beliefer edited a comment on issue #24372: [SPARK-27462][SQL] Enhance insert into hive table that could choose some columns in target table flexibly. URL: https://github.com/apache/spark/pull/24372#issuecomment-495915028 > This requires the concept of the DEFAULT value when we create a table. Thus, we need to first change our schema definition before improving INSERT. Thanks for your review first of all. AFAIK, @wangyum will change schema definition with `DEFAULT` value and Hive3.0+ could support it. I have two suggestions about the development order between `DEFAULT` and improving INSERT. First, improving INSERT before support `DEFAULT` value which is considered after @wangyum's work. This PR will insert `NULL` value in column that not selected as target column. When Spark support `DEFAULT`, I will create another PR to replace `NULL` with `DEFAULT`. Second, waiting @wangyum to support Hive3.0+ and `DEFAULT` value, then we improving INSERT. Maybe there exists other better suggestion. Hive 1.2+ already have this syntax and Hive 3.0+ begin support `DEFAULT` value. Hive insert `NULL` value between version 1.2 and 3.0 too.
---------------------------------------------------------------- This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL above to go to the specific comment. For queries about this service, please contact Infrastructure at: [email protected] With regards, Apache Git Services --------------------------------------------------------------------- To unsubscribe, e-mail: [email protected] For additional commands, e-mail: [email protected]
