the previous recipe by installing R on your operating system.

Perform the following steps to create a function with different types of argument lists:

R provides a flexible argument binding mechanism when creating functions. In this recipe, we first create a function called defaultag with two formal arguments: x and y. Here, the y argument has a default value, defined as 5. Then, when we make a function call by passing 3 to defaultarg, it passes 3 to x and 5 to y in the function and returns 13. Besides passing a scalar as the function input, we can pass a vector (or any other data type) to the function. In this example, if we pass a 1:3 vector to defaultarg, it returns a vector.

Moving on, we can see how arguments bind to the function. When calling a function by passing an argument without a parameter name, the function binds the passing value by position. Take step 4 as an example; the first 3 argument matches to x, and 6 matches to y, and it returns 15. On the other hand, you can pass arguments by name. In step 5, we can pass named arguments to the function in any order. Thus, if we pass y=6 and x=3 to defaultarg, the function returns 15.

Furthermore, we can use an argument as a control statement. In step 6, we specify three formal arguments: x, y, and type, in which the type argument has the default value defined as sum. Next, we can specify the value for the type argument as a condition in the if-else control flow. That is, when we pass sum to type, it returns the summation of x and y. When we pass mean to type, it returns the average of x and y. When we pass any value other than sum and mean to the type argument, it returns the product of x and y.

Lastly, we can pass an unspecified number of arguments to the function using the ... notation. In the final step of this example, if we pass only 3 and 5 to the function, the function first passes 3 to x and 5 to y. Then, the function adds 2 to x, multiplies y by 2, and sums the value of both x and y. However, if we pass more than two arguments to the function, the function will also sum the additional parameters.

In addition to giving a full argument name, we can abbreviate the argument name when making a function call:

>funcarg(3,5, t = 'unknown')
[1] 15

Here, though we do not specify the argument's name, type, correctly, the function passes a value of unknown to the argument type, and returns 15 as the output.

Ensure that you completed the previous recipes by installing R on your operating system.

Perform the following steps to work with the environment:

We can regard an R environment as a place to store and manage variables. That is, whenever we create an object or a function in R, we add an entry to the environment. By default, the top-level environment is the R_GlobalEnv global environment, and we can determine the current environment using the environment function. Then, we can use either .GlobalEnv or globalenv to print the global environment, and we can compare the environment with the identical function.

Besides the global environment, we can actually create our environment and assign variables into the new environment. In the example, we created the myenv environment and then assigned x <- 3 to myenv by placing a dollar sign after the environment name. This allows us to use the ls function to list all variables in myenv and global environment. At this point, we find x in myenv, but we can only find myenv in the global environment.

Moving on, we can determine the environment of a function. By creating a function called addnum, we can use environment to get its environment. As we created the function under global environment, the function obviously belongs to the global environment. On the other hand, when we get the environment of the lm function, we get the package name instead. That means that the lm function is in the namespace of the stat package.

Furthermore, we can print out the current environment inside a function. By invoking the addnum2 function, we can determine that the environment function outputs a different environment name from the global environment. That is, when we create a function, we also create a new environment for the global environment and link a pointer to its parent environment. To further examine this characteristic, we create another addnum3 function with a func1 nested function inside. At this point, we can print out the environment inside func1 and addnum3, and it is possible that they have completely different environments.

To get the parent environment, we can use the parent.env function. In the following example, we can see that the parent environment of parentenv is R_GlobalEnv:

>parentenv<- function(){
+ e <- environment()
+ print(e)
+ print(parent.env(e))
+ }
>parentenv()
<environment: 0x0000000019456ed0>
<environment: R_GlobalEnv>

Ensure that you completed the previous recipes by installing R on your operating system.

Perform the following steps to understand how the scoping rule works:

There are two different types of variable binding methods: one is lexical binding, and the other is dynamic binding. Lexical binding is also called static binding in which every binding scope manages variable names and values in the lexical environment. That is, if a variable is lexically bound, it will search the binding of the nearest lexical environment. In contrast to this, dynamic binding keeps all variables and values in the global state. That is, if a variable is dynamically bound, it will bind to the most recently created variable.

