wmoustafa commented on PR #7992: URL: https://github.com/apache/iceberg/pull/7992#issuecomment-1763172619
> Also, I'm not convinced that you can/should have `field-aliases` and `field-comments` at the top-level. Some engines may support field aliases, but others do not so they don't apply equally to the representations and is inconsistent with the spec, which has them at the representation level. +1. Also worth considering whether "Representation" really refers to an engine or a dialect. Right now it looks like dialect, but even if we go with the above we will be making an implicit correlation between dialect and engine. > I think it's fine to omit the default catalog and default namespaces, especially if we can appropriately qualify them in the engine when constructing the sql representation. +1. How are we thinking about the fact that catalog names are not the same across engines, even when the underlying tables end up being looked up from the same metastore/catalog? E.g., `hive.db.t` in Trino might refer to the same table as `spark_catalog.db.t` in Spark? -- 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]
