use function name to name new columns with across()

tidyverse
1

Is there anyway to use the function name for the glue syntax of .names in across()?

this gives you columns named mean_hp and mean_disp

library(tidyverse)
mtcars %>% summarise(across(c(hp,disp), list(mean=mean), .names="{fn}_{col}"))

it appearse that {fn} is using the name in the list instead of the function name. Id like to have the function be an parameter in rmarkdown but I cant figure how to make a named list out of it. The result of embracing the variable with curlycurly is simply 1_hp and 1_disp because it is unnamed.

centralstat = 'mean'
mtcars %>% summarise(across(c(hp,disp), {{centralstat}}, .names="{fn}_{col}"))

i'm guessing there is either a way to make a named list or some amount of rlang magic to unquote or quote it.

ideally i could do

mtcars %>% summarise(across(c(hp,disp), list({{centralstat}},sd), .names="{fn}_{col}"))

and get e.g. median_hp, median_disp,, sd_hp, sd_disp
any suggestions??

2

This topic was automatically closed 21 days after the last reply. New replies are no longer allowed.