Embabel Agent Documentation/com.embabel.agent.rag.lucene

Package-level declarations

Types

class HighDimensionVectorCodec constructor(maxDimensions: Int = DEFAULT_MAX_DIMENSIONS)

Custom codec that supports higher vector dimensions than the default 1024. This enables use of OpenAI embeddings (1536 dimensions) and other large embedding models.

LuceneSearchOperations

class LuceneSearchOperations constructor(val name: String, val enhancers: List<<Error class: unknown class>> = emptyList(), embeddingService: <Error class: unknown class>? = null, keywordExtractor: KeywordExtractor? = null, vectorWeight: Double = 0.5, chunkerConfig: <Error class: unknown class> = ContentChunker.Config(), chunkTransformer: <Error class: unknown class> = ChunkTransformer.NO_OP, indexPath: Path? = null) : Closeable

Lucene RAG facet with optional vector search support via an EmbeddingService. Supports both in-memory and disk-based persistence. Implements WritableContentElementRepository so we can add to the store.

LuceneSearchOperationsBuilder

data class LuceneSearchOperationsBuilder(name: String = "lucene-search", embeddingService: <Error class: unknown class>? = null, chunkerConfig: <Error class: unknown class> = ContentChunker.Config(), chunkTransformer: <Error class: unknown class> = ChunkTransformer.NO_OP, indexPath: Path? = null)

Builder for LuceneSearchOperations instances.

LuceneStatistics

data class LuceneStatistics(val chunkCount: Int, val documentCount: Int, val contentElementCount: Int, val averageChunkLength: Double, val hasEmbeddings: Boolean, val vectorWeight: Double, val isPersistent: Boolean, val indexPath: String?)

Statistics about the Lucene RAG service state