Struct MemoryCardContext 

pub struct MemoryCardContext<G = MemvidStore>
where
    G: MemoryGraph,
{ /* private fields */ }

A VectorStoreIndex that returns formatted MemoryCards instead of frame text. Generic over any MemoryGraph backend; defaults to MemvidStore so the common call site MemoryCardContext::new(store, _) continues to work without naming the type parameter.

Returned scores combine deterministic query/card relevance with recency as a tie-breaker. They are intended for stable context ordering, not as embedding similarity scores.

Implementations

impl<G> MemoryCardContext<G>

where G: MemoryGraph,

pub const DEFAULT_MAX_CARDS: usize = 8

Default cap on cards returned per query when the caller doesn’t override via Self::with_max_cards or VectorSearchRequest::samples.

pub fn new(graph: G, strategy: CardSelection) -> Self

Build a context view over graph using strategy.

pub fn with_max_cards(self, max_cards: usize) -> Self

Cap on the number of cards returned per query. The effective cap is min(max_cards, request.samples()).

pub fn graph(&self) -> &G

Borrow the underlying graph backend.

pub fn strategy(&self) -> &CardSelection

Selection strategy currently in use.

pub fn select(&self, query: &str) -> Result<Vec<MemoryCard>, G::Error>

Resolve query into the set of candidate cards according to the configured CardSelection. Public so tests and tools can inspect what the agent would see.

Trait Implementations

impl<G> Clone for MemoryCardContext<G>

where G: MemoryGraph + Clone,

fn clone(&self) -> MemoryCardContext<G>

Returns a duplicate of the value.

fn clone_from(&mut self, source: &Self)

Performs copy-assignment from source.

impl<G> Debug for MemoryCardContext<G>

where G: MemoryGraph + Debug,

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

Formats the value using the given formatter.

impl<G> VectorStoreIndex for MemoryCardContext<G>

where G: MemoryGraph + WasmCompatSend + Sync,

type Filter = MemvidFilter

Card-context selection ignores filters; using MemvidFilter keeps the type aligned with MemvidStore so callers can stack both views under a single filter type when composing dynamic_context.

async fn top_n<T>( &self, req: VectorSearchRequest<Self::Filter>, ) -> Result<Vec<(f64, String, T)>, VectorStoreError>

where T: for<'a> Deserialize<'a> + WasmCompatSend,

Returns the top N most similar documents as (score, id, document) tuples.

async fn top_n_ids( &self, req: VectorSearchRequest<Self::Filter>, ) -> Result<Vec<(f64, String)>, VectorStoreError>

Returns the top N most similar document IDs as (score, id) tuples.