[ https://issues.apache.org/jira/browse/FLINK-1633?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14363234#comment-14363234 ]
ASF GitHub Bot commented on FLINK-1633: --------------------------------------- Github user vasia commented on a diff in the pull request: https://github.com/apache/flink/pull/452#discussion_r26487167 --- Diff: flink-staging/flink-gelly/src/main/java/org/apache/flink/graph/example/EuclideanGraphExample.java --- @@ -0,0 +1,231 @@ +/* + * 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.flink.graph.example; + +import org.apache.flink.api.common.ProgramDescription; +import org.apache.flink.api.common.functions.MapFunction; +import org.apache.flink.api.java.DataSet; +import org.apache.flink.api.java.ExecutionEnvironment; +import org.apache.flink.api.java.tuple.Tuple3; +import org.apache.flink.graph.Edge; +import org.apache.flink.graph.Graph; +import org.apache.flink.graph.Triplet; +import org.apache.flink.graph.Vertex; +import org.apache.flink.graph.example.utils.EuclideanGraphData; + +import java.io.Serializable; + +/** + * Given a directed, unweighted graph, with vertex values representing points in a plan, + * return a weighted graph where the edge weights are equal to the Euclidean distance between the + * src and the trg vertex values. + * + * <p> + * Input files are plain text files and must be formatted as follows: + * <ul> + * <li> Vertices are represented by their vertexIds and vertex values and are separated by newlines, + * the value being formed of two doubles separated by a comma. + * For example: <code>1,1.0,1.0\n2,2.0,2.0\n3,3.0,3.0\n</code> defines a data set of three vertices + * <li> Edges are represented by triples of srcVertexId, srcEdgeId, weight which are + * separated by commas. Edges themselves are separated by newlines. The initial edge value will be overwritten. + * For example: <code>1,2,13.0\n1,3,6.0\n</code> defines two edges 1-2 and 1-3 having + * weights 13 and 6 respectively. + * </ul> + * </p> + * + * Usage <code>EuclideanDistanceExample <vertex path> <edge path> <result path></code><br> + * If no parameters are provided, the program is ran with default data from + * {@link org.apache.flink.graph.example.utils.EuclideanGraphData} + */ +@SuppressWarnings("serial") +public class EuclideanGraphExample implements ProgramDescription { + + public static void main(String[] args) throws Exception { + + if (!parseParameters(args)) { + return; + } + + ExecutionEnvironment env = ExecutionEnvironment.getExecutionEnvironment(); + + DataSet<Vertex<Long, Point>> vertices = getVerticesDataSet(env); + + DataSet<Edge<Long, Double>> edges = getEdgesDataSet(env); + + Graph<Long, Point, Double> graph = Graph.fromDataSet(vertices, edges, env); + + DataSet<Triplet<Long, Point, Double>> triplets = graph.getTriplets(); + + // the edge value will be the Euclidean distance between its src and trg vertex + DataSet<Triplet<Long, Point, Double>> resultedTriplets = triplets + .map(new MapFunction<Triplet<Long, Point, Double>, Triplet<Long, Point, Double>>() { + + @Override + public Triplet<Long, Point, Double> map(Triplet<Long, Point, Double> triplet) throws Exception { + + Vertex<Long, Point> srcVertex = triplet.getSrcVertex(); + Vertex<Long, Point> trgVertex = triplet.getTrgVertex(); + Edge<Long, Double> edge = triplet.getEdge(); + + edge.setValue(srcVertex.getValue().euclideanDistance(trgVertex.getValue())); + + return new Triplet<Long, Point, Double>(srcVertex, trgVertex, edge); + } + }); + + // retrieve the edges from the final result + DataSet<Edge<Long, Double>> result = resultedTriplets + .map(new MapFunction<Triplet<Long, Point, Double>, Edge<Long, Double>>() { + + @Override + public Edge<Long, Double> map(Triplet<Long, Point, Double> triplet) throws Exception { + return triplet.getEdge(); + } + }); + + // emit result + if (fileOutput) { + result.writeAsCsv(outputPath, "\n", ","); + } else { + result.print(); + } + + env.execute("Euclidean Graph Example"); + } + + @Override + public String getDescription() { + return "Weighing a graph by computing the Euclidean distance " + + "between its vertices"; + } + + // ************************************************************************* + // DATA TYPES + // ************************************************************************* + + /** + * A simple two-dimensional point. + */ + public static class Point implements Serializable { + + public double x, y; + + public Point() {} + + public Point(double x, double y) { + this.x = x; + this.y = y; + } + + public Point add(Point other) { --- End diff -- is this used anywhere? > Add getTriplets() Gelly method > ------------------------------ > > Key: FLINK-1633 > URL: https://issues.apache.org/jira/browse/FLINK-1633 > Project: Flink > Issue Type: New Feature > Components: Gelly > Affects Versions: 0.9 > Reporter: Vasia Kalavri > Assignee: Andra Lungu > Priority: Minor > Labels: starter > > In some graph algorithms, it is required to access the graph edges together > with the vertex values of the source and target vertices. For example, > several graph weighting schemes compute some kind of similarity weights for > edges, based on the attributes of the source and target vertices. This issue > proposes adding a convenience Gelly method that generates a DataSet of > <srcVertex, Edge, TrgVertex> triplets from the input graph. -- This message was sent by Atlassian JIRA (v6.3.4#6332)