kosiew commented on PR #20485: URL: https://github.com/apache/datafusion/pull/20485#issuecomment-3951325973
@adriangb , In file scans, we can legitimately rewrite expressions between a logical file schema and a physical file schema whose field order differs. So relying on a `Column` index alone is brittle at adaptation boundaries. This change makes the adapter defensive by resolving the field via `column.name()` against the physical schema at the cast-construction point, instead of trusting `column.index()`. That avoids selecting an unrelated field when an index is stale or mismatched. -- 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]
