Struct CachedEmbeddingProvider 

pub struct CachedEmbeddingProvider { /* private fields */ }

Wraps any EmbeddingProvider with a thread-safe in-memory LRU-ish cache.

Embeddings are cached by FNV-1a hash of the input text. When the cache exceeds max_cache_size, the oldest entry (by insertion order) is evicted.

Implementations

impl CachedEmbeddingProvider

pub fn new(inner: Arc<dyn EmbeddingProvider>, max_cache_size: usize) -> Self

Wrap an existing provider with caching.

pub async fn cache_stats(&self) -> CacheStats

Returns current cache statistics.

pub async fn clear(&self)

Clears the cache and resets statistics.

Trait Implementations

impl EmbeddingProvider for CachedEmbeddingProvider

fn embed<'life0, 'life1, 'async_trait>( &'life0 self, text: &'life1 str, ) -> Pin<Box<dyn Future<Output = ArgentorResult<Vec<f32>>> + Send + 'async_trait>>

where Self: 'async_trait, 'life0: 'async_trait, 'life1: 'async_trait,

Compute embedding vector for a single text.

fn dimension(&self) -> usize

Dimension of the embedding vectors produced by this provider.

fn embed_batch<'life0, 'life1, 'life2, 'async_trait>( &'life0 self, texts: &'life1 [&'life2 str], ) -> Pin<Box<dyn Future<Output = ArgentorResult<Vec<Vec<f32>>>> + Send + 'async_trait>>

where Self: 'async_trait, 'life0: 'async_trait, 'life1: 'async_trait, 'life2: 'async_trait,

Compute embeddings for a batch of texts.