findepi commented on code in PR #13147:
URL: https://github.com/apache/datafusion/pull/13147#discussion_r1826332616
##########
datafusion/physical-plan/src/joins/utils.rs:
##########
@@ -246,6 +268,11 @@ pub trait JoinHashMapType {
let next_chain = self.get_list();
for (row_idx, hash_value) in iter {
// Get the hash and find it in the index
+ if let Some(bloom_filter) = self.get_bloom_filter() {
+ if !bloom_filter.contains(hash_value) {
Review Comment:
> It needs to be pushed down below repartition / aggregate / scan etc. to be
of any benefit
That's a good point.
if the dynamic filter was range-based and we could push it down to the file
scan, it could allow file- and row-group-level pruning in Parquet.
--
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]
