[
https://issues.apache.org/jira/browse/RYA-416?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16321132#comment-16321132
]
ASF GitHub Bot commented on RYA-416:
------------------------------------
Github user kchilton2 commented on a diff in the pull request:
https://github.com/apache/incubator-rya/pull/254#discussion_r160799419
--- Diff:
dao/mongodb.rya/src/main/java/org/apache/rya/mongodb/aggregation/AggregationPipelineQueryOptimizer.java
---
@@ -0,0 +1,67 @@
+/*
+ * Licensed to the Apache Software Foundation (ASF) under one
+ * or more contributor license agreements. See the NOTICE file
+ * distributed with this work for additional information
+ * regarding copyright ownership. The ASF licenses this file
+ * to you under the Apache License, Version 2.0 (the
+ * "License"); you may not use this file except in compliance
+ * with the License. You may obtain a copy of the License at
+ *
+ * http://www.apache.org/licenses/LICENSE-2.0
+ *
+ * Unless required by applicable law or agreed to in writing,
+ * software distributed under the License is distributed on an
+ * "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY
+ * KIND, either express or implied. See the License for the
+ * specific language governing permissions and limitations
+ * under the License.
+ */
+package org.apache.rya.mongodb.aggregation;
+
+import org.apache.hadoop.conf.Configurable;
+import org.apache.hadoop.conf.Configuration;
+import org.apache.rya.mongodb.StatefulMongoDBRdfConfiguration;
+import org.openrdf.query.BindingSet;
+import org.openrdf.query.Dataset;
+import org.openrdf.query.algebra.TupleExpr;
+import org.openrdf.query.algebra.evaluation.QueryOptimizer;
+import org.slf4j.Logger;
+import org.slf4j.LoggerFactory;
+
+/**
+ * MongoDB-specific query optimizer that replaces part or all of a SPARQL
query
+ * tree with a MongoDB aggregation pipeline.
+ * <p>
+ * Transforms query trees using {@link SparqlToPipelineTransformVisitor}.
If
+ * possible, this visitor will replace portions of the query tree, or the
entire
+ * query, with an equivalent aggregation pipeline (contained in an
+ * {@link AggregationPipelineQueryNode}), thereby allowing query logic to
be
+ * evaluated by the MongoDB server rather than by the client.
+ */
+public class AggregationPipelineQueryOptimizer implements QueryOptimizer,
Configurable {
+ private Configuration conf;
+ private Logger logger = LoggerFactory.getLogger(getClass());
+
+ @Override
+ public void optimize(TupleExpr tupleExpr, Dataset dataset, BindingSet
bindings) {
+ if (conf instanceof StatefulMongoDBRdfConfiguration) {
+ StatefulMongoDBRdfConfiguration mongoConf =
(StatefulMongoDBRdfConfiguration) conf;
+ SparqlToPipelineTransformVisitor pipelineVisitor = new
SparqlToPipelineTransformVisitor(mongoConf);
+ try {
+ tupleExpr.visit(pipelineVisitor);
+ } catch (Exception e) {
+ logger.error("Error attempting to transform query using
the aggregation pipeline", e);
+ }
+ }
+ }
+
+ @Override
+ public void setConf(Configuration conf) {
+ this.conf = conf;
--- End diff --
This should probably throw an IllegalArgumentException if it is provided a
conf object that is not of type StatefulMongoDBRdfConfiguration because the
code will not work if that is the case. Throwing an exception make it easier to
hunt down that bug.
> Add the ability use the MongoDB aggregation pipeline to evaluate simple
> SPARQL expressions
> ------------------------------------------------------------------------------------------
>
> Key: RYA-416
> URL: https://issues.apache.org/jira/browse/RYA-416
> Project: Rya
> Issue Type: New Feature
> Reporter: Jesse Hatfield
> Assignee: Jesse Hatfield
>
> MongoDB provides the [aggregation pipeline
> framework|https://docs.mongodb.com/manual/core/aggregation-pipeline/] for
> multi-stage data processing. Currently, the query engine invokes this
> framework to apply individual statement patterns (using a "$match" expression
> for each and iterating through the results), then applies higher-level query
> operations (join, filter, select, project, etc) client-side.
> In principle, those high-level query operations could be rewritten as
> aggregation pipeline stages as well ($group, $match, $project, etc). This
> would allow more query evaluation logic to be executed by the MongoDB server
> itself, enabling server-side optimization. This could be used as a general
> query optimization, but would additionally be useful for any tool that only
> needed to write query results back to the server: adding a write step to the
> end of the resulting pipeline could obviate the need to communicate
> individual results to the client at all.
--
This message was sent by Atlassian JIRA
(v6.4.14#64029)
