To adjust the code to match your requirements, you can use mutate with a conditional statement that checks if there’s an action == 'Return' within each group and uses the difference between these two times.

Here is how you could do it:

library(dplyr)

df %>% 
  mutate(
    timediffsecs = if (any(action == 'Return')) 
                       dt[action == 'Return'] - dt[action == 'Release']
                     else Sys.time() - as.POSIXct(dt),
    action = replace(action, n() > 1 & action == "Release", NA)
  )

This will calculate the difference between dt and Sys.time() for each row if there’s no action == 'Return', otherwise it calculates the difference between dt[action == 'Return'] and dt[action == 'Release'].

Also, note that in if (any(action == 'Return')), I used as.POSIXct(dt) to convert dt into POSIXct format which is required for time difference operation.

Note: The value returned for timediffsecs is in seconds.

Last modified on 2023-06-15