Github user meiercaleb commented on a diff in the pull request:
https://github.com/apache/incubator-rya/pull/255#discussion_r160317536
--- Diff:
dao/mongodb.rya/src/main/java/org/apache/rya/mongodb/aggregation/SparqlToPipelineTransformVisitor.java
---
@@ -0,0 +1,196 @@
+/*
+ * 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 java.util.Arrays;
+
+import org.apache.rya.mongodb.StatefulMongoDBRdfConfiguration;
+import org.bson.Document;
+import org.openrdf.query.algebra.Distinct;
+import org.openrdf.query.algebra.Extension;
+import org.openrdf.query.algebra.Filter;
+import org.openrdf.query.algebra.Join;
+import org.openrdf.query.algebra.MultiProjection;
+import org.openrdf.query.algebra.Projection;
+import org.openrdf.query.algebra.Reduced;
+import org.openrdf.query.algebra.StatementPattern;
+import org.openrdf.query.algebra.helpers.QueryModelVisitorBase;
+
+import com.google.common.base.Preconditions;
+import com.mongodb.MongoClient;
+import com.mongodb.client.MongoCollection;
+import com.mongodb.client.MongoDatabase;
+
+/**
+ * Visitor that transforms a SPARQL query tree by replacing as much of the
tree
+ * as possible with one or more {@code AggregationPipelineQueryNode}s.
+ * <p>
+ * Each {@link AggregationPipelineQueryNode} contains a MongoDB aggregation
+ * pipeline which is equivalent to the replaced portion of the original
query.
+ * Evaluating this node executes the pipeline and converts the results into
+ * query solutions. If only part of the query was transformed, the
remaining
+ * query logic (higher up in the query tree) can be applied to those
+ * intermediate solutions as normal.
+ * <p>
+ * In general, processes the tree in bottom-up order: A leaf node
+ * ({@link StatementPattern}) is replaced with a pipeline that matches the
+ * corresponding statements. Then, if the parent node's semantics are
supported
+ * by the visitor, stages are appended to the pipeline and the subtree at
the
+ * parent node is replaced with the extended pipeline. This continues up
the
+ * tree until reaching a node that cannot be transformed, in which case
that
+ * node's child is now a single {@code AggregationPipelineQueryNode} (a
leaf
+ * node) instead of the previous subtree, or until the entire tree has been
+ * subsumed into a single pipeline node.
+ * <p>
+ * Nodes which are transformed into pipeline stages:
+ * <p><ul>
+ * <li>A {@code StatementPattern} node forms the beginning of each
pipeline.
+ * <li>Single-argument operations {@link Projection}, {@link
MultiProjection},
+ * {@link Extension}, {@link Distinct}, and {@link Reduced} will be
transformed
+ * into pipeline stages whenever the child {@link TupleExpr} represents a
+ * pipeline.
+ * <li>A {@link Filter} operation will be appended to the pipeline when its
+ * child {@code TupleExpr} represents a pipeline and the filter condition
is a
+ * type of {@link ValueExpr} understood by {@code
AggregationPipelineQueryNode}.
+ * <li>A {@link Join} operation will be appended to the pipeline when one
child
+ * is a {@code StatementPattern} and the other is an
+ * {@code AggregationPipelineQueryNode}.
+ * </ul>
+ */
+public class SparqlToPipelineTransformVisitor extends
QueryModelVisitorBase<Exception> {
--- End diff --
Does the AggregationPipelineQueryNode have to be evaluated first in the
query plan? Do you do anything to group nodes that can be evaluated in an
AggregationPipelineQueryNode together?
---
