pvary commented on pull request #3947: URL: https://github.com/apache/iceberg/pull/3947#issuecomment-1019864347
> @pvary, it sounds like what you're saying is that you're generating a schema, then a name mapping, and then creating a table and want the table schema to match the name mapping. I think it makes the most sense to create the table, then base the name mapping on the schema, and then add it to the table. That way you don't rely on the IDs that you pass into create table -- the IDs passed in are reassigned and we make no guarantees about them. Wouldn't this create a transiently invalid table? Also this would create an extra write to the S3 as we need to create a second metadata file. Maybe some other way to provide the mapping would be better. -- 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]
