Ordering variables in R alphabetically without necessarily using relocate function-CodePudding

I would like to order the variables in the below tibble so I have a1, a2, ...d2. How can I achieve that?

library(tidyverse)

dat1 <- tibble(
  b1 = c("Math", "Kisw", "Computer", "Biology"),
  a1 = c("Science", "Geog", "Studies", "Math"),
  a2 = c("Yes", "No", "Yes", "No"),
  b2 = rnorm(4, 80, 5),
  c2 = rnorm(4, 67, 5),
  b4 = rnorm(4, 98, 5),
  d2 = rnorm(4, 23, 5),
  d1 = rnorm(4, 76, 5),
  a5 = rnorm(4, 87, 5)

)

CodePudding user response：

It's not tidyverse, but

dat1[sort(names(dat1))]

will work.

If you have more than 10 sequentially numbered variable names (e.g. a1 up to a20) you might want to use gtools::mixedsort() instead of sort() so that they get sorted a1, ... a10, a11, ..., a20 rather than lexicographically (a1, a10, a11, ... , a2, a20, a3, ...) (gtools::mixedsort will also work with the select() and relocate()-based answers posted by others.)

CodePudding user response：

How about this:

dat1 %>% select(sort(colnames(.)))
# A tibble: 4 × 9
#  a1      a2       a5 b1         b2    b4    c2    d1    d2
#  <chr>   <chr> <dbl> <chr>   <dbl> <dbl> <dbl> <dbl> <dbl>
#1 Science Yes    93.3 Math     72.9 102.   67.1  83.7  24.0
#2 Geog    No     90.9 Kisw     77.4 106.   67.2  70.2  23.3
#3 Studies Yes    94.7 Comput…  79.3 109.   67.5  82.1  24.5
#4 Math    No     85.0 Biology  75.5  97.4  69.3  83.5  28.5

CodePudding user response：

You don't have to list out all the variables with relocate:

library(dplyr)
dat1 %>% 
  relocate(sort(names(.)))

# A tibble: 4 × 9
#  a1      a2       a5 b1         b2    b4    c2    d1    d2
#  <chr>   <chr> <dbl> <chr>   <dbl> <dbl> <dbl> <dbl> <dbl>
#1 Science Yes    93.3 Math     72.9 102.   67.1  83.7  24.0
#2 Geog    No     90.9 Kisw     77.4 106.   67.2  70.2  23.3
#3 Studies Yes    94.7 Comput…  79.3 109.   67.5  82.1  24.5
#4 Math    No     85.0 Biology  75.5  97.4  69.3  83.5  28.5

If you want a11 to be after a9, you can use str_sort instead of sort:

library(stringr)
library(dplyr)
dat1 %>% 
  relocate(str_sort(names(.), numeric = TRUE))

# A tibble: 4 × 9
#  a2       a5 a11     b1          b2    b4    c2    d1    d2
#  <chr> <dbl> <chr>   <chr>    <dbl> <dbl> <dbl> <dbl> <dbl>
#1 Yes    92.5 Science Math      81.7 100.   67.1  72.8  27.9
#2 No     88.0 Geog    Kisw      70.2  95.4  66.0  66.5  26.5
#3 Yes    97.2 Studies Computer  81.3  98.8  64.3  72.4  27.9
#4 No     78.6 Math    Biology   76.6 105.   62.9  68.3  23.6