3.4 Attributes

Attributes are name-value pairs that attach metadata to an object(vector).

  • Name-value pairs: attributes have a name and a value
  • Metadata: not data itself, but data about the data

3.4.1 How?

3.4.1.1 Getting and Setting

Three functions:

  1. retrieve and modify single attributes with attr()
  2. retrieve en masse with attributes()
  3. set en masse with structure()

Single attribute

Use attr()

# some object
a <- c(1, 2, 3)

# set attribute
attr(x = a, which = "attribute_name") <- "some attribute"

# get attribute
attr(a, "attribute_name")
#> [1] "some attribute"

Multiple attributes

To set multiple attributes, use structure() To get multiple attributes, use attributes()

a <- 1:3
attr(a, "x") <- "abcdef"
attr(a, "x")
#> [1] "abcdef"

attr(a, "y") <- 4:6
str(attributes(a))
#> List of 2
#>  $ x: chr "abcdef"
#>  $ y: int [1:3] 4 5 6

# Or equivalently
a <- structure(
  1:3, 
  x = "abcdef",
  y = 4:6
)
str(attributes(a))
#> List of 2
#>  $ x: chr "abcdef"
#>  $ y: int [1:3] 4 5 6
Image Credit: Advanced R
Image Credit: Advanced R

3.4.2 Why

Three particularly important attributes:

  1. names - a character vector giving each element a name
  2. dimension - (or dim) turns vectors into matrices and arrays
  3. class - powers the S3 object system (we’ll learn more about this in chapter 13)

Most attributes are lost by most operations. Only two attributes are routinely preserved: names and dimension.

3.4.2.1 Names

Three Four ways to name:

# (1) When creating it: 
x <- c(A = 1, B = 2, C = 3)
x
#> A B C 
#> 1 2 3

# (2) By assigning a character vector to names()
y <- 1:3
names(y) <- c("a", "b", "c")
y
#> a b c 
#> 1 2 3

# (3) Inline, with setNames():
z <- setNames(1:3, c("a", "b", "c"))
z
#> a b c 
#> 1 2 3
proper diagram
proper diagram
# (4) By setting names--with {rlang}
a <- 1:3
rlang::set_names(
  x = a,
  nm = c("a", "b", "c")
)
#> a b c 
#> 1 2 3
simplified diagram
simplified diagram
  • You can remove names from a vector by using x <- unname(x) or names(x) <- NULL.
  • Thematically but not directly related: labelled class vectors with haven::labelled()

3.4.2.2 Dimensions

Create matrices and arrays with matrix() and array(), or by using the assignment form of dim():

# Two scalar arguments specify row and column sizes
x <- matrix(1:6, nrow = 2, ncol = 3)
x
#>      [,1] [,2] [,3]
#> [1,]    1    3    5
#> [2,]    2    4    6

# One vector argument to describe all dimensions
y <- array(1:24, c(2, 3, 4)) # rows, columns, no of arrays
y
#> , , 1
#> 
#>      [,1] [,2] [,3]
#> [1,]    1    3    5
#> [2,]    2    4    6
#> 
#> , , 2
#> 
#>      [,1] [,2] [,3]
#> [1,]    7    9   11
#> [2,]    8   10   12
#> 
#> , , 3
#> 
#>      [,1] [,2] [,3]
#> [1,]   13   15   17
#> [2,]   14   16   18
#> 
#> , , 4
#> 
#>      [,1] [,2] [,3]
#> [1,]   19   21   23
#> [2,]   20   22   24

