Concatenate strings with NA handling

str_c_narm concatenates strings similar to base::paste() or stringr::str_c(), but with different NA handling. NAs are dropped row-wise prior to concatenation. See Details and Examples.

Usage

str_c_narm(
  ...,
  sep = "",
  collapse = NULL,
  if_all_na = c("empty", "NA", "remove")
)

Arguments

...

character vectors or vectors coercible to character. May also be a single data frame (to accommodate dplyr::across() and pick()).

sep

separator to insert between input vectors.

collapse

optional character string to combine results into a single string.

if_all_na

what to do if all values in a row are NA:

• "empty": (default) returns an empty string.

• "NA": returns NA.

• "remove": removes the row from the output.

Value

A character vector.

Details

str_c_narm() provides alternative missing value handling compared to base::paste() and stringr::str_c(). Given some vectors with missing values:

#> (dat <- tibble::tibble(v1 = c("a1", "b1", "c1", NA),
                      v2 = c("a2", NA, "c2", NA),
                      v3 = c("a3", "b3", NA, NA)))
# A tibble: 4 × 3
  v1    v2    v3
  <chr> <chr> <chr>
1 a1    a2    a3
2 b1    NA    b3
3 c1    c2    NA
4 NA    NA    NA

paste() and paste0() convert NAs to the string "NA" before concatenating:

#> dplyr::mutate(dat, paste = paste(v1, v2, v3, sep = " | "))
# A tibble: 4 × 4
  v1    v2    v3    paste
  <chr> <chr> <chr> <chr>
1 a1    a2    a3    a1 | a2 | a3
2 b1    NA    b3    b1 | NA | b3
3 c1    c2    NA    c1 | c2 | NA
4 NA    NA    NA    NA | NA | NA

...while stringr::str_c returns NA if any value in a row is NA:

#> dplyr::mutate(dat, str_c = stringr::str_c(v1, v2, v3, sep = " | "))
# A tibble: 4 × 4
  v1    v2    v3    str_c
  <chr> <chr> <chr> <chr>
1 a1    a2    a3    a1 | a2 | a3
2 b1    NA    b3    NA
3 c1    c2    NA    NA
4 NA    NA    NA    NA

In contrast, str_c_narm() removes NAs from each row before concatenating:

#> dplyr::mutate(dat, str_c_narm = str_c_narm(v1, v2, v3, sep = " | "))
# A tibble: 4 × 4
  v1    v2    v3    str_c_narm
  <chr> <chr> <chr> <chr>
1 a1    a2    a3    "a1 | a2 | a3"
2 b1    NA    b3    "b1 | b3"
3 c1    c2    NA    "c1 | c2"
4 NA    NA    NA    ""

...with options for the case where all values in a row are NA:

#> dplyr::mutate(dat, str_c_narm = str_c_narm(v1, v2, v3, sep = " | ", if_all_na = "NA"))
# A tibble: 4 × 4
  v1    v2    v3    str_c_narm
  <chr> <chr> <chr> <chr>
1 a1    a2    a3    a1 | a2 | a3
2 b1    NA    b3    b1 | b3
3 c1    c2    NA    c1 | c2
4 NA    NA    NA    NA

See also

str_c_tidy

Examples

str_c_narm(c("a", NA), c("b", "c"), sep = "_")
#> [1] "a_b" "c"  
str_c_narm(c("a", NA), c("b", NA), if_all_na = "NA")
#> [1] "ab" NA  

# compare behavior to `paste()` and `str_c()`
dat <- tibble::tibble(v1 = c("a1", "b1", "c1", NA),
                      v2 = c("a2", NA, "c2", NA),
                      v3 = c("a3", "b3", NA, NA))

dplyr::mutate(
  dat,
  paste = paste(v1, v2, v3, sep = " | "),
  str_c = stringr::str_c(v1, v2, v3, sep = " | "),
  str_c_narm = str_c_narm(v1, v2, v3, sep = " | ")
)
#> # A tibble: 4 × 6
#>   v1    v2    v3    paste        str_c        str_c_narm    
#>   <chr> <chr> <chr> <chr>        <chr>        <chr>         
#> 1 a1    a2    a3    a1 | a2 | a3 a1 | a2 | a3 "a1 | a2 | a3"
#> 2 b1    NA    b3    b1 | NA | b3 NA           "b1 | b3"     
#> 3 c1    c2    NA    c1 | c2 | NA NA           "c1 | c2"     
#> 4 NA    NA    NA    NA | NA | NA NA           ""