arrange not working with integer64 + another column

[jameelalsalam]

It appears that using arrange with an integer64 column and another column does not re-order the second column as expected. I think I can work around this with group_by + arrange, but seemed worth a report.

library(dplyr)
#> 
#> Attaching package: 'dplyr'
#> The following objects are masked from 'package:stats':
#> 
#>     filter, lag
#> The following objects are masked from 'package:base':
#> 
#>     intersect, setdiff, setequal, union

tibble(
  id = bit64::as.integer64(1),
  year = c(1, 3, 2)
) %>%
  arrange(id, year)
#> # A tibble: 3 x 2
#>        id  year
#>   <int64> <dbl>
#> 1       1     1
#> 2       1     3
#> 3       1     2

^{Created on 2022-05-12 by the reprex package (v2.0.1)}

Session info

sessioninfo::session_info()
#> - Session info ---------------------------------------------------------------
#>  setting  value
#>  version  R version 4.1.3 (2022-03-10)
#>  os       Windows 10 x64 (build 19042)
#>  system   x86_64, mingw32
#>  ui       RTerm
#>  language (EN)
#>  collate  English_United States.1252
#>  ctype    English_United States.1252
#>  tz       America/New_York
#>  date     2022-05-12
#>  pandoc   2.17.1.1 @ C:/Program Files/RStudio/bin/quarto/bin/ (via rmarkdown)
#> 
#> - Packages -------------------------------------------------------------------
#>  ! package     * version date (UTC) lib source
#>  P assertthat    0.2.1   2019-03-21 [?] CRAN (R 4.1.1)
#>  P bit           4.0.4   2020-08-04 [?] CRAN (R 4.1.1)
#>  P bit64         4.0.5   2020-08-30 [?] CRAN (R 4.1.1)
#>  P cli           3.2.0   2022-02-14 [?] CRAN (R 4.1.3)
#>  P crayon        1.5.1   2022-03-26 [?] CRAN (R 4.1.3)
#>  P DBI           1.1.2   2021-12-20 [?] CRAN (R 4.1.3)
#>  P digest        0.6.29  2021-12-01 [?] CRAN (R 4.1.3)
#>  P dplyr       * 1.0.9   2022-04-28 [?] CRAN (R 4.1.3)
#>  P ellipsis      0.3.2   2021-04-29 [?] CRAN (R 4.1.1)
#>  P evaluate      0.15    2022-02-18 [?] CRAN (R 4.1.3)
#>  P fansi         1.0.3   2022-03-24 [?] CRAN (R 4.1.3)
#>  P fastmap       1.1.0   2021-01-25 [?] CRAN (R 4.1.1)
#>  P fs            1.5.2   2021-12-08 [?] CRAN (R 4.1.3)
#>  P generics      0.1.2   2022-01-31 [?] CRAN (R 4.1.3)
#>  P glue          1.6.2   2022-02-24 [?] CRAN (R 4.1.3)
#>  P highr         0.9     2021-04-16 [?] CRAN (R 4.1.1)
#>  P htmltools     0.5.2   2021-08-25 [?] CRAN (R 4.1.1)
#>  P knitr         1.38    2022-03-25 [?] CRAN (R 4.1.3)
#>  P lifecycle     1.0.1   2021-09-24 [?] CRAN (R 4.1.1)
#>  P magrittr      2.0.3   2022-03-30 [?] CRAN (R 4.1.3)
#>  P pillar        1.7.0   2022-02-01 [?] CRAN (R 4.1.3)
#>  P pkgconfig     2.0.3   2019-09-22 [?] CRAN (R 4.1.1)
#>  P purrr         0.3.4   2020-04-17 [?] CRAN (R 4.1.1)
#>  P R6            2.5.1   2021-08-19 [?] CRAN (R 4.1.1)
#>  P reprex        2.0.1   2021-08-05 [?] CRAN (R 4.1.1)
#>  P rlang         1.0.2   2022-03-04 [?] CRAN (R 4.1.3)
#>  P rmarkdown     2.13    2022-03-10 [?] CRAN (R 4.1.3)
#>  P rstudioapi    0.13    2020-11-12 [?] CRAN (R 4.1.1)
#>  P sessioninfo   1.2.2   2021-12-06 [?] CRAN (R 4.1.3)
#>  P stringi       1.7.6   2021-11-29 [?] CRAN (R 4.1.2)
#>  P stringr       1.4.0   2019-02-10 [?] CRAN (R 4.1.1)
#>  P tibble        3.1.6   2021-11-07 [?] CRAN (R 4.1.3)
#>  P tidyselect    1.1.2   2022-02-21 [?] CRAN (R 4.1.3)
#>  P utf8          1.2.2   2021-07-24 [?] CRAN (R 4.1.1)
#>  P vctrs         0.4.1   2022-04-13 [?] CRAN (R 4.1.3)
#>  P withr         2.5.0   2022-03-03 [?] CRAN (R 4.1.3)
#>  P xfun          0.30    2022-03-02 [?] CRAN (R 4.1.3)
#>  P yaml          2.3.5   2022-02-21 [?] CRAN (R 4.1.2)
#> 
#>  [1] C:/Users/jalsal02/R/renv/library/ONGwells-7ad48f06/R-4.1/x86_64-w64-mingw32
#>  [2] C:/Program Files/R/R-4.1.3/library
#> 
#>  P -- Loaded and on-disk path mismatch.
#> 
#> ------------------------------------------------------------------------------

[DavisVaughan]

Yea, this isn't quite right

dplyr/R/arrange.R

Lines 156 to 158 in dbda0c7

	proxy <- order(vec_order(proxy,
	direction = direction,
	na_value = if(desc) "smallest" else "largest"

You lose the fact that the 3 1s are identical. i.e. it basically does this

# Lose that fact that the 3 `1`s are identical
x <- bit64::as.integer64(c(1, 1, 1))
order(vctrs::vec_order(x))
#> [1] 1 2 3

So then it thinks the first column is already sorted and doesn't look to the second one to break ties.

Luckily this will be fixed by #6263 because it removes the hack we had to do there