blackmwk commented on issue #2478: URL: https://github.com/apache/iceberg-rust/issues/2478#issuecomment-4999580734
> Though I think it may be correct that down in the ParquetWriter we may not want to encode this logic, I do think it's pretty important that we fix this in the datafusion integration. I prefer to do it in query engine. If we do it in datafusion integration, we need to check the schema for every batch in the physical executor. But it we do it in query engine, it only needs to do it in planning phase, and insert a conversion executor node when necessary. -- 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]
