syun64 commented on PR #473: URL: https://github.com/apache/iceberg-python/pull/473#issuecomment-1966411962
> Thanks @syun64 for raising this. I agree with @HonahX that setting it to `-1` is not a good idea, for mainly two reasons: > > > > - There are two options for representing saying that the `current-snapshot-id` should be ignored, both `-1` and `None`. > > - Where `-1` does not point to a valid snapshot. > > > > However Java is the reference implementation, we should not blindly follow everything that they did there. There are some things in there that we can improve on since we don't carry the full history that Java does. Can you elaborate on why you need this change? Hi @Fokko thank you for the response. Tables created by PyIceberg currently does not persist a current_snapshot_id. This means that if no further commits were made after table creation, the table cannot be read by older versions of Java code (Trino, Spark, etc) that require that the current_snapshot_id be a valid long value, until a commit is made that commits a current_snapshot_id. Otherwise, it will throw above error trace when trying to parse the table metadata. Hence I thought it was a backward incompatible change that's worth noting here -- 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
