用 R 中的唯一字符值填充 NA

Question

I'm trying to get unique ID for NAs' in my data. Example will clears my idea.

library(dplyr)
tbl <-  tibble(ID = c(rep("A", 3), rep("B", 3)),
               SecondID = c("AAA", NA, NA, "BBB", "BBB", NA),
               ThirdID = c("CCC", NA, NA, "DDD", "DDD", NA))

I need unique values for Second and Third ID. Here is how solved it, but now I'm just waiting crisis to happen when sample will give me exactly same number twice for certain group.

tbl %>% 
  rowwise() %>% 
  mutate(SecondID = if_else(is.na(SecondID), paste0(ID, ".", sample(1:100, 1, replace=FALSE)), SecondID)) %>% 
  ungroup() %>% 
  mutate(ThirdID = if_else(is.na(ThirdID), paste0(SecondID, ".1"), ThirdID))

  ID    SecondID ThirdID
  <chr> <chr>    <chr>  
1 A     AAA      CCC    
2 A     A.54     A.54.1 
3 A     A.65     A.65.1 
4 B     BBB      DDD    
5 B     BBB      DDD    
6 B     B.8      B.8.1  

Is there a foolproof method creating IDs'?

Answer 1

Get all the values from sample together not in rowwise fashion so that it is guaranteed that you get different numbers.

library(dplyr)


tbl %>%
  mutate(SecondID = replace(SecondID, is.na(SecondID), 
   paste(ID[is.na(SecondID)], sample(1:100, sum(is.na(SecondID))), sep = ".")), 
         ThirdID = ifelse(is.na(ThirdID), paste0(SecondID, '.1'), ThirdID))

# ID    SecondID ThirdID
#  <chr> <chr>    <chr>  
#1 A     AAA      CCC    
#2 A     A.29     A.29.1 
#3 A     A.13     A.13.1 
#4 B     BBB      DDD    
#5 B     BBB      DDD    
#6 B     B.22     B.22.1