R Paste - ignore cases where at least one element I am pasting is NA

Question

I have the following vector:

vector1 <- c("A", "B", "C" , NA, NA, "D")

I want to apply this function paste ignoring NA values in vector1

vector2 <- paste("#", vector1, "something", sep = "")

and obtain this

vector2 <- c("#Asomething", "#Bsomething" , "#Csomething", NA, NA, "#Dsomething")

I want to avoid ex-post solutions using sub where I just get rid of elements containing the letters "NA" in the string.

I saw a similar question: suppress NAs in paste() however there they want to ignore NA and simply paste #something whereas I want NA to be displayed.

score 2 · Answer 1 · answered May 04 '20 at 11:35

2

Base R solution:

vector2 <- ifelse(is.na(vector1), NA_character_, paste("#", vector1, "something" ,sep = ""))

answered May 04 '20 at 11:35

hello_friend

5,682
1
11
15

Did not see your answer which is pretty much the same. Why do you use `NA_character_` and not just `NA`? – May 04 '20 at 11:42
@machine NA is of type logical, NA_character is of type character. vector_1 is a character vector, vector_2 is also explicitly a character vector. – hello_friend May 04 '20 at 11:44
But isn't it another vector than the OP asked for? `c("#Asomething", "#Bsomething" , "#Csomething" ,NA , NA, "#Dsomething")` – May 04 '20 at 11:45
@machine No. Its a matter of typing. As in, the data.type of the vector, explicitly typing prevents any problems of unintended type conversion; in this instance an unintended consquence may be conversion to a factor vetor, when OP wants a character vector. – hello_friend May 04 '20 at 12:32
I see. +1 from me! Did not know that it is possible to define NA in a certain class. – May 04 '20 at 13:36
@machine no worries, cheers ! Yes NA alone is of type logical ! Also interesting discussion on the negatives of the tidyverse post, post should not have been marked as duplicate! – hello_friend May 04 '20 at 13:37

score 2 · Accepted Answer · 2020-05-04T13:41:18.577

2

You can combine it with ifelse like this

ifelse(is.na(vector1), NA, paste("#", vector1, "something" ,sep = ""))

Although this was marked as answered please also see the answer of @hello_friend which may be more appropriate.

edited May 04 '20 at 13:41

answered May 04 '20 at 11:40

score 1 · Answer 3 · answered May 04 '20 at 11:35

I think you need to use sapply() with a anonymous function.

Please, have a look at the following code which does what you want.

vector1 <- c("A", "B" , "C" ,NA_character_ , NA_character_, "D")
vector2 <- sapply( vector1, function( x ) 
  ifelse( is.na( x ), 
          NA_character_, 
          paste("#", x, "something" ,sep = "") ),
  USE.NAMES = FALSE
)
vector2
#> [1] "#Asomething" "#Bsomething" "#Csomething" NA            NA           
#> [6] "#Dsomething"

^{Created on 2020-05-04 by the reprex package (v0.3.0)}

R Paste - ignore cases where at least one element I am pasting is NA

3 Answers3