Saving output from a function containing loop and ifelse

Question

Basically I wan to be able to save a function output in numeric format into an R object. The function contains a loop and ifelse arguments which may contribute to my difficulties in achieving this.

I would like to create a function like the following:

fnct <- function(x, y) {
  for (i in x) {
    ifelse(i >= y[1] & i <= y[2], y[3], 
            ifelse(i < y[1], y[1], 
                    ifelse((max(x) - y[2] > y[1]), y[2], 
                           NA)))
  }
}

If I run this code, nothing happens.

fnct(c(1:150), c(20, 80, 50)) 

If I modify the code as follows by adding print arguments, I get an ouptut:

fnct <- function(x, y) {
  for (i in x) {
    ifelse(i >= y[1] & i <= y[2],
            print(y[3]),
            ifelse(i < y[1],
                    print(y[1]), ifelse((max(x) - y[2] > y[1]),
                                        print(y[2]),
                                        print(NA))))
  }
}

But I cannot save it to an r object.

I have also tried the following:

fnct <- function(x, y) {
  res <- matrix(nrow=length(x))
  for (i in x) {
    ifelse(i >= y[1] & i <= y[2],
            res[i] <- (y[3]),
            ifelse(i<y[1],
                   res[i] <- (y[1]), ifelse((max(x) - y[2] > y[1]),
                                           res[i] <- (y[2]),
                                           res[i] <- (NA))))
  }
  return(res)
}

which works in this case:

fnct(c(1:150), c(20, 80, 50)) 

but not for this:

fnct(seq(1.575, 5.125, by=0.05), c(1.5, 3, 3.5))

Answer 1

1. With a for loop.

You must create a vector where you save the ifelse results.
Also, note that the question's last ifelse does not depend on the index so it can be pre-computed.

fnct <- function(x,y) {
  out <- numeric(length(x))
  pre_comp <- ifelse(max(x) - y[2] > y[1], y[2], NA)
  for (i in seq_along(x)) {
    out[i] <- ifelse (x[i] >= y[1] & x[i] <= y[2],
            y[3],
            ifelse (x[i] < y[1],
                    y[1], pre_comp))
  }
  out
}
fnct(c(1:150),c(20,80,50))
#>   [1] 20 20 20 20 20 20 20 20 20 20 20 20 20 20 20 20 20 20 20 50 50 50 50 50 50
#>  [26] 50 50 50 50 50 50 50 50 50 50 50 50 50 50 50 50 50 50 50 50 50 50 50 50 50
#>  [51] 50 50 50 50 50 50 50 50 50 50 50 50 50 50 50 50 50 50 50 50 50 50 50 50 50
#>  [76] 50 50 50 50 50 80 80 80 80 80 80 80 80 80 80 80 80 80 80 80 80 80 80 80 80
#> [101] 80 80 80 80 80 80 80 80 80 80 80 80 80 80 80 80 80 80 80 80 80 80 80 80 80
#> [126] 80 80 80 80 80 80 80 80 80 80 80 80 80 80 80 80 80 80 80 80 80 80 80 80 80

^{Created on 2024-04-12 with reprex v2.1.0}

---

2. With a *apply loop

The loop but with a *apply function becomes simpler, there is no need to create the results vector beforehand.

fnct <- function(x, y) {
  pre_comp <- ifelse(max(x) - y[2] > y[1], y[2], NA)
  sapply(x, \(xx) {
    ifelse (xx >= y[1] & xx <= y[2],
            y[3],
            ifelse (xx < y[1],
                    y[1], pre_comp))
  })
}
fnct(c(1:150),c(20,80,50))
#>   [1] 20 20 20 20 20 20 20 20 20 20 20 20 20 20 20 20 20 20 20 50 50 50 50 50 50
#>  [26] 50 50 50 50 50 50 50 50 50 50 50 50 50 50 50 50 50 50 50 50 50 50 50 50 50
#>  [51] 50 50 50 50 50 50 50 50 50 50 50 50 50 50 50 50 50 50 50 50 50 50 50 50 50
#>  [76] 50 50 50 50 50 80 80 80 80 80 80 80 80 80 80 80 80 80 80 80 80 80 80 80 80
#> [101] 80 80 80 80 80 80 80 80 80 80 80 80 80 80 80 80 80 80 80 80 80 80 80 80 80
#> [126] 80 80 80 80 80 80 80 80 80 80 80 80 80 80 80 80 80 80 80 80 80 80 80 80 80

^{Created on 2024-04-12 with reprex v2.1.0}

---

Edit

Following user @lotus's comment I realized that the loop is not needed at all.

fnct <- function(x, y) {
  ifelse (x >= y[1] & x <= y[2],
          y[3],
          ifelse (x < y[1],
                  y[1], 
                  ifelse(max(x) - y[2] > y[1], 
                         y[2], NA)))
}
fnct(c(1:150),c(20,80,50))
#>   [1] 20 20 20 20 20 20 20 20 20 20 20 20 20 20 20 20 20 20 20 50 50 50 50 50 50
#>  [26] 50 50 50 50 50 50 50 50 50 50 50 50 50 50 50 50 50 50 50 50 50 50 50 50 50
#>  [51] 50 50 50 50 50 50 50 50 50 50 50 50 50 50 50 50 50 50 50 50 50 50 50 50 50
#>  [76] 50 50 50 50 50 80 80 80 80 80 80 80 80 80 80 80 80 80 80 80 80 80 80 80 80
#> [101] 80 80 80 80 80 80 80 80 80 80 80 80 80 80 80 80 80 80 80 80 80 80 80 80 80
#> [126] 80 80 80 80 80 80 80 80 80 80 80 80 80 80 80 80 80 80 80 80 80 80 80 80 80

^{Created on 2024-04-12 with reprex v2.1.0}