Expand description
Query types and search execution
Structs§
- Bm25
Params - BM25 parameters
- Boolean
Query - Boolean query with MUST, SHOULD, and MUST_NOT clauses
- Boost
Query - Boost query - multiplies the score of the inner query
- Dense
Vector Query - Dense vector query for similarity search
- DocAddress
- Unique document address: segment_id (hex) + local doc_id within segment
- Empty
Scorer - Empty scorer for terms that don’t exist
- Global
Stats - Global statistics aggregated across all segments
- Global
Stats Builder - Builder for aggregating statistics from multiple segments
- Global
Stats Cache - Cached global statistics with automatic invalidation
- Heap
Entry - Entry for top-k min-heap
- Matched
Field - Matched field info with ordinals (for multi-valued fields)
- Phrase
Query - Phrase query - matches documents containing terms in consecutive positions
- Score
Collector - Efficient top-k collector using min-heap
- Scored
Doc - Search result from WAND execution
- Search
Hit - Search hit with unique document address and score
- Search
Response - Search response with hits (IDs only, no documents)
- Search
Result - Search result with doc_id and score (internal use)
- Sparse
Field Stats - Statistics for a sparse vector field
- Sparse
Term Scorer - Scorer for sparse vector dimensions
- Sparse
Vector Query - Sparse vector query for similarity search
- Term
Query - Term query - matches documents containing a specific term
- Term
Query Info - Info for WAND-optimizable term queries
- Text
Field Stats - Statistics for a full-text field
- Text
Term Scorer - Scorer for full-text terms using WAND optimization
- TopK
Collector - Collector for top-k results
- Wand
Executor - Generic MaxScore WAND executor for top-k retrieval
- Wand
OrQuery - WAND-optimized OR query for multiple terms
Enums§
- Multi
Value Combiner - Strategy for combining scores when a document has multiple values for the same field
Constants§
- BM25_B
- BM25 b parameter - controls length normalization 0 = no length normalization, 1 = full normalization
- BM25_K1
- BM25 k1 parameter - controls term frequency saturation Higher values give more weight to term frequency
Traits§
- Query
- A search query (async)
- Scorer
- Scorer that iterates over matching documents and computes scores
- Scoring
Iterator - Common interface for scoring iterators (text terms or sparse dimensions)
Functions§
- bm25_
idf - Compute IDF (Inverse Document Frequency) using BM25 variant
- bm25_
score - Compute BM25 score for a term occurrence
- bm25_
upper_ bound - Compute BM25 upper bound score for WAND pruning
- bm25f_
score - Compute BM25F score with field boost
- bm25f_
upper_ bound - Compute BM25F upper bound score for WAND pruning with field boost
- search_
segment - Execute a search query on a single segment (async)
- search_
segment_ with_ positions - Execute a search query on a single segment with optional position collection (async)
Type Aliases§
- Count
Future - Future type for count estimation
- Matched
Positions - Matched positions for a field (field_id, list of encoded positions)
- Scorer
Future - Future type for scorer creation