[ https://issues.apache.org/jira/browse/HIVE-1309?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13637388#comment-13637388 ]
Henry Robinson commented on HIVE-1309: -------------------------------------- What was the original motivation for disallowing NULL / '' as valid partition keys? The current behaviour seems hard to reason about, e.g. the following is confusing on a recent Hive: {code} INSERT INTO TABLE part_table(part) SELECT 1, NULL FROM source_table LIMIT 1; SELECT part FROM part_table; __HIVE_DEFAULT_PARTITION__ {code} - even more so when {{part}} is an integer column. Would it be better to allow 'bad' values to map onto directory-friendly strings like {{__HIVE_NULL_PARTITION_KEY__}} etc? > let user specify "bad file" rather than HIVE_DEFAULT_PARTITION in dynamic > partition insert > ------------------------------------------------------------------------------------------ > > Key: HIVE-1309 > URL: https://issues.apache.org/jira/browse/HIVE-1309 > Project: Hive > Issue Type: New Feature > Reporter: Ning Zhang > Assignee: Ning Zhang > > Currently if the dynamic partition column value is "bad" -- null, empty > string, etc., the row will be put into the __HIVE_DEFAULT_PARTITION__ where > the bad column value will be lost (replaced by __HIVE_DEFAULT_PARTITION__) if > user select from that partition. It would be useful to put the bad record > into an file specified by the user at DML/DDL time and the user can check the > rows afterward. -- This message is automatically generated by JIRA. If you think it was sent incorrectly, please contact your JIRA administrators For more information on JIRA, see: http://www.atlassian.com/software/jira