To use a case is that I wanted to return a DataFrame as an object to serve the service. The recreation service has no control over the spark context. So, is there a way that I can execute ANSI requests, for example, how I execute on registerAsTemptable.
I will pass the table names and queries to the recreation service. Then I have to return something as an object that I can show as a table in sight.
If there is an alternative way, then please suggest this as well. But I wanted to use a spark as a base structure.
source
share