How to order, group, mutate in data.table?-CodePudding

I am new to data.table & trying to replicate some dplyr code but not getting same results when I mutate columns.

libs

library(data.table)
library(lubridate)
library(tidyverse)

test_df <- data.frame(id = c(1234, 1234, 5678, 5678),
           date = c("2021-10-10","2021-10-10", "2021-8-10", "2021-8-15"),
           Amount  = c(54767, 96896, 34534, 79870)) %>% 
  
  mutate(date = ymd(date))

dplyr code:

test_df %>% 
  group_by(id) %>% 
  arrange(date) %>% 
  mutate(Amt_first = first(Amount),
         Amt_last = last(Amount)) %>%
  ungroup()

results:

# A tibble: 4 x 5
     id date       Amount Amt_first Amt_last
  <dbl> <date>      <dbl>     <dbl>    <dbl>
1  5678 2021-08-10  34534     34534    79870
2  5678 2021-08-15  79870     34534    79870
3  1234 2021-10-10  54767     54767    96896
4  1234 2021-10-10  96896     54767    96896

data.table attempt (returns me nothing):

setDT(test_df)[order(date),
             `:=`(Amt_first = data.table::first(Amount),
                   Amt_last = data.table::last(Amount)), 
             by = id]

I am not sure what is wrong, it seems its not selecting any columns but I as am mutating columns so ideally it should return all the columns.

CodePudding user response：

This is described in data.table's FAQ - 2.23.
You just need to add an extra [] at the end of your code:

setDT(test_df)[order(date),
             `:=`(Amt_first = data.table::first(Amount),
                   Amt_last = data.table::last(Amount)), 
             by = id][]

     id       date Amount Amt_first Amt_last
1: 1234 2021-10-10  54767     54767    96896
2: 1234 2021-10-10  96896     54767    96896
3: 5678 2021-08-10  34534     34534    79870
4: 5678 2021-08-15  79870     34534    79870