You are viewing a plain text version of this content. The canonical link for it is here.
Posted to jira@arrow.apache.org by "SHIMA Tatsuya (Jira)" <ji...@apache.org> on 2022/09/15 13:00:03 UTC

[jira] [Updated] (ARROW-17737) [R] Continue to retain grouping metadata even if ungroup arrow dplyr query

     [ https://issues.apache.org/jira/browse/ARROW-17737?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]

SHIMA Tatsuya updated ARROW-17737:
----------------------------------
    Description: 
Perhaps {{metadata$r$attributes$.group_vars}} needs to be removed when it becomes arrow dplyr query.
And it must also be written back again when converted to Table.

{code:r}
mtcars |> dplyr::group_by(cyl) |> arrow::arrow_table() |> dplyr::ungroup() |> as.data.frame() |> dplyr::group_vars()
#> character(0)
mtcars |> dplyr::group_by(cyl) |> arrow:::as_adq() |> dplyr::ungroup() |> as.data.frame() |> dplyr::group_vars()
#> [1] "cyl"
{code}

  was:
Perhaps {{metadata$r$attributes$.group_vars}} needs to be removed when it becomes arrow dplyr query.

{code:r}
mtcars |> dplyr::group_by(cyl) |> arrow::arrow_table() |> dplyr::ungroup() |> as.data.frame() |> dplyr::group_vars()
#> character(0)
mtcars |> dplyr::group_by(cyl) |> arrow:::as_adq() |> dplyr::ungroup() |> as.data.frame() |> dplyr::group_vars()
#> [1] "cyl"
{code}


> [R] Continue to retain grouping metadata even if ungroup arrow dplyr query
> --------------------------------------------------------------------------
>
>                 Key: ARROW-17737
>                 URL: https://issues.apache.org/jira/browse/ARROW-17737
>             Project: Apache Arrow
>          Issue Type: Improvement
>          Components: R
>    Affects Versions: 9.0.0
>            Reporter: SHIMA Tatsuya
>            Priority: Major
>
> Perhaps {{metadata$r$attributes$.group_vars}} needs to be removed when it becomes arrow dplyr query.
> And it must also be written back again when converted to Table.
> {code:r}
> mtcars |> dplyr::group_by(cyl) |> arrow::arrow_table() |> dplyr::ungroup() |> as.data.frame() |> dplyr::group_vars()
> #> character(0)
> mtcars |> dplyr::group_by(cyl) |> arrow:::as_adq() |> dplyr::ungroup() |> as.data.frame() |> dplyr::group_vars()
> #> [1] "cyl"
> {code}



--
This message was sent by Atlassian Jira
(v8.20.10#820010)