Vectorize addition whilst removing NA

Question

I am trying to add a scalar to a vector. However, when my scalar is NA, I want it to be ignored, similar to the na.rm = TRUE argument used in sum.

ie.

1:3 + x

returns NA NA NA if x is assigned NA. I would like it to return 1 2 3. I tried

sum(1:3, NA, na.rm = T)

but this does not vectorize the addition and just returns 1+2+3 = 6.

Is there a neat way to do this?

Thank you

I'd use `1:3 + coalesce(x, 0)` (or `1:3 + ifelse(is.na(x), 0, x)`, if you don't have `dplyr` loaded) — Gregor Thomas, Dec 03 '20 at 02:32
Another option would be `rowSums(cbind(1:3, x), na.rm = TRUE)` — Gregor Thomas, Dec 03 '20 at 02:33
You could get tricky and define your own new operator - `\`%+%\` <- function(x,y) if(is.na(y)) x else x + y` then `1:3 %+% NA` — thelatemail, Dec 03 '20 at 04:53

score 1 · Accepted Answer · answered Dec 03 '20 at 04:39

1

You can use replace :

x <- NA
1:3 + replace(x, is.na(x), 0)
#[1] 1 2 3

x <- 2
1:3 + replace(x, is.na(x), 0)
#[1] 3 4 5

answered Dec 03 '20 at 04:39

Ronak Shah

377,200
20
156
213

Vectorize addition whilst removing NA

1 Answers1