# You can also modify an object in place by setting dim()
z <- 1:6
dim(z) <- c(2, 3) # rows, columns
z
#>      [,1] [,2] [,3]
#> [1,]    1    3    5
#> [2,]    2    4    6

a <- 1:24
dim(a) <- c(2, 3, 4) # rows, columns, no of arrays
a
#> , , 1
#> 
#>      [,1] [,2] [,3]
#> [1,]    1    3    5
#> [2,]    2    4    6
#> 
#> , , 2
#> 
#>      [,1] [,2] [,3]
#> [1,]    7    9   11
#> [2,]    8   10   12
#> 
#> , , 3
#> 
#>      [,1] [,2] [,3]
#> [1,]   13   15   17
#> [2,]   14   16   18
#> 
#> , , 4
#> 
#>      [,1] [,2] [,3]
#> [1,]   19   21   23
#> [2,]   20   22   24
3.4.2.2.1 Functions for working with vectors, matrices and arrays:
Vector Matrix Array
names() rownames(), colnames() dimnames()
length() nrow(), ncol() dim()
c() rbind(), cbind() abind::abind()
— t() aperm()
is.null(dim(x)) is.matrix() is.array()
  • Caution: A vector without a dim attribute set is often thought of as 1-dimensional, but actually has NULL dimensions.
  • One dimension?
str(1:3)                   # 1d vector
#>  int [1:3] 1 2 3
str(matrix(1:3, ncol = 1)) # column vector
#>  int [1:3, 1] 1 2 3
str(matrix(1:3, nrow = 1)) # row vector
#>  int [1, 1:3] 1 2 3
str(array(1:3, 3))         # "array" vector
#>  int [1:3(1d)] 1 2 3

3.4.3 Exercises

  1. How is setNames() implemented? How is unname() implemented? Read the source code.
Answer(s)

setNames() is implemented as:

setNames <- function(object = nm, nm) {
  names(object) <- nm
  object
}

Because the data argument comes first, setNames() also works well with the magrittr-pipe operator. When no first argument is given, the result is a named vector (this is rather untypical as required arguments usually come first):

setNames( , c("a", "b", "c"))
#>   a   b   c 
#> "a" "b" "c"

unname() is implemented in the following way:

unname <- function(obj, force = FALSE) {
  if (!is.null(names(obj))) 
    names(obj) <- NULL
  if (!is.null(dimnames(obj)) && (force || !is.data.frame(obj))) 
    dimnames(obj) <- NULL
  obj
}
unname() removes existing names (or dimnames) by setting them to NULL.
  1. What does dim() return when applied to a 1-dimensional vector? When might you use NROW() or NCOL()?
Answer(s)

dim() will return NULL when applied to a 1d vector.

One may want to use NROW() or NCOL() to handle atomic vectors, lists and NULL values in the same way as one column matrices or data frames. For these objects nrow() and ncol() return NULL:

x <- 1:10

# Return NULL
nrow(x)
#> NULL
ncol(x)
#> NULL

# Pretend it's a column vector
NROW(x)
#> [1] 10
NCOL(x)
#> [1] 1
  1. How would you describe the following three objects? What makes them different from 1:5?
x1 <- array(1:5, c(1, 1, 5))
x2 <- array(1:5, c(1, 5, 1))
x3 <- array(1:5, c(5, 1, 1))
Answer(s) 
x1 <- array(1:5, c(1, 1, 5))  # 1 row,  1 column,  5 in third dim.
x2 <- array(1:5, c(1, 5, 1))  # 1 row,  5 columns, 1 in third dim.
x3 <- array(1:5, c(5, 1, 1))  # 5 rows, 1 column,  1 in third dim.
  1. An early draft used this code to illustrate structure():
structure(1:5, comment = "my attribute")
#> [1] 1 2 3 4 5

But when you print that object you don’t see the comment attribute. Why? Is the attribute missing, or is there something else special about it?

Answer(s)

The documentation states (see ?comment):

Contrary to other attributes, the comment is not printed (by print or print.default).

Also, from ?attributes:

Note that some attributes (namely class, comment, dim, dimnames, names, row.names and tsp) are treated specially and have restrictions on the values which can be set.

We can retrieve comment attributes by calling them explicitly:

foo <- structure(1:5, comment = "my attribute")

attributes(foo)
#> $comment
#> [1] "my attribute"
attr(foo, which = "comment")
#> [1] "my attribute"