blackmwk commented on PR #2477: URL: https://github.com/apache/iceberg-rust/pull/2477#issuecomment-4517217268
> Looked at this a little more closely. The coercion has to happen per-batch since we don't know the incoming timezone until we see data, but I can move the utility into the arrow module to keep it consistent with the int96.rs pattern. Would that address your concern? The code linked only seems to be used on the read-side? Yes, let's move it arrow module. The actually problem is the misalignment between arrow and iceberg schema. The only difference is that int96 is about schema only, but we need to convert arrow data here. This is somehow similar to dealing with schema evolution when reading from parquet. Anyway, let's make it a crate private standalone module, and we can refine the api and processing logic in future. -- 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]
