Get Back for R

David Hugh-Jones

This is lame

letters[length(letters)]   # so unreadable

[1] "z"

letters[(length(letters) - 3):(length(letters))]   # even worse

[1] "w" "x" "y" "z"

letters[length(letters) - 3:0]   # what am I a maths nerd

[1] "w" "x" "y" "z"

Tail

tail(letters, 4)   # ok

[1] "w" "x" "y" "z"

tail(letters, 2:4)  # boring

Error in tail.default(letters, 2:4): invalid 'n' - must have length one when dim(x) is NULL, got 3

tail(letters, 1) <- "Z"   # boooring

Error in tail(letters, 1) <- "Z": could not find function "tail<-"

Reverse of helpful

rev(letters)[1:3]                     # ok...

[1] "z" "y" "x"

rev(letters)[1:3] <- c("Z", "Y", "X") # why :-(

Error in rev(letters)[1:3] <- c("Z", "Y", "X"): could not find function "rev<-"

Why don’t people do this, are they stupid

.back_ix <- function (x, ix) {
  stopifnot(is.numeric(ix), all(abs(ix) <= length(x)))
  ix <- ix[ix != 0]
  neg <- ix < 0
  ix <- length(x) + 1 - abs(ix)
  ix[neg] <- -1 * ix[neg]
  ix
}

back <- function (x, ix) {
  x[.back_ix(x, ix)]
}

`back<-` <- function (x, ix, value) {
  x[.back_ix(x, ix)] <- value
  x
}

So obvious

back(letters, 4)

[1] "w"

back(letters, 1:4) # as it should be

[1] "z" "y" "x" "w"

back(letters, 2:4) # not a problem

[1] "y" "x" "w"

back(letters, 4:1) # just count from the back

[1] "w" "x" "y" "z"

As God intended

back(letters, 4) <- "W"
back(letters, 2:3) <- c("Y", "X")

cat(back(letters, 10:1)) 

q r s t u v W X Y z

Backwardation

letters[-(1:21)]

[1] "v" "W" "X" "Y" "z"

back(letters, -(1:21)) # of course!

[1] "a" "b" "c" "d" "e"

letters[0]

character(0)

back(letters, 0)

character(0)

letters[30]

[1] NA

back(letters, 30) # don't get over-excited

Error in .back_ix(x, ix): all(abs(ix) <= length(x)) is not TRUE

Gist

No I will not be producing a CRAN package

Conclusion