Merged
Conversation
generall
approved these changes
Oct 6, 2025
timvisee
pushed a commit
that referenced
this pull request
Nov 14, 2025
Merged
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
2 participants
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
There is a use-case where a filter is to be used as a primary source, and then use vector search as fallback. With this PR, assuming cosine similarity which has output of 0-1, a query like this can be made.
{ "prefetch": [ { "filter": { "must": { "key": "description", "match": { "text": ... } } }, // full-text filter }, { "query": ..., // vector query } ], "query": { "formula": { "sum": ["$score[0]", "$score[1]"] } } }This way, results from the full-text filter will rank higher, but if the filter does not have enough results, we'll still have results from vector search.
Alternatively, something that can be done without this PR is to use the condition within the formula query again. But this method will check the same condition twice, and has a worse DX.
{ "prefetch": [ { "filter": { "must": { "key": "description", "match": { "text": ... } } }, // full-text filter }, { "query": ..., // vector query } ], "query": { "formula": { "sum": [ { "key": "description", "match": { "text": ... } }, // same condition again "$score[1]", ] } }