Struct AnnScan 

pub struct AnnScan<'a, E: StorageEngine> { /* private fields */ }

Approximate nearest neighbor search operator.

Uses an HNSW index to efficiently find the K nearest neighbors to a query vector. The search is approximate, trading some accuracy for speed.

Example

use manifoldb_vector::ops::{AnnScan, VectorOperator, SearchConfig};

let config = SearchConfig::k_nearest(10).with_ef_search(50);
let mut scan = AnnScan::new(&index, query, config)?;

while let Some(m) = scan.next()? {
    println!("Entity {:?} at distance {}", m.entity_id, m.distance);
}

Implementations

impl<'a, E: StorageEngine> AnnScan<'a, E>

pub fn new( index: &'a HnswIndex<E>, query: &Embedding, config: SearchConfig, ) -> Result<Self, VectorError>

Create a new ANN scan operator.

Performs the HNSW search immediately and buffers the results for iteration. This is because HNSW search is most efficient when performed as a single operation.

Arguments

• index - Reference to the HNSW index
• query - The query embedding
• config - Search configuration (k, max_distance, ef_search)

Errors

Returns an error if the query dimension doesn’t match the index.

pub fn k_nearest( index: &'a HnswIndex<E>, query: &Embedding, k: usize, ) -> Result<Self, VectorError>

Create an ANN scan with simple k-nearest configuration.

Convenience method for common case of finding K nearest neighbors.

Arguments

• index - Reference to the HNSW index
• query - The query embedding
• k - Number of nearest neighbors to find

Errors

Returns an error if the query dimension doesn’t match the index.

pub fn within_distance( index: &'a HnswIndex<E>, query: &Embedding, max_distance: f32, max_results: usize, ) -> Result<Self, VectorError>

Create an ANN scan to find vectors within a distance threshold.

Note: HNSW is optimized for k-NN search. For within-distance queries, this searches for a large k and filters by distance. Consider using ExactKnn for precise distance-based filtering on small sets.

Arguments

• index - Reference to the HNSW index
• query - The query embedding
• max_distance - Maximum distance threshold
• max_results - Maximum number of results to return

Errors

Returns an error if the query dimension doesn’t match the index.

pub const fn index(&self) -> &'a HnswIndex<E>

Get the underlying index.

pub fn len(&self) -> usize

Get the number of results found.

pub fn is_empty(&self) -> bool

Check if no results were found.

pub fn peek(&self) -> Option<&VectorMatch>

Peek at the next result without consuming it.

pub fn reset(&mut self)

Reset the iterator to the beginning.

Trait Implementations

impl<E: StorageEngine> VectorOperator for AnnScan<'_, E>

fn next(&mut self) -> Result<Option<VectorMatch>, VectorError>

Get the next match from the operator.

fn dimension(&self) -> usize

Get the dimension of vectors this operator works with.

fn collect_all(&mut self) -> Result<Vec<VectorMatch>, VectorError>

Collect all remaining matches into a vector.