Module lancedb 

LanceDB vector store implementation.

LanceDB is a serverless, low-latency vector database for AI applications. This implementation uses the LanceDB HTTP API for remote/server mode. For embedded/local mode, use the lancedb crate directly.

Example

use lc_vector_stores::lancedb::{LanceDBVectorStore, LanceDBConfig};

let config = LanceDBConfig::new("http://localhost:1337", "my_table");
let store = LanceDBVectorStore::new(config);
store.add_documents(docs, embeddings).await?;
let results = store.similarity_search(&query_embedding, 5).await?;

Structs

LanceDBConfig

LanceDB configuration.

LanceDBVectorStore

LanceDB vector store.

Functions

filter_to_sql

S3: MetadataFilter → LanceDB SQL where 子句。