David Snelling
dd325f2f94
Filtered semantic search now keeps its recall. A find({ query, where }) that
pairs vector search with a metadata filter restricts the HNSW beam walk to the
matching candidates INSIDE the traversal (walk-all, collect-allowed) rather than
filtering the top-k afterward — so a query whose nearest vectors are all filtered
out still returns the best matches that DO pass the filter, instead of empty.
- JS HNSW search() honors the 8.0 `allowedIds: OpaqueIdSet | ReadonlySet<string>`
contract param (ANDs with candidateIds; ignores the opaque Buffer form it can't
decode — only a native provider consumes that).
- MetadataIndexProvider gains an OPTIONAL `getIdSetForFilter(filter): OpaqueIdSet`
producer — the native (cor) metadata index returns its roaring filter result as
a serialized buffer; the JS index does not implement it.
- find() forwards the matched universe to the vector walk: the opaque buffer
(zero id materialization) when the native producer is present, alongside the
materialized visibility-precise candidateIds for the JS index. Visibility stays
correct via the existing post-search hard filter.
Tested: JS recall/restriction/AND/opaque-ignore on the index directly, plus
find() forwarding the opaque universe through to the beam walk via a stubbed
native producer.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
|
||
|---|---|---|
| .. | ||
| aggregation | ||
| brainy | ||
| db | ||
| graph | ||
| hnsw | ||
| import | ||
| indexes/columnStore | ||
| neural | ||
| performance | ||
| storage | ||
| transaction | ||
| types | ||
| universal | ||
| utils | ||
| vfs | ||
| boundary-no-native.test.ts | ||
| brainy-core.unit.test.ts | ||
| brainy-get-optimization.test.ts | ||
| create-entities-default.test.ts | ||
| fieldTypeInference.test.ts | ||
| plugin-version-coupling.test.ts | ||
| plugin.test.ts | ||
| type-filtering.unit.test.ts | ||
| vfs-multi-instance-diagnostic.test.ts | ||
| vfs-restart-fix.test.ts | ||