To demonstrate how lexical binding works, we first create an x variable and assign 5 to x in the global environment. Then, we can create a function named tmpfunc. The function outputs x + 3 as the return value. Even though we do not assign any value to x within the tmpfunc function, x can still find the value of x as 5 in the global environment.

Next, we create another function named parentfunc. In this function, we assign x to 3 and create a childfunc nested function (a function defined within a function). At the bottom of the parentfunc body, we invoke childfunc as the function return. Here, we find that the function uses the x defined in parentfunc instead of the one defined outside parentfunc. This is because R searches the global environment for a matched symbol name, and then subsequently searches the namespace of packages on the search list.

Moving on, let's take a look at what will return if we create an x variable as a string in the global state and assign an x local variable to 5 within the function. When we invoke the localassign function, we discover that the function returns 5 instead of the string value. On the other hand, if we print out the value of x, we still see string in return. While the local variable and global variable have the same name, the assignment of the function does not alter the value of x in global state. If you need to revise the value of x in the global state, you can use the <<- notation instead.

In order to examine the search list (or path) of R, you can type search() to list the search list:

>search()
[1] ".GlobalEnv""tools:rstudio"
[3] "package:stats" "package:graphics"
[5] "package:grDevices" "package:utils"
[7] "package:datasets" "package:methods"
[9] "Autoloads" "package:base"

Ensure that you completed the previous recipes by installing R on your operating system.

Perform the following steps to create a closure in function:

In R, you do not have to create a function with the actual name. Instead, you can use closure to integrate methods within objects. Thus, you can create a smaller and simpler function within another object to accomplish complicated tasks.

In our first example, we illustrated how a normally-named function is created. We can simply invoke the function by passing values into the function. On the other hand, we demonstrate how closure works in our second example. In this case, we do not need to assign a name to the function, but we can still pass the value to the anonymous function and obtain the return value.

Next, we demonstrate how to add a closure within a maxval named function. This function simply returns the maximum value of two passed parameters. However, it is possible to use closure within any other function. Moreover, we can use closure as an argument in higher order functions, such as lapply and sapply. Here, we can input an anonymous function as a function argument to return the multiplication of 10 and the first value of any vector within a given list.

Furthermore, we can specify a single function, or we can store functions in a list. Therefore, when we want to apply multiple functions to a given vector, we can pass the function calls as an argument list to the lapply function.

Besides using closure within a lapply function, we can also pass a closure to other functions of the apply function family. Here, we demonstrate how we can pass the same closure to the sapply function:

> x <- c(1,10,100)
> y <- c(2,4,6)
> z <- c(30,60,90)
> a <- list(x,y,z)
>sapply(a, function(e){e[1] * 10})
[1] 10 20 300

Ensure that you completed the previous recipes by installing R on your operating system.

Perform the following steps to see how lazy evaluation works:

R performs a lazy evaluation to evaluate an expression if its value is needed. This type of evaluation strategy has the following three advantages:

In this recipe, we demonstrate some lazy evaluation examples in the R code. In our first example, we create a function with two arguments, x and y, but return only x. Due to the characteristics of lazy evaluation, we can successfully obtain function returns even though we pass the value of x to the function. However, if the function return includes both x and y, as step 2 shows, we will get an error message because we only passed one value to the function. If we set a default value to y, then we do not necessarily need to pass both x and y to the function.

As lazy evaluation has the advantage of creating an infinite data structure without an infinite loop, we use a Fibonacci number generator as the example. Here, this function first creates an infinite list of Fibonacci numbers and then extracts the nth element from the list.

Additionally, we can use the force function to check whether y exists:

>lazyfunc3<- function(x, y){
+ force(y)
+ x
+ }
>lazyfunc3(3)
Error in force(y) : argument "y" is missing, with no default
>input_function<- function(x, func){
+ func(x)
+ }
>input_function(1:10, sum)
[1] 55

Ensure that you completed the previous recipes by installing R on your operating system.

