returnString commented on a change in pull request #8917:
URL: https://github.com/apache/arrow/pull/8917#discussion_r543504211
##########
File path: rust/datafusion/src/datasource/datasource.rs
##########
@@ -48,9 +66,19 @@ pub trait TableProvider {
&self,
projection: &Option<Vec<usize>>,
batch_size: usize,
+ filters: &[Expr],
Review comment:
That's an interesting idea! In my use case, the TableProvider impl
interprets the filters synchronously after _asynchronously_ preparing the
necessary data externally from DataFusion, but yeah, I could see that happening
too.
If we didn't make this change, I suppose implementations could still use
these filters within the returned `ExecutionPlan`'s `execute` method for now to
enable async use cases?
Edit: my main concern here would be code churn, propagating that new
async-ness up through all the callsites.
----------------------------------------------------------------
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]
