wypoon commented on pull request #3269: URL: https://github.com/apache/iceberg/pull/3269#issuecomment-941826001
I agree with @pvary that it would be better if Spark supported the same SQL syntax for time travel queries as Hive and Impala, especially if the syntax in question is the SQL standard. I understand the difficulty of getting the support into Spark itself. Is it not possible to implement the support using SQL extensions in Iceberg? Orthogonal to that, I think that metadata tables are conceptually different from snapshots, so while I think something like `<catalog>.<database>.<table>.history` or `<catalog>.<database>.<table>.snapshots` have a certain naturalness, `<catalog>.<database>.<table>.snapshot_id_<id>` is a different kind of beast than the metadata tables, and using the same syntax for the snapshots is lumping two different kinds of things into the same category. Pardon my ignorance, but does the code for table identifiers limit us to using `.` for delimiting the name? Is it possible to use `@` for time travel, as @pvary suggests? (I expect we'd need something like `@snapshot(<id>)` and `@time(<timestamp>)` to distinguish the two different ways of specifying time travel.) -- 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]
