Struct QueryPlanner 

pub struct QueryPlanner { /* private fields */ }

Query planner that converts Query into optimized PhysicalPlan

The planner applies predicate pushdown, filter merging, and cost-based selection to produce an execution plan that minimises expensive FHE operations and I/O.

Optionally maintains a plan cache to avoid re-planning identical queries.

Implementations

impl QueryPlanner

pub fn new() -> Self

Create a new query planner with default statistics

pub fn with_stats(stats: Arc<PlannerStats>) -> Self

Create a planner with custom statistics

pub fn with_cache(self, config: PlanCacheConfig) -> Self

Enable plan caching with the given configuration

pub fn stats(&self) -> &PlannerStats

Get a reference to the planner statistics

pub fn plan_cache(&self) -> Option<&PlanCache>

Get a reference to the plan cache, if enabled

pub fn cache_stats(&self) -> CacheStats

Return cache statistics, or default stats if caching is not enabled

pub fn invalidate_all(&self)

Invalidate all cached plans (e.g., after a schema change)

pub fn invalidate_prefix(&self, prefix: &str)

Invalidate cached plans matching a prefix (e.g., a collection name)

pub fn plan(&self, query: &Query) -> Result<PhysicalPlan>

Plan a query

If caching is enabled, checks the cache first and returns a cached plan if one exists and has not expired. Otherwise, plans the query from scratch and inserts the result into the cache.

pub fn estimate_cost(&self, plan: &PhysicalPlan) -> PlanCost

Estimate the cost of a physical plan

pub fn choose_cheaper<'a>( &self, a: &'a PhysicalPlan, b: &'a PhysicalPlan, ) -> &'a PhysicalPlan

Compare two physical plans by cost and return the cheaper one

Trait Implementations

impl Default for QueryPlanner

fn default() -> Self

Returns the “default value” for a type.