alamb commented on issue #970: URL: https://github.com/apache/arrow-datafusion/issues/970#issuecomment-1793411154
> Would this be a sensible approach to take? The basic idea seems reasonable to me. Given your description it seems like the same thing could be accomplished today in DataFusion by implementing and registering the appropriate Logical [OptimizerRule](https://docs.rs/datafusion/latest/datafusion/optimizer/trait.OptimizerRule.html#) or ExecutionPlan [PhysicalOptimizerRule](https://docs.rs/datafusion/latest/datafusion/physical_optimizer/optimizer/trait.PhysicalOptimizerRule.html#) that would walk the plans, identify subtrees that they knew how to push down, and rewrite them to use specialized `ExecutionPlan`s DataFusion is already setup to run arbitrary ExecutionPlans / Sources / Passes. I wonder if there is anything specific we need to add 🤔 Perhaps an example and improve the docs would suffice -- 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]
