mun1r0b0t commented on PR #11623: URL: https://github.com/apache/iceberg/pull/11623#issuecomment-2515392653
With the abstraction, it doesn't need to check the configuration for each record. I want to use the connector with ~30 tables and not having to parse them all for each record will help with the performance. It also allows saving routing specific state within the router instead of having it all in the SinkWriter, which I think is easier to manage rather than doing it all in one method. I also like that it allows users to plugin their own routing solution easily. I can think of many other ways someone might determine the destination table, and it is good functionality to have in the connector. I intend to use the plugin feature myself as well. So, overall, I think the abstraction is worth the performance and usability benefits, and it is not that much more code. Could you explain your concerns around complexity? Perhaps I can address them in other ways. -- 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: issues-unsubscr...@iceberg.apache.org For queries about this service, please contact Infrastructure at: us...@infra.apache.org --------------------------------------------------------------------- To unsubscribe, e-mail: issues-unsubscr...@iceberg.apache.org For additional commands, e-mail: issues-h...@iceberg.apache.org
