Fokko commented on issue #6853: URL: https://github.com/apache/iceberg/issues/6853#issuecomment-1433594980
@Saranviveka I agree that this might be confusing, but I see Ryan's point. Querying the table can be done using equality using `cast(trans_ts as date) = date '2019-01-13'`, then the widening will be avoided. Keep in mind that this is up to the query engine how to implement this, and Spark and Trino chose to handle it the same way. > Idea of creating/specifying date based partition on timestamp is that when a corresponding date is provided without time, end user should get the result. [Iceberg Partitions](https://iceberg.apache.org/docs/latest/partitioning/) differ from your traditional Hive partitions. If you do an `hour(trans_ts)` partition then `2019-06-13 13:22:30` will be truncated into `2019-06-13 13:00:00`, and therefore you already have partitioning on a day, month, and year level. -- 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]
