gengliangwang removed a comment on issue #23383: [SPARK-23817][SQL] Migrate ORC file format read path to data source V2 URL: https://github.com/apache/spark/pull/23383#issuecomment-452660424 Since e54caaf188c4575bb6d5e4b4a93d79fce2757547, I tried to avoid changing `DataSource.providingClass` for falling back to V1 when with catalog. However, there is a lot test failures including Hive/Streaming. So I tried to pass canonical name of file format to construct the V1 data source, but that will be behavior change. Another approach is to modify CreateTableCommand, but it is too tricky and it is hard to cover every command. Hence, in d8f117e26189020a3c7f549f6918c3f9bfd51485 I use `DataSource.providingClass` again, which turn out to be a simple and great solution.
---------------------------------------------------------------- This is an automated message from the Apache Git Service. To respond to the message, please log on GitHub and use the URL above to go to the specific comment. For queries about this service, please contact Infrastructure at: [email protected] With regards, Apache Git Services --------------------------------------------------------------------- To unsubscribe, e-mail: [email protected] For additional commands, e-mail: [email protected]
