Column value incremental update

Fellow R-Stackoverflowers,

I have a data table with 2 columns and I'm trying to calculate a new third column based on both the 2 existing column values and the new column calculated value for the previous row.

I have been checking the forums and I have tried a couple of answers but I don't get it right. I hope you can help me.

Here is a reproducible example:

error <- c(1,1,0,0,0,1,1,1,1,0)
trigger <- c(FALSE, TRUE, FALSE, FALSE, FALSE, FALSE, TRUE, TRUE, TRUE, FALSE)
expected <- c(1,2,0,0,0,1,2,3,4,0)
DTtest <- data.table(error, trigger, expected)
DTtest
     error trigger expected 
 1:     1   FALSE        1     
 2:     1    TRUE        2     
 3:     0   FALSE        0     
 4:     0   FALSE        0     
 5:     0   FALSE        0     
 6:     1   FALSE        1     
 7:     1    TRUE        2     
 8:     1    TRUE        3     
 9:     1    TRUE        4     
10:     0   FALSE        0     

The "expected" column includes the values I expect to calculate with the "error" and "trigger" columns. The formula I would like to apply would be the following:

if(trigger) {
    new_column = new_column(previous_row) + 1
} else {
    new_column = error
}

My first try was to use an ifelse directly to update the new column. I found that I had to actually initialize the new column for it to run:

DTtest <- DTtest[, impact:=0]
DTtest[, impact:=ifelse(trigger, lag(impact)+1, error)]

This option does calculate the new column "impact" but the results do not match the expected value I got calculating the column in Excel (something I can't do with the whole data table, as it's pretty big):

    error trigger expected impact
 1:     1   FALSE        1      1
 2:     1    TRUE        2      1
 3:     0   FALSE        0      0
 4:     0   FALSE        0      0
 5:     0   FALSE        0      0
 6:     1   FALSE        1      1
 7:     1    TRUE        2      1
 8:     1    TRUE        3      1
 9:     1    TRUE        4      1
10:     0   FALSE        0      0

Then I tried a for loop but the results are not correct either:

for(index in nrow(DTtest)){
  imp <- 0
  if(index==1){
    imp <- DTtest[index]$error
  } else {
    imp <- DTtest[index-1]$impact+1
  }
  set(DTtest, i=index, j=as.integer(4), value=imp )
}

I have the feeling that lag(impact) does not get the updated value for some reason, but I can't fancy why.

Would you please help me? .

Thank you!