fresh-borzoni commented on issue #547: URL: https://github.com/apache/fluss-rust/issues/547#issuecomment-4611784548
@loserwang1024 +1 on the direction - on the log path, return the row as-written + schemaId + type instead of projecting to the startup schema. The conform-on-read really doesn't fit CDC. The one thing I'd push back on is having users do schema resolution + field mapping themselves. We already have that logic in the KV path, FixedSchemaDecoder + index_mapping + ProjectedRow. So, in theory, we can do exactly what you want - raw row + schemaId + type as the base, and just expose that same mapper for readers who want a stable shape. -- 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]
