An R function to work out the preceding value - r

Question

I'm trying to create a table of staff, who they report to and what level they are.
I've been working with a similar table, and #TonakShah was kind enough to help me with calculating the lowest level location is and the level above is using the solution below.
My employee table looks like this:
input = structure(list(Level.1 = structure(c(1L, 1L, 1L, 1L, 1L, 1L,
1L), .Label = "Board", class = "factor"), Level.2 = structure(c(2L,
2L, 2L, 1L, 1L, 3L, 3L), .Label = c("Aasha", "Grace", "Marisol"
), class = "factor"), Level.3 = structure(c(4L, 4L, 3L, 1L, 1L,
2L, 2L), .Label = c("Alex", "Chandler", "Millie", "Tushad"), class = "factor"),
Level.4 = structure(c(2L, 2L, 6L, 1L, 5L, 3L, 4L), .Label = c("#",
"Frank", "Joey", "Rachel", "Sarah", "Tony"), class = "factor"),
Level.5 = structure(c(3L, 2L, 1L, 1L, 1L, 4L, 1L), .Label = c("#",
"Lela", "Millie", "Ross"), class = "factor"), Level.6 = structure(c(1L,
1L, 1L, 1L, 1L, 1L, 1L), .Label = "#", class = "factor")), class = "data.frame", row.names = c(NA,
-7L))
and using the technique described here by Ronak (stackoverflow.com/questions/56903188/create-a-table-from-a-hierarchy/)
which is,
as.data.frame(t(apply(input, 1, function(x)
{new_x = x[x != "###"]; c(rev(tail(new_x, 2)), length(new_x)) })))
I can get most of the required table. But I'm having trouble trying to get "the bosses" (eg. those with employees but are not "the board").
My ideal output would look something like this (I've added colnames to make it easier to understand):
structure(list(Subordinate = structure(c(9L, 4L, 14L, 5L, 7L,
13L, 9L, 2L, 1L, 12L, 11L, 6L, 3L, 8L, 10L), .Label = c("Aasha",
"Alex", "Chandler", "Frank", "Grace", "Joey", "Lela", "Marisol",
"Millie", "Rachel", "Ross", "Sarah", "Tony", "Tushad"), class = "factor"),
Boss = structure(c(5L, 10L, 6L, 3L, 5L, 9L, 6L, 1L, 3L, 2L,
7L, 4L, 8L, 3L, 4L), .Label = c("Aasha", "Alex", "Board",
"Chandler", "Frank", "Grace", "Joey", "Marisol", "Millie",
"Tushad"), class = "factor"), Level = c(5L, 4L, 3L, 2L, 5L,
4L, 3L, 3L, 2L, 4L, 5L, 4L, 3L, 2L, 4L)), class = "data.frame", row.names = c(NA,
-15L))
I think I maybe do it with a loop, but this doesn't seem to be the best answer. Can anyone offer any other tips?

Answer 1

Couldn't come up with a prettier solution but this works. Using a while loop in the apply call used previously, we can do
output <- do.call(rbind.data.frame, apply(input, 1, function(x) {
new_x = as.character(x[x != "#"])
list_df <- list()
i = 1
while(length(new_x) >= 2) {
#Get last 2 eneteries
list_df[[i]] <- c(rev(tail(new_x, 2)), length(new_x))
#Go one level deeper
new_x = head(new_x, -1)
i = i +1
}
do.call(rbind, list_df)
}))
#To remove duplicate enteries
output[!duplicated(output), ]
# V1 V2 V3
#1 Millie Frank 5
#2 Frank Tushad 4
#3 Tushad Grace 3
#4 Grace Board 2
#5 Lela Frank 5
#9 Tony Millie 4
#10 Millie Grace 3
#12 Alex Aasha 3
#13 Aasha Board 2
#14 Sarah Alex 4
#17 Ross Joey 5
#18 Joey Chandler 4
#19 Chandler Marisol 3
#20 Marisol Board 2
#21 Rachel Chandler 4