Creating ID based in two columns string and value and increase after change conditions in r (Hard)-CodePudding

How to create a new column so that a.) when the userID reaches the value one, the next row changes the ID or b.) the next row changes the userID.

For example:

df_base <- structure(list(
  userid    =  c("5465d","5465d","5465d","5465d","6765b","6765b","7995b","7995b","7988b","7778b"), 
  sells     =  c(1, 0, 0, 1,0,0,0,1,0,0)), 
  .Names=c("userid", "sells"), 
  row.names = c("1", "2", "3", "4","5","6","7","8","9","10"), class =("data.frame"))

After creating column

df_base_result <- structure(list(
userid    =  c("5465d","5465d","5465d","5465d","6765b","6765b","7995b","7995b","7988b","7778b"), 
sells     =  c(1, 0, 0, 1,0,0,0,1,0,0),
ids     =  c("ID1","ID2","ID2","ID2","ID3","ID3","ID4","ID4","ID5","ID6")),
.Names=c("userid", "sells","ids"), 
row.names = c("1", "2", "3", "4","5","6","7","8","9","10"), class =("data.frame"))

CodePudding user response：

df_base %>%
  group_by(userid) %>%
  mutate(a = cumsum(lag(sells,default = 0)))%>%
  ungroup() %>%
  mutate(ids = paste0('ID',data.table::rleid(userid, a))) %>%
  select(-a)
         
# A tibble: 10 x 3
   userid sells ids  
   <chr>  <dbl> <chr>
 1 5465d      1 ID1  
 2 5465d      0 ID2  
 3 5465d      0 ID2  
 4 5465d      1 ID2  
 5 6765b      0 ID3  
 6 6765b      0 ID3  
 7 7995b      0 ID4  
 8 7995b      1 ID4  
 9 7988b      0 ID5  
10 7778b      0 ID6

CodePudding user response：

You may try

x <- 1
res <- c(x)

for (i in 2:nrow(df_base)){
  if(df_base$userid[i] != df_base$userid[i-1] | (df_base$sells[i] == 0 && df_base$sells[i-1] ==1)){
    x <- x   1
    res <- c(res, x)
  } else {
    x <- x
    res <- c(res, x)
  }
}

df_base$ids <- paste0("ID",res)
df_base

   userid sells ids
1   5465d     1 ID1
2   5465d     0 ID2
3   5465d     0 ID2
4   5465d     1 ID2
5   6765b     0 ID3
6   6765b     0 ID3
7   7995b     0 ID4
8   7995b     1 ID4
9   7988b     0 ID5
10  7778b     0 ID6