[
https://issues.apache.org/jira/browse/ARROW-6368?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16916880#comment-16916880
]
Benjamin Kietzman commented on ARROW-6368:
------------------------------------------
If the operation is potentially expensive then we might want to avoid hiding it
in the projector; in the case of augmenting as described above the columns to
augment can be generated once then reused for each yielded batch.
> [C++] Add RecordBatch projection functionality
> ----------------------------------------------
>
> Key: ARROW-6368
> URL: https://issues.apache.org/jira/browse/ARROW-6368
> Project: Apache Arrow
> Issue Type: Improvement
> Components: C++
> Reporter: Benjamin Kietzman
> Assignee: Benjamin Kietzman
> Priority: Minor
> Labels: dataset
>
> define classes RecordBatchProjector (which projects from one schema to
> another, augmenting with null/constant columns where necessary) and a subtype
> of RecordBatchIterator which projects each batch yielded by a wrapped
> iterator.
--
This message was sent by Atlassian Jira
(v8.3.2#803003)