Perform the following steps to create an infix operator in R:

In a standard function, if we want to perform some operations on the a and b variables, we would probably create a function in the form of func(a,b). While this form is the standard function syntax, it is harder to read than regular mathematical notation, (that is, a * b). However, we can create an infix operator to simplify the function syntax.

Before creating our infix operator, we examine different syntax when we apply a binary operator on two variables. In the first step, we demonstrate how to perform arithmetic operations with binary operators. Similar to a standard mathematical formula, all we need to do is to place a binary operator between two variables. On the other hand, we can transform the representation from infix form to prefix form. Like a standard function, we can use the binary operator as the function name, and then we can place the variables in between the parentheses.

In addition to using a predefined infix operator in R, the user can define the infix operator. To create an operator, we need to name the function that starts and ends with %, and surround the name with a single quote (') or back tick (`). Here, we create an infix operator named %match% to extract the interaction between two vectors. We can also create another infix function named %diff% to extract the set difference between two vectors. Lastly, though we can apply the created infix function to more than two vectors, we can use the reduce function to apply the %match% operation on the list.

We can also overwrite the existing operator by creating an infix operator with the same name:

>'+' <- function(x,y) paste(x,y, sep = "|")
> x = '123'
> y = '456'
>x+y
[1] "123|456"

Here, we can concatenate two strings with the + operator.

Ensure that you completed the previous recipes by installing R on your operating system.

Perform the following steps to create a replacement function in R:

In this recipe, we first demonstrated how we could use the names function to assign argument names for each value. This type of function method may appear confusing, but it is actually what the replacement function does: assigning the value to the function call. We then illustrated how this function works in a standard function form, which we achieved by placing an assignment arrow (<-) after the function name and placing the x object and value between the parentheses.

Next, we learned how to create our replacement function. We made a function named erase, which removed certain values from a given object. We invoked the function by wrapping the vector to replace within the erase function and assigning the value to remove on the right-hand side of the assignment notation. Alternatively, we can still call the replacement function by placing an assignment arrow after erase as the function name. In addition to removing a single value from a given vector object, we can also remove multiple values by placing a vector on the right-hand side of the assignment function.

Furthermore, we can remove the values of certain positions with the replacement function. Here, we only needed to add a position argument between the object and value within the parentheses. As our last step shows, we removed 2 from the second value of the list with our newly-created replacement function.

As mentioned earlier, names<- is a replacement function. To examine whether a function is a replacement function or not, use the get function:

>get("names<-")
function (x, value) .Primitive("names<-")

Ensure that you completed the previous recipes by installing R on your operating system.

Perform the following steps to handle errors in an R function:

Similar to other programming languages, R provides developers with an error-handling mechanism. However, the error-handling mechanism in R is implemented in the function instead of a pure code block. This is due to the fact that all operations are pure function calls.

In the first step, we demonstrate what will output if we add an integer to a string. If the operation is invalid, the system will print an error message on the console. There are three basic types of error handling messages in R, which are error, warning, and interrupt.

Next, we create a function named addnum, which is designed to return the addition of two arguments. However, sometimes you will pass an unexpected type of input (for example, string) into a function. For this condition, we can add an argument type check condition before the return statement. If none of the input data types is numeric, the stop function will print an error message quoted in the stop function.

Besides using the stop function, we can use a warning function instead to handle an error. However, only using a warning function, the function process will not terminate but proceed to return a + b. Thus, we might find both an error and warning message displayed on the console. To suppress the warning message, we can set warn=2 in the options function, or we can use suppressWarnings instead to mute the warning message. On the other hand, we can also use the stopifnot function to check whether the argument is valid or not. If the input argument is invalid, we can stop the program and print an error message on the screen.

