EvaluationStrategyFactory.java

/*******************************************************************************
 * Copyright (c) 2016 Eclipse RDF4J contributors.
 *
 * All rights reserved. This program and the accompanying materials
 * are made available under the terms of the Eclipse Distribution License v1.0
 * which accompanies this distribution, and is available at
 * http://www.eclipse.org/org/documents/edl-v10.php.
 *
 * SPDX-License-Identifier: BSD-3-Clause
 *******************************************************************************/
package org.eclipse.rdf4j.query.algebra.evaluation;

import java.util.Optional;
import java.util.function.Supplier;

import org.eclipse.rdf4j.collection.factory.api.CollectionFactory;
import org.eclipse.rdf4j.query.Dataset;
import org.eclipse.rdf4j.query.algebra.evaluation.impl.EvaluationStatistics;

/**
 * Factory for {@link EvaluationStrategy}s.
 */
public interface EvaluationStrategyFactory {

	/**
	 * Set the number of query solutions the {@link EvaluationStrategy} will keep in main memory before it attempts to
	 * sync to a temporary disk cache. If set to 0, no disk caching will occur. EvaluationStrategies that provide no
	 * disk caching functionality are free to ignore this parameter.
	 *
	 * @param threshold the number of query solutions that the EvaluationStrategy can cache in main memory before
	 *                  attempting disk sync.
	 */
	@Deprecated
	void setQuerySolutionCacheThreshold(long threshold);

	/**
	 * Get the number of query solutions the {@link EvaluationStrategy} will keep in main memory before it attempts to
	 * sync to a temporary disk cache. If set to 0, no disk caching will occur. EvaluationStrategies that provide no
	 * disk caching functionality are free to ignore this parameter.
	 */
	long getQuerySolutionCacheThreshold();

	/**
	 * Set a {@link QueryOptimizerPipeline} to be used for query execution planning by the {@link EvaluationStrategy}.
	 *
	 * @param pipeline a {@link QueryOptimizerPipeline}
	 */
	void setOptimizerPipeline(QueryOptimizerPipeline pipeline);

	/**
	 * Get the {@link QueryOptimizerPipeline} that this factory will inject into the {@link EvaluationStrategy}, if any.
	 * If no {@link QueryOptimizerPipeline} is defined, the {@link EvaluationStrategy} itself determines the pipeline.
	 *
	 * @return a {@link QueryOptimizerPipeline}, or {@link Optional#empty()} if no pipeline is set on this factory.
	 */
	Optional<QueryOptimizerPipeline> getOptimizerPipeline();

	/**
	 * Returns the {@link EvaluationStrategy} to use to evaluate queries for the given {@link Dataset} and
	 * {@link TripleSource}.
	 *
	 * @param dataset              the DataSet to evaluate queries against.
	 * @param tripleSource         the TripleSource to evaluate queries against.
	 * @param evaluationStatistics the store evaluation statistics to use for query optimization.
	 * @return an EvaluationStrategy.
	 */
	EvaluationStrategy createEvaluationStrategy(Dataset dataset, TripleSource tripleSource,
			EvaluationStatistics evaluationStatistics);

	/**
	 * Returns the status of the result size tracking for the query plan. Useful to determine which parts of a query
	 * plan generated the most data.
	 *
	 * @return true if result size tracking is enabled.
	 */
	default boolean isTrackResultSize() {
		return false;
	}

	/**
	 * Enable or disable results size tracking for the query plan. Useful to determine which parts of a query plan
	 * generated the most data.
	 *
	 * @param trackResultSize true to enable tracking.
	 */
	default void setTrackResultSize(boolean trackResultSize) {
		// no-op for backwards compatibility
	}

	/**
	 * Set the collection factory to use.
	 *
	 * @param collectionFactory
	 */
	default void setCollectionFactory(Supplier<CollectionFactory> collectionFactory) {
		// no-op for backwards compatibility
	}

}