gengliangwang commented on PR #45571: URL: https://github.com/apache/spark/pull/45571#issuecomment-2005751959
@yaooqinn For parquet files, it's possible that a file directory contains two different kinds of timestamp columns. Before this change, only timestamp type can be used to read such a file directory. While for JDBC, a table type is certain. But you make a good point here. I am also considering allowing reading INT96 as TimestampNTZ, since there is no clear definition of this Parquet physical type. We can still fobid reading Int64 timestamps with `isAdjustedToUTC=true` as TimestampNTZ. -- 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]
