alamb commented on issue #25572: URL: https://github.com/apache/datafusion/issues/25572#issuecomment-5801826254
Sorry for the late reply here -- i am finding it hard to parse through these proposals. For example > not "my pass leaves the plan invalid, who cleans up", but "my pass only finds work on an enforced plan". Think of a rule that eliminates enforced sorts by reorganizing what runs beneath them: it pattern-matches the very operators enforcement materializes, produces nothing invalid itself, and simply finds no work at all if it runs before enforcement. Maybe if you could provide a specific example or perhaps just try to make this more concise I could provide more timely feedback > An enforcement pass that finds nothing to do still pays a full-tree requirement re-derivation today, which is most of what our six passes spend (4 of 6 are byte-identical no-ops). If the enforcement phase is gated on check_invariants we could potentially do that but I don't think check_invariants actually covers all the issues that could arise. Maybe also you could summarize anything specific you would like feedback I will now look at https://github.com/apache/datafusion/pull/25585 and maybe this will make more sense after that -- 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]
