slice.() reorders columns when using .by

[markfairbanks]

All slice variants have this issue:

pacman::p_load(tidytable)

test_df <- data.table(x = 1:3, y = c("a", "a", "b"))

test_df %>% slice.(1, .by = y)
#>    y x
#> 1: a 1
#> 2: b 3

test_df %>% slice_head.(1, .by = y)
#>    y x
#> 1: a 1
#> 2: b 3

test_df %>% slice_tail.(1, .by = y)
#>    y x
#> 1: a 2
#> 2: b 3

test_df %>% slice_max.(1, .by = y)
#>    y x
#> 1: a 1

test_df %>% slice_min.(1, .by = y)
#>    y x
#> 1: a 1