CODEWITHSUNDEEP
rdataframedplyr
jangorecki
jangorecki

Reputation: 16697

mutate tibble for particular rows specified by a numbers

Trying to find dplyr canonical way to mutate a tibble by row number.
I went through many Stack Overflow questions and GH issues, functions like row_number, if_else, case_when, but it is still not clear what is the proper way to achieve the following.

Minimal example: having a data.frame, I would like to update one of its columns to a particular value (here NA) for particular rows by providing row indices (here rows 2 and 4).
Column to update doesn't have to be parametrized, but only row numbers.
Below base R to achieve the following.

DF = data.frame(x=5:1)
idx = c(2L, 4L)
DF[idx, "x"] = NA_integer_
DF
#   x
#1  5
#2 NA
#3  3
#4 NA
#5  1

Upvotes: 1

Views: 4351

Answers (1)

Artem
Artem

Reputation: 3414

Based on @camille feedback and the consent by other commentors the most tidyverse-like code is mutate(x = ifelse(row_number() %in% idx, NA_integer_, x)).

BTW. It is better to use <- instead of = assignments, and avoid usage capital letters in the names of objects. For more information please see The tidyverse style guide

So the full code:

library(dplyr)
df <- data.frame(x = 5:1)
idx <- c(2L, 4L)
df %>% mutate(x = ifelse(row_number() %in% idx, NA_integer_, x))

#   x
#1  5
#2 NA
#3  3
#4 NA
#5  1

Upvotes: 2

Related Questions