Struct FactorizedExpandOperator 

pub struct FactorizedExpandOperator { /* private fields */ }

An expand operator that produces factorized output.

Instead of duplicating input rows for each neighbor (Cartesian product), this operator adds neighbors as a new factorization level. This avoids exponential blowup in multi-hop queries.

Memory Comparison

For a query MATCH (a)-[:KNOWS]->(b)-[:KNOWS]->(c) with:

• 100 source nodes
• Average 10 neighbors per hop

Regular Expand (flat):

• After hop 1: 100 * 10 = 1,000 rows
• After hop 2: 1,000 * 10 = 10,000 rows
• Memory: ~10,000 * row_size

Factorized Expand:

• Level 0: 100 source nodes
• Level 1: 1,000 first-hop neighbors
• Level 2: 10,000 second-hop neighbors
• Memory: ~11,100 values (no duplication)

Implementations

impl FactorizedExpandOperator

pub fn new( store: Arc<LpgStore>, input: Box<dyn Operator>, source_column: usize, direction: Direction, edge_type: Option<String>, ) -> Self

Creates a new factorized expand operator.

pub fn with_tx_context(self, epoch: EpochId, tx_id: Option<TxId>) -> Self

Sets the transaction context for MVCC visibility.

pub fn with_column_names(self, names: Vec<String>) -> Self

Sets the input column names for schema tracking.

pub fn next_factorized(&mut self) -> FactorizedResult

Gets the next factorized chunk.

This is the main method for factorized execution. For compatibility with the regular Operator trait, use next() which flattens the result.

Trait Implementations

impl FactorizedOperator for FactorizedExpandOperator

fn next_factorized(&mut self) -> FactorizedResult

Returns the next chunk as factorized data.

impl Operator for FactorizedExpandOperator

fn next(&mut self) -> OperatorResult

Pulls the next batch of data. Returns None when exhausted.

fn reset(&mut self)

Resets to initial state so you can iterate again.

fn name(&self) -> &'static str

Returns a name for debugging/explain output.