alamb commented on code in PR #14138:
URL: https://github.com/apache/datafusion/pull/14138#discussion_r1918227595
##########
datafusion/physical-expr/src/equivalence/ordering.rs:
##########
@@ -156,6 +156,26 @@ impl OrderingEquivalenceClass {
}
}
+ /// Trims `orderings[idx]` if some suffix of it overlaps with a prefix of
+ /// `orderings[pre_idx]`. Returns `true` if there is any overlap, `false`
otherwise.
+ ///
+ /// For example, if `orderings[idx]` is `[a ASC, b ASC, c DESC]` and
+ /// `orderings[pre_idx]` is `[b ASC, c DESC]`, then the function will trim
+ /// `orderings[idx]` to `[a ASC]`.
+ fn resolve_overlap(&mut self, idx: usize, pre_idx: usize) -> bool {
+ let length = self.orderings[idx].len();
+ let other_length = self.orderings[pre_idx].len();
+ for overlap in 1..=length.min(other_length) {
Review Comment:
FWIW I think this is a very hot path in the particular queries. as it is
something like O(N^3) in the number of potential sort orders. I am still
investigating
--
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]
