thomasrebele opened a new pull request, #6420: URL: https://github.com/apache/hive/pull/6420
[HIVE-29554](https://issues.apache.org/jira/browse/HIVE-29554) ### What changes were proposed in this pull request? Extract a common superclass MColumnStatistics from MPartitionColumnStatistics and MColumnStatistics and combine some methods that use the latter two. ### Why are the changes needed? There is a lot of code duplication around column statistics, which increases the risk of diverging code and partially applied fixes. ### Does this PR introduce _any_ user-facing change? No ### How was this patch tested? 1. I've run Hive from the master branch (13f3208c01fec2d20108302efc3fd033d1d76a19), created a table with some values, and executed `analyze table tab1 compute statistics for columns;`. I verified the statistics with `describe formatted tab1 a;`. 2. I stopped Hive, and run Hive with the changes of the PR. I checked that the statistics are shown correctly. I inserted more values into the table and executed the ANALYZE TABLE ... command. I verified the statistics. 3. I stopped Hive, and run Hive from the master branch again. I verified the statistics. I inserted more values into the table and executed the ANALYZE TABLE ... command. I verified the statistics. I started and run the DESCRIBE FORMATTED, ANALYZE TABLE, DESCRIBE FORMATTED sequence for * a metastore server with the changes from the PR, and * a hiveserver and beeline from the master branch and vice versa, i.e., * a metastore server from the master branch, and * a hiveserver and beeline with the changes from the PR All commands worked as expected. -- 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. To unsubscribe, e-mail: [email protected] For queries about this service, please contact Infrastructure at: [email protected] --------------------------------------------------------------------- To unsubscribe, e-mail: [email protected] For additional commands, e-mail: [email protected]
