Struct SimpleVectorBackend 

pub struct SimpleVectorBackend { /* private fields */ }

Brute-force vector search backend.

Stores documents in memory and computes cosine similarity for each query. Used as a fallback when LanceDB is not available or for small collections.

Caching

Supports cache persistence via save_cache and load_cache. Use is_cache_fresh to check if a cached index is still valid.

Limitations

• O(n) search time
• All documents must fit in memory

Implementations

impl SimpleVectorBackend

pub fn new(provider: Arc<dyn EmbeddingProvider>) -> Self

Create a new empty simple vector backend.

pub fn add_documents(&mut self, documents: Vec<EmbeddedDocument>)

Add documents to the backend.

pub fn save_cache(&self, path: &Path, content_hash: &str) -> Result<()>

Save the backend’s documents to a cache file.

Stores documents and a content hash for freshness checking. Uses JSON format for simplicity and debuggability.

pub fn load_cache( path: &Path, provider: Arc<dyn EmbeddingProvider>, ) -> Result<Option<Self>>

Load a cached backend from disk.

Returns Ok(Some(backend)) if the cache exists and loaded successfully, Ok(None) if the cache doesn’t exist, or Err on read/parse errors.

pub fn is_cache_fresh(path: &Path, content_hash: &str) -> bool

Check if the cache is fresh (content hasn’t changed).

Trait Implementations

impl Debug for SimpleVectorBackend

fn fmt(&self, f: &mut Formatter<'_>) -> Result

Formats the value using the given formatter.

impl VectorBackend for SimpleVectorBackend

fn search<'life0, 'async_trait>( &'life0 self, params: VectorSearchParams, ) -> Pin<Box<dyn Future<Output = Result<VectorSearchResults>> + Send + 'async_trait>>

where Self: 'async_trait, 'life0: 'async_trait,

Execute a vector similarity search.

fn name(&self) -> &str

Get the backend name for diagnostics.

fn document_count(&self) -> Result<usize>

Get the number of indexed documents.

fn is_ready(&self) -> bool

Check if the backend is ready to handle queries.