Moving on, we can catch the error using the try function. Here, we store the error message into errormsg in the operation of adding a character string to an integer. However, the function will still print the error message on the screen. We can mute the message by setting a silent argument to TRUE. Furthermore, the try function is very helpful if don't want a for-loop being interrupted by unexpected errors. Therefore, we first demonstrate how an error may unexpectedly interrupt the loop execution. In that step, we may find that the loop execution stops, and we have successfully assigned only three variables to res. However, we can actually proceed with the for-loop execution by wrapping the code into a try function.

Besides the try function, we can use a more advanced error-handling function, tryCatch, to handle errors including warning and error. We use the tryCatch function in the following manner:

tryCatch({
result<- expr
}, warning = function(w) {
# handling warning
}, error = function(e) {
# handling error
}, finally = {
#Cleanup
})

In this function, we can catch warning and error messages in different function code blocks. By following the function form, we can create a function named dividenum. The function first performs numeric division; if any error occurs, we can catch the error and print an error message in the error function. At the end of the block, we remove any defined value within the function and print the message of clean variable. At this point, we can test how this function works in three different situations: performing a normal division, dividing a string from a string, and passing only one parameter into the function. We can now observe the output message under different conditions. In the first condition, the function prints out the division result, followed by clean variable because it is coded in the block of finally. For the second condition, the function first catches the error of missing value in the error block and then outputs clean variable at the end. For the last condition, while we do not catch the error of not passing a value to the b parameter, the function still returns an error message first and then prints clean variable on the console.

If you want to catch the error message while using the tryCatch function, you can put a conditionMessage to the error argument of the tryCatch function:

>dividenum<- function(a,b){
+ result<- tryCatch({
+ a/b
+ }, error = function(e) {
+ conditionMessage(e)
+ }
+ )
+ result
+ }
>dividenum(3,5)
[1] 0.6
>dividenum(3,"hello")
[1] "non-numeric argument to binary operator"

In this example, if you pass two valid numeric arguments to the dividenum function, the function returns a computation of 3/5 as output. On the other hand, if you pass a non-numeric value to the function, the function catches the error with the conditionMessage function and returns the error as the function output.

Make sure that you know how a function and works and how to create a new function.

Perform the following steps to debug an R function:

As it is inevitable for all code to include bugs, an R programmer has to be well prepared for them with a good debugging toolset. In this recipe, we showed you how to debug a function with the debug, browser, trace, and traceback functions.

In the first section, we explained how to debug a function by applying debug to an existing function. We first made a function named debugfunc, with two input arguments: x and y. Then, we applied a debug function onto debugfunc. Here, we applied the debug function on the name, argument, or function. At this point, whenever we invoke debugfunc, our R console will enter into a browser mode with Browse as the prompt at the start of each line.

Browser mode enables us to make a single step through the execution of the function. We list the single-letter commands that one can use while debugging here:

In the following operations, we first use help to list all possible commands. Then, we type n to step to the next line. Next, we type objects and ls to list all current objects. At this point, we can type the variable name to find out the current value of each object. Finally, we can type Q to quit debugging mode, and use undebug to unflag the function.

Besides using the debug function, we can insert the browser function within the code to debug it. After we have inserted browser() into debugfunc2, whenever we invoke the function, the R function will step right into the next line below the browser function. Here, we can perform any command mentioned in the previous command table. If you want to move among frames or return to the top level of debugging mode, we can use the recover function. Additionally, we can use the trace function to insert debugging code into the function. Here, we assign what to trace as debugfunc2, and set the tracer to examine whether y is missing. If y is missing, it will execute the browser() function. At that argument, we set 4 to the argument so that the tracer code will be inserted at line 4 of debugfunc2. Then, when we call the debugfunc2 function, the function enters right into where the tracer is located and executes the browser function as the y argument is missing.

Finally, we introduce the traceback function, which prints the calling stack of the function. At this step, we pass two unassigned parameters, x and y, into an lm linear model fitting function. As we do not assign any value to these two parameters, the function returns an error message in the console output. To understand the calling stack sequence, we can use the traceback function to print out the stack.

Besides using the command line, we can use RStudio to debug functions:

You can now use the command line or dropdown menu of Debug to debug the function:

Figure 5: Use debug functions