tustvold commented on code in PR #6049:
URL: https://github.com/apache/arrow-datafusion/pull/6049#discussion_r1179333939
##########
datafusion/core/src/physical_plan/memory.rs:
##########
@@ -223,15 +245,365 @@ impl RecordBatchStream for MemoryStream {
}
}
+/// Execution plan for writing record batches to an in-memory table.
+pub struct MemoryWriteExec {
Review Comment:
I agree that a single plan is likely not tractable, it will be highly
specific to the `TableProvider` in question. What I had meant was simply not
returning anything and leaving everything an internal implementation detail of
the `TableProvider` from how it executes the provided `ExecutionPlan` through
to how that is exposed. Not only would this make this PR significantly simpler,
obviating the need for any of the ExecutionPlan or Stream shenanigans, but
would provide the most flexibility for table implementations that have more
complex scheduling or transaction handling requirements. Exposing
`ExecutionPlan` is a pretty strong API commitment, and I _personally_ am not
confident this won't be painting ourselves into a corner, but if the consensus
is otherwise :shrug:
--
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]
