Subtract above number that is not NA

Question

I have a table similar to this minimal example without the difference column:

trigger	values	difference
0	3
0	NA
0	NA
1	5	2
0	4
0	NA
1	10	6

I want to subtract the above number (and leave out the NAs) from the number at each trigger point (trigger = 1)

Is there a way to do this in R?

Edit:

I have now the situation where the triggers lie close together like in this example:

trigger	values	difference
0	3
0	NA
0	NA
1	5	2
0	4
1	5	1
0	10

How can I tackle this problem?

Answer 1

Create a grouping column with cumsum on the 'trigger' and taking the lag, then do the difference between the first and last element and replace it as the last value per group

library(dplyr)
df1 %>%
    group_by(grp = lag(cumsum(trigger), default = 0)) %>% 
    mutate(difference = replace(rep(NA, n()), n(),
         values[n()] - values[1])) %>%
    ungroup %>%
    select(-grp)

-output

# A tibble: 7 × 3
  trigger values difference
    <int>  <int>      <int>
1       0      3         NA
2       0     NA         NA
3       0     NA         NA
4       1      5          2
5       0      4         NA
6       0     NA         NA
7       1     10          6

---

For the second case, we may need a condition with if/else that checks the number of rows i.e. if the number of rows is greater than 1 only need the computation to replace

df2 %>%  
  group_by(grp = lag(cumsum(trigger), default = 0)) %>% 
  mutate(difference = if(n() > 1) replace(rep(NA, n()), n(), 
         values[n()] - values[1]) else NA) %>%
  ungroup

-output

# A tibble: 7 × 4
  trigger values   grp difference
    <int>  <int> <dbl>      <int>
1       0      3     0         NA
2       0     NA     0         NA
3       0     NA     0         NA
4       1      5     0          2
5       0      4     1         NA
6       1      5     1          1
7       0     10     2         NA

data

df1 <- structure(list(trigger = c(0L, 0L, 0L, 1L, 0L, 0L, 1L), values = c(3L, 
NA, NA, 5L, 4L, NA, 10L)), class = "data.frame", row.names = c(NA, 
-7L))

df2 <- structure(list(trigger = c(0L, 0L, 0L, 1L, 0L, 1L, 0L), values = c(3L, 
NA, NA, 5L, 4L, 5L, 10L)), class = "data.frame", row.names = c(NA, 
-7L))