• R Console
• Read-Eval-Print Loop
• The Prompt
  • By default… >
  • Continuation… +

• R code can be run in batch mode i.e. non-interactively.
• Instruction are saved in a text file known as a script.
• File extension .R or .r.
• Script is read line by line from top to bottom.
• Each line is read from left to right.
• This is how R code is shared, including that in packages.
• Can be written with any text editor e.g. Notepad.

• These are ‘scalar’ quantities that are sent to the R interpreter
• Usually these scalars are numeric data like integers or doubles, booleans (TRUE/FALSE),
• When used like this, they are returned as-is.

26L

## [1] 26

2.71828

## [1] 2.71828

0xef

## [1] 239

"Hello"

## [1] "Hello"

NULL

## NULL

Some other special constants that one may come across in mathematical computations

• Inf, -Inf
• NaN

1/0
0/0

## [1] Inf
## [1] NaN

• R instructions are called statements
• Statements are always executed one at a time.
• It is possible to have 2 or more statements on a single line, separated by a semicolon (;).
• If a statement cannot be successfully evaluated, an ERROR is thrown and execution terminated. The REPL is returned to the prompt, ready to receive new input.
• Sometimes, a problematic operation can be signalled by a WARNING - code executes successfully but notice is given after the fact.

1 + 1

## [1] 2

6 / 4

## [1] 1.5

1 + 1; 6 / 4  # identical 

## [1] 2

## [1] 1.5

1 + "a"

## Error in 1 + "a": non-numeric argument to binary operator

• Bounded by double quotes
• Created using single or double quotes
• Use of escapes with the backslash (\) for nested quotes and special characters.

"He said: "I have a dream""

## Error: <text>:1:12: unexpected symbol
## 1: "He said: "I
##                ^

"He said: \"I have a dream!\""

## [1] "He said: \"I have a dream!\""

"The right word is 'cute'."

## [1] "The right word is 'cute'."

• Objects created in R can be bound to symbols
• These named objects are variables
• When the name is read and evaluated, the object’s value is printed by R.
• The operation that binds a name to an object is called assignment.
• <- is the assignment operator.
• = works as well

## Create a variable
life <- 42L

## Evaluate the variable by printing out its value
life

## [1] 42

Rules:

• Can contain any alphanumeric character, dot (.) or underscore (_)
• Cannot start with _
• Cannot start with number
• Cannot contain special characters other than e.g. +, -, ?, etc
• Cannot contain whitespace
• Can be overridden by surrounding with backticks `

myobject <- "a"

my.object <- TRUE

.myObject <- 1e-12

_myObject <- 6.3328   # illegal
4myobject <- 6798L    # illegal

`4myobject` <- 6798L  # backquotes make this permissible

• R code is case sensitive - sum is different from Sum

• Everything that does something in R is a function
• Either return a value or have side effects.
• Functions in R are held in packages.
• R comes with default packages that are made available during each R session
• For the contents of a package to be accessible, it must be attached to the search path.
• When a function is called, R searches for it in the attached packages.
• The base package is the most important packages that is ALWAYS available.
• Other packages that are usually attached by default are utils, stats, methods, graphics, grDevices, datasets
• Between these packages, we have some 2,512 objects.

• Arithmetic
  • +, -, *, /
  • ^, %%, %/% (exponentiation, modulo, integer division)
• Logical
  • &, |, !
• Comparison
  • ==, !=
  • <, >, <=, >=

Arithmetic

4^3

## [1] 64

7 / 5

## [1] 1.4

7 %% 5

## [1] 2

7 %/% 5

## [1] 1

Logical

TRUE & FALSE

## [1] FALSE

TRUE | FALSE

## [1] TRUE

!FALSE

## [1] TRUE

!TRUE

## [1] FALSE

Comparison

x <- 7L
x == 7

## [1] TRUE

x != 7

## [1] FALSE

x < 8

## [1] TRUE

x > 7

## [1] FALSE

x >= 7

## [1] TRUE

x <= 3

## [1] FALSE

To use a function in R:

• Write its name with open and closing parentheses e.g. ()
• Write its arguments, if any, within the parentheses
• Run i.e. evaluate the function

func(arg1, arg2)                     # 1. signature

func(arg1 = "input1", arg2 = 4.333)  # 2. function call

func("input1", 4.333)                # 3. Same as 2

func(arg2 = 4.333, arg1 = "input1")  # 4. Same as 2 & 3

func(4.333, "input1")                # 5. Likely to fail

• Any code, whether in full or in part, that is started with a pound (#) sign.
• Such portions of code are ignored by the R interpreter
• An important part of coding - in any language
• Used to describe what the code is doing and expantiate on constructs used.
• Vital for fellow programmers to interact with code, as well as “future self”.

# This is a comment - the whole line will be ignored
"This is a valid string; it will be evaluated" # but this part won't

## [1] "This is a valid string; it will be evaluated"

• The working directory
• The Workspace
• The search path

getwd()              # Displays the current working directory

setwd("path/to/dir") # Sets working directory to a new location

list.files()         # Lists contents of directory, defaults to getwd()

• Also known as the global environment
• Names of objects created by session are bound here
• Chained to attached packages so their contents are accessible

ls()

## character(0)

x <- 1e7
ls()

## [1] "x"

rm(x)    # appropriate arguments to remove multiple objects
ls()

## character(0)

• Using 3rd party packages
  • Installing packages (install.packages())
  • Attaching or loading packages (library() and :: operator)
• The search path
  • base package first, then other default packages.
  • Additional package(s) attached by user/code

search()

## [1] ".GlobalEnv"        "package:stats"     "package:graphics" 
## [4] "package:grDevices" "package:utils"     "package:datasets" 
## [7] "package:methods"   "Autoloads"         "package:base"

library(MASS)

search()

##  [1] ".GlobalEnv"        "package:MASS"      "package:stats"    
##  [4] "package:graphics"  "package:grDevices" "package:utils"    
##  [7] "package:datasets"  "package:methods"   "Autoloads"        
## [10] "package:base"

• Your R installation comes with a rich set of documentation and help files.
• Every R package has its own set of help files.
• Function documentation can be read offline with ? operator.
• Also use online resources:
  • Web search
  • StackOverflow
  • Social media, notably Twitter (#rstats) and Reddit

R Manuals

help.start()           # Opens R Manuals (best for reference purpose)

Function help

?sum

help("sum")            # Help for the function 'sum'

Package documentation

help(package = "MASS") # Help on the MASS package

library(help = "MASS") # Summary info on the MASS package