we have following data

input <- data.frame(
  id = c(1, 1, 2, 1),
  val1 = c(NA, "world", "good", NA),
  val2 = c(NA, NA, "morning", "spark"),
  val3 = c("hello", NA, "spark", NA)
)

  id  val1    val2  val3
1  1  <NA>    <NA> hello
2  1 world    <NA>  <NA>
3  2  good morning spark
4  1  <NA>   spark  <NA>

expected output

this is the code i tried and got expected output, but it is a bit lengthy. Any alternate simple approach

# dummy list
list_outcome <- list()

# Get the columnnames to loop over

loop_var <- colnames(input)[2:4]

for(i in loop_var){
 nr <- input %>% group_by(id) %>% filter(!is.na(.data[[i]])) %>% select(id, all_of(i))
 # Save every iteration in the loop as the ith element of the list
 list_outcome[[i]] <- nr
}

# merge the elements within list by Reduce function, uses common variable name within the list on which the merge will take place 

Reduce(merge, list_outcome)