Replace last non-NA value(s) with NA [R]

I have a dataframe with 7 character columns and 1 column with integers (x).
Column x denotes the amount of non-NA values that I want to replace with NA (in this example x goes as high as 2, but could also be higher than that).

library(data.table)
d1 <- data.table(K0 = c("A", "B", "C", "D", "E", "E", "E", "F", "F", "F", "G", "G"),  
K1 = c("H", "I", "J", "J", "J", "K", "K", "K", "1111", "1111", "L", "M"),
K2 = c("N", "O", "P", "P", "P", "Q", "Q", "Q", "1112", "1112", NA, "R"),
K3 = c("S", "T", "1113", "1113", "1113", "U", "U", "U", NA, NA, NA, "1114"),
K4 = c("P", NA, "1115", "1115", "1115", "1116", "1116", "1116", NA, NA, NA, NA),
K5 = c(NA, NA, "1117", "1117", "1117", "1118", "1118", "1118", NA, NA, NA, NA),
K6 = c(NA, NA, NA, NA, NA, "1119","1119","1119", NA, NA, NA, NA),
x = c(0,0,0,1,2,0,1,2,0,1,0,0)) 

I tried a lot (what is not worth mentioning here), but I can't figure out how to solve this problem.

The desired output should look like this.

d2 <- data.table(K0 = c("A", "B", "C", "D", "E", "E", "E", "F", "F", "F", "G", "G"),  
K1 = c("H", "I", "J", "J", "J", "K", "K", "K", "1111", "1111", "L", "M"),
K2 = c("N", "O", "P", "P", "P", "Q", "Q", "Q", "1112", NA, NA, "R"),
K3 = c("S", "T", "1113", "1113", "1113", "U", "U", "U", NA, NA, NA, "1114"),
K4 = c("P", NA, "1115", "1115", NA, "1116", "1116", "1116", NA, NA, NA, NA),
K5 = c(NA, NA, "1117", NA, NA, "1118", "1118", NA, NA, NA, NA, NA),
K6 = c(NA, NA, NA, NA, NA, "1119",NA,NA, NA, NA, NA, NA),
x = c(0,0,0,1,2,0,1,2,0,1,0,0)) 

An example for clarification; row 3, 4 and 5 are duplicates and this should become

Row 3: C, J, P, 1113, 1115, 1117, NA 0 (0 = no replacement needed)

Row 4: C, J, P, 1113, 1115, NA, NA 1 (1 = last non-NA replaced with NA)

Row 5: C, J, P, 1113, NA, NA, NA 2 (2 = last 2 non-NA replaced with NA)

If some R mastermind could help me out, that would be great.
(I prefer dplyr, but at this point anything will do).
Using R version 4.1.0.

One way, not using dplyr but using apply(), would be to create a function that fills the last n non-missing values with NA, then use apply() to do this for every row:

replace_last_n <- function(x){
  # Get the last column - the number of NA's to fill
  n <- as.numeric(x[length(x)])
  
  # Remove that column
  x <- x[1:length(x)-1]
  
  # If nothing to do
  if(n == 0) return(c(x, n))

  # consider if there n is greater than the number of non-missing values
  if(n > length(x[!is.na(x)])) return(c(rep(NA, length(x)), n))
  
  # Determine number of nonmissing values
  tmp <- length(x[!is.na(x)])
  
  # fill the last n non missing values with NA
  x[!is.na(x)][(tmp - n + 1):tmp] <- NA
  
  # return entire row
  return(c(x, n))
}

Where the last value in x is the number of NA's to fill. For example:

replace_last_n(c(1,2,3,4,5, 1))
# [1]  1  2  3  4 NA  1

replace_last_n(c(1,2,3,4,NA, 1))
# [1]  1  2  3 NA NA  1

replace_last_n(c(1,2,3,NA,5, 2))
# [1]  1  2 NA NA NA  2

Then you can use apply() to use this function on every single row:

apply(d1, 1, \(x) replace_last_n(x)) %>% t() %>% as_tibble
#   K0    K1    K2    K3    K4    K5    K6    V8   
#    <chr> <chr> <chr> <chr> <chr> <chr> <chr> <chr>
#  1 A     H     N     S     P     NA    NA    0    
#  2 B     I     O     T     NA    NA    NA    0    
#  3 C     J     P     1113  1115  1117  NA    0    
#  4 D     J     P     1113  1115  NA    NA    1    
#  5 E     J     P     1113  NA    NA    NA    2    
#  6 E     K     Q     U     1116  1118  1119  0    
#  7 E     K     Q     U     1116  1118  NA    1    
#  8 F     K     Q     U     1116  NA    NA    2    
#  9 F     1111  1112  NA    NA    NA    NA    0    
# 10 F     1111  NA    NA    NA    NA    NA    1    
# 11 G     L     NA    NA    NA    NA    NA    0    
# 12 G     M     R     1114  NA    NA    NA    0  

This should work for your sample data, but not sure if the structure in your actual data would fit. The big assumption here is that there is no NA in the middle columns.

library(tidyverse)
d1 %>% 
rowwise() %>% 
mutate(string_collection = list(c(K0, K1, K2, K3, K4, K5, K6))) %>% 
mutate(column_to_keep = length(na.omit(string_collection)) - x) %>% # assume NA only found at the end
mutate(string_collection = list((string_collection)[1:column_to_keep])) %>% 
mutate(string_collection = paste0(string_collection, collapse = ",")) %>% # assume character column do not have ","
dplyr::select(x, string_collection) %>% 
separate(string_collection, into = c("K0", "K1", "K2", "K3", "K4", "K5", "K6"), sep = ",") %>% 
relocate(x, .after = last_col()) %>% # can skip the lines below
ungroup()