Dandandan commented on issue #251: URL: https://github.com/apache/arrow-datafusion/issues/251#issuecomment-833793043
Just some design input here, IMO it is best to not generate the timestamp at the moment of the first invocation when evaluating the query, but "just" always at the start of the query. An important optimization is to evaluate / replace the timestamp at planning time as a optimization rule, so it can be used to prune partitions and can be used for predicate pushdown. -- 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. For queries about this service, please contact Infrastructure at: [email protected]
