[
https://issues.apache.org/jira/browse/ARROW-5492?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Neal Richardson updated ARROW-5492:
-----------------------------------
Description:
read_feather, read_parquet, read_csv_arrow (and read_json, when it exists)
should take a `col_select` argument, following the model of
[vroom|http://vroom.r-lib.org/articles/vroom.html#column-selection] (readr and
base R file readers also support this feature, just much more awkwardly).
Currently, read_feather has a "columns" argument and none of the other readers
expose it. Parquet can certainly support it; cf. {{pyarrow.parquet.read_table.}}
was:This is just like like the same option in {{pyarrow.parquet.read_table}}
> [R] Add "col_select" argument to read_* functions to read subset of columns
> ----------------------------------------------------------------------------
>
> Key: ARROW-5492
> URL: https://issues.apache.org/jira/browse/ARROW-5492
> Project: Apache Arrow
> Issue Type: Improvement
> Components: R
> Reporter: Wes McKinney
> Priority: Major
> Fix For: 0.14.0
>
>
> read_feather, read_parquet, read_csv_arrow (and read_json, when it exists)
> should take a `col_select` argument, following the model of
> [vroom|http://vroom.r-lib.org/articles/vroom.html#column-selection] (readr
> and base R file readers also support this feature, just much more awkwardly).
> Currently, read_feather has a "columns" argument and none of the other
> readers expose it. Parquet can certainly support it; cf.
> {{pyarrow.parquet.read_table.}}
>
--
This message was sent by Atlassian JIRA
(v7.6.3#76005)
