sadikovi commented on PR #36094: URL: https://github.com/apache/spark/pull/36094#issuecomment-1091045907
Can you define the risk? Lack of support in data sources is rather a limitation that could be prioritised and resolved. Is there a risk of regression in users' queries, performance implications, or backward compatibility? As far as I am concerned, TimestampNTZ is not a default type anyway and you have to explicitly opt-in to use this type which is up to users. If you try writing this type in JDBC data source, I expect an error to be thrown. Is the behaviour different from that? We can state in the release notes that support is limited to certain data sources but I would not necessarily disable it for everything. -- 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]
