Extracting specific variable names into a single variable

Question

I would to pick out any variable with ## in each row then store such variables in the vars_extract variable. Any idea?

library(tidyverse)

df <- tibble(
  "a1" = c("##", 3, NA, 4, 5),
  "a2" = c(10, 38, "##", 4, 5),
  "a3" = c(11, 34, NA, 4, 5),
  "a4" = c("##", 35, 67, 4, "##"),
  "fname" = c("Megan", "John", "Terry", "Kim", "Anne")
)

# Trial (though not modified)
df %>%
  mutate(vars_extract = names(.))

I would like to something like this

for Anne, I have vars_extract as "a1, a4"

Answer 1

Something like this?

library(dplyr)
library(tidyr)

df %>%
  mutate(across(a1:a4, ~case_when(. == '##' ~ cur_column()), .names = 'new_{col}')) %>%
  unite(New_Col, starts_with('new'), na.rm = TRUE, sep = ' ')

 a1    a2       a3 a4    fname New_Col
  <chr> <chr> <dbl> <chr> <chr> <chr>  
1 ##    10       11 ##    Megan "a1 a4"
2 3     38       34 35    John  ""     
3 NA    ##       NA 67    Terry "a2"   
4 4     4         4 4     Kim   ""     
5 5     5         5 ##    Anne  "a4"