Skip to content

Conversation

@Karakatiza666
Copy link
Contributor

No description provided.

@Karakatiza666
Copy link
Contributor Author

@copilot finish the implementation of percentile_cont and percetile_disc. Note that Calcite already supports these. See other implementations for aggregate functions as a reference point. To start, run the current regression test with: mvn test -Dtest=RegressionTests#issue3778, analyze and fix the error

Copy link
Contributor

Copilot AI commented Jan 20, 2026

@Karakatiza666 I've opened a new pull request, #5471, to work on those changes. Once the pull request is ready, I'll request review from you.

@mihaibudiu
Copy link
Contributor

The brute-force solution is actually not particularly difficult; just use ARRAY_AGG and run PERCENTILE for the arrays. But ARRAY_AGG is the worst possible aggregate function in terms of performance: https://docs.feldera.com/sql/aggregates/#efficient-aggregation-functions

@ryzhyk
Copy link
Contributor

ryzhyk commented Jan 21, 2026

This data structure looks like a group. This aggregate may be quasi-linear, similar to Avg.

tree: BTreeMap<T, i64>,

… functions

Signed-off-by: Karakatiza666 <bulakh.96@gmail.com>
…for insert, update, select and rank

Signed-off-by: Karakatiza666 <bulakh.96@gmail.com>
Signed-off-by: feldera-bot <feldera-bot@feldera.com>
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

5 participants