Details
First, it requires usage of sort() over .[order(.)].
sort() is the dedicated option to sort a list or vector. It is more legible
and around twice as fast as .[order(.)], with the gap in performance
growing with the vector size.
Second, it requires usage of is.unsorted() over equivalents using sort().
The base function is.unsorted() exists to test the sortedness of a vector.
Prefer it to inefficient and less-readable equivalents like
x != sort(x). The same goes for checking x == sort(x) – use
!is.unsorted(x) instead.
Moreover, use of x == sort(x) can be risky because sort() drops missing
elements by default, meaning == might end up trying to compare vectors
of differing lengths.
See also
linters for a complete list of linters available in lintr.
Examples
# will produce lints
lint(
text = "x[order(x)]",
linters = sort_linter()
)
#> <text>:1:3: warning: [sort_linter] sort(x, na.last = TRUE) is better than x[order(x)]. Note that it's always preferable to save the output of order() for the same variable as a local variable than to re-compute it.
#> x[order(x)]
#> ^~~~~~~~
lint(
text = "x[order(x, decreasing = TRUE)]",
linters = sort_linter()
)
#> <text>:1:3: warning: [sort_linter] sort(x, decreasing = TRUE, na.last = TRUE) is better than x[order(x, decreasing = TRUE)]. Note that it's always preferable to save the output of order() for the same variable as a local variable than to re-compute it.
#> x[order(x, decreasing = TRUE)]
#> ^~~~~~~~~~~~~~~~~~~~~~~~~~~
lint(
text = "sort(x) == x",
linters = sort_linter()
)
#> <text>:1:1: warning: [sort_linter] Use !is.unsorted(x) to test the sortedness of a vector.
#> sort(x) == x
#> ^~~~~~~~~~~~
# okay
lint(
text = "x[sample(order(x))]",
linters = sort_linter()
)
#> ℹ No lints found.
lint(
text = "y[order(x)]",
linters = sort_linter()
)
#> ℹ No lints found.
lint(
text = "sort(x, decreasing = TRUE) == x",
linters = sort_linter()
)
#> ℹ No lints found.
# If you are sorting several objects based on the order of one of them, such
# as:
x <- sample(1:26)
y <- letters
newx <- x[order(x)]
newy <- y[order(x)]
# This will be flagged by the linter. However, in this very specific case,
# it would be clearer and more efficient to run order() once and assign it
# to an object, rather than mix and match order() and sort()
index <- order(x)
newx <- x[index]
newy <- y[index]