[
https://issues.apache.org/jira/browse/SPARK-38869?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17531444#comment-17531444
]
Daniel commented on SPARK-38869:
--------------------------------
Thanks for pointing this out. We can skip adding new columns with default
values if the inserted relation has fewer columns than the target table. If any
of the inserted columns are explicit DEFAULT references that would have
otherwise failed analysis, we can replace them with corresponding values as
with other tables (either the provided DEFAULT values from the target schema,
or NULL otherwise).
> Respect Table capability `ACCEPT_ANY_SCHEMA` in default column resolution
> -------------------------------------------------------------------------
>
> Key: SPARK-38869
> URL: https://issues.apache.org/jira/browse/SPARK-38869
> Project: Spark
> Issue Type: Sub-task
> Components: SQL
> Affects Versions: 3.4.0
> Reporter: Gengliang Wang
> Assignee: Daniel
> Priority: Major
>
> If a V2 table has the capability of
> [ACCEPT_ANY_SCHEMA|https://github.com/apache/spark/blob/master/sql/catalyst/src/main/java/org/apache/spark/sql/connector/catalog/TableCapability.java#L94],
> we should skip adding default column values to the insert schema.
>
--
This message was sent by Atlassian Jira
(v8.20.7#820007)
---------------------------------------------------------------------
To unsubscribe, e-mail: issues-unsubscr...@spark.apache.org
For additional commands, e-mail: issues-h...@spark.apache.org
