punAhuja opened a new pull request, #5119: URL: https://github.com/apache/solr/pull/5119
https://issues.apache.org/jira/browse/SOLR-18542 <!-- _(If you are a project committer then you may remove some/all of the following template.)_ Before creating a pull request, please file an issue in the ASF Jira system for Solr: * https://issues.apache.org/jira/projects/SOLR For something minor (i.e. that wouldn't be worth putting in release notes), you can skip JIRA. To create a Jira issue, you will need to create an account there first. The title of the PR should reference the Jira issue number in the form: * SOLR-####: <short description of problem or changes> SOLR must be fully capitalized. A short description helps people scanning pull requests for items they can work on. Properly referencing the issue in the title ensures that Jira is correctly updated with code review comments and commits. --> # Description Combined Query RRF gives every query list equal weight, so a caller can't favor one list, e.g. keep an exact lexical match above a loosely similar vector match. Elasticsearch and OpenSearch both support per-list weights. Also, `ReciprocalRankFusion.getRanks` iterates `queriesDocMap.values()` (a `HashMap`) but labels the rank array with `combiner.query` keys in request order, so the debug explain can mislabel which rank belongs to which query when the two orders differ. # Solution Please provide a short description of the approach taken to implement your solution. * Add `combiner.rrf.weight.<queryName>`, a non-negative multiplier on that query's reciprocal rank contribution, defaulting to `1.0`. Requests without it score exactly as before. * Reject negative weights, and weights for names that aren't in `combiner.query`, with a 400. * Make `getRanks` iterate in `combiner.query` order, so each rank and weight in the explain lines up with the right query. The explain prints `weight/(k+rank)` when a weight is set. * Update the ref guide: new param row, a weighted example, and an entry in major-changes-in-solr-11. AI disclosure: parts of this change were developed with the help of an AI coding assistant. I reviewed and tested everything and take full responsibility for it. # Tests * `ReciprocalRankFusionTest`: no-weight order unchanged, a higher weight flips the fused order, an omitted weight defaults to 1, a zero weight contributes nothing, negative weight throws, unknown weight name throws, and the explain carries the weight with the right query label. * `DistributedCombinedQueryComponentTest`: two queries with fixed scores where weights 1 and 3 flip the winner. # Checklist Please review the following and check all that apply: - [ ] I have reviewed the guidelines for [How to Contribute](https://github.com/apache/solr/blob/main/CONTRIBUTING.md) and my code conforms to the standards described there to the best of my ability. - [ ] I have created a Jira issue and added the issue ID to my pull request title. - [ ] I have given Solr maintainers [access](https://help.github.com/en/articles/allowing-changes-to-a-pull-request-branch-created-from-a-fork) to contribute to my PR branch. (optional but recommended, not available for branches on forks living under an organisation) - [ ] I have developed this patch against the `main` branch. - [ ] I have run `./gradlew check`. - [ ] I have added tests for my changes. - [ ] I have added documentation for the [Reference Guide](https://github.com/apache/solr/tree/main/solr/solr-ref-guide) - [ ] I have added a [changelog entry](https://github.com/apache/solr/blob/main/dev-docs/changelog.adoc) for my change -- 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]
