LakshSingla commented on PR #12431: URL: https://github.com/apache/druid/pull/12431#issuecomment-1102007061
Is it possible that we can do this on `RelNodes` somehow instead of `SqlNodes`? With SqlNodes, we would have to account for every shape and form of the given statement, but I am assuming that if we can work on `RelNodes` itself, Calcite will convert it into some standard form and we won't have to worry about the cases like gian mentioned in the comments. This might be tricky because there is no distinction between SELECT and INSERT query I think when we plan (because we extract out the INSERT into data source at the beginning itself). If it is not possible and we can handle everything on the SQLNodes layer itself, maybe we should start breaking out `ParsedNodes` into a separate file, and have some utils method because the `DruidPlanner` file contains a lot of implementation details (even the changes related to extracting PARTITIONED and CLUSTERED can be separated out). -- 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]
