Struct VectorJoinOp 

pub struct VectorJoinOp {

    pub input: Box<LogicalOperator>,
    pub left_vector_variable: Option<String>,
    pub left_property: Option<String>,
    pub query_vector: LogicalExpression,
    pub right_variable: String,
    pub right_property: String,
    pub right_label: Option<String>,
    pub index_name: Option<String>,
    pub k: usize,
    pub metric: Option<VectorMetric>,
    pub min_similarity: Option<f32>,
    pub max_distance: Option<f32>,
    pub score_variable: Option<String>,
}

Join graph patterns with vector similarity search.

This operator takes entities from the left input and computes vector similarity against a query vector, outputting (entity, distance) pairs.

Use Cases

1. Hybrid graph + vector queries: Find similar nodes after graph traversal
2. Aggregated embeddings: Use AVG(embeddings) as query vector
3. Filtering by similarity: Join with threshold-based filtering

Example

// Find movies similar to what the user liked
MATCH (u:User {id: $user_id})-[:LIKED]->(liked:Movie)
WITH avg(liked.embedding) AS user_taste
VECTOR JOIN (m:Movie) ON m.embedding
WHERE vector_similarity(m.embedding, user_taste) > 0.7
RETURN m.title

Fields

input: Box<LogicalOperator>

Input operator providing entities to match against.

left_vector_variable: Option<String>

Variable from input to extract vectors from (for entity-to-entity similarity). If None, uses query_vector directly.

left_property: Option<String>

Property containing the left vector (used with left_vector_variable).

query_vector: LogicalExpression

The query vector expression (constant or computed).

right_variable: String

Variable name to bind the right-side matching entities.

right_property: String

Property containing the right-side vector embeddings.

right_label: Option<String>

Optional label filter for right-side entities.

index_name: Option<String>

Name of vector index on right side (None = brute-force).

k: usize

Number of nearest neighbors per left-side entity.

metric: Option<VectorMetric>

Distance metric.

min_similarity: Option<f32>

Minimum similarity threshold.

max_distance: Option<f32>

Maximum distance threshold.

score_variable: Option<String>

Variable to bind the distance/similarity score.

Trait Implementations

impl Clone for VectorJoinOp

fn clone(&self) -> VectorJoinOp

Returns a duplicate of the value.

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

Performs copy-assignment from source.

impl Debug for VectorJoinOp

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

Formats the value using the given formatter.