Add columns using tidyverse language, but use column numbers instead of column names

Question

library(tidyverse)
df <- tibble(col1 = c(5, 2), col2 = c(6, 4), col3 = c(9, 9))
# # A tibble: 2 x 3
#    col1  col2  col3
#   <dbl> <dbl> <dbl>
# 1     5     6     9
# 2     2     4     9

I need to add columns 1 and 3. But the column names often change. So I can only use column numbers as opposed to the actual column name.

Attempt 1 works as expected.
Attempt 2 and 3 don't work.

What's wrong with my syntax? I can't use attempt 1 because next month the column names may be something else, but their relative positions will remain the same.

df %>% mutate(col4 = col1 + col3)        # attempt 1
df %>% mutate(col4 = .[, 1] + .[, 3])    # attempt 2
df %>% {mutate(col4 = .[, 1] + .[, 3])}  # attempt 3

akrun · Accepted Answer · 2019-04-25 15:54:37Z

2

If it is based on position, use rowSums by subsetting the columns based on the column index. The advantage is that we can also take care of NA elements (if any)

df %>% 
  mutate(col4 = rowSums(.[c(1, 3)], na.rm = TRUE))
# A tibble: 2 x 4
#   col1  col2  col3  col4
#  <dbl> <dbl> <dbl> <dbl>
#1     5     6     9    14
#2     2     4     9    11

Regarding the issue in OP's case, we need [[ instead of [ for subsetting a single column as a vector. With df[, 1] or .[,1] it would still be a tibble with one column instead of converting to a vector as we think about the behavior we find with data.frame

df %>% 
     mutate(col4 = .[[1]] + .[[3]])
# A tibble: 2 x 4
#   col1  col2  col3  col4
#  <dbl> <dbl> <dbl> <dbl>
#1     5     6     9    14
#2     2     4     9    11

answered Apr 25, 2019 at 15:54

akrun

891k38 gold badges590 silver badges700 bronze badges

Sign up to request clarification or add additional context in comments.

Collectives™ on Stack Overflow

Add columns using tidyverse language, but use column numbers instead of column names

1 Answer 1

Comments

Your Answer

Hot Network Questions

Collectives™ on Stack Overflow

1 Answer 1

Comments

Your Answer

Sign up or log in

Post as a guest

Related