shortest path QW compiles, run enron

amcp · Apr 7, 2016 · 3222088 · 3222088
1 parent fadc19f
commit 3222088
Show file tree

Hide file tree

Showing 5 changed files with 23 additions and 21 deletions.
diff --git a/.gitignore b/.gitignore
@@ -9,3 +9,5 @@ target/*
 /metrics
 /storage
 /results
+/.idea/
+/graphdb-benchmarks.iml
diff --git a/src/main/java/eu/socialsensor/graphdatabases/GraphDatabaseBase.java b/src/main/java/eu/socialsensor/graphdatabases/GraphDatabaseBase.java
@@ -137,6 +137,8 @@ public void shortestPaths(Set<Integer> nodes) {
             ctxt = shortestPathTimes.time();
             try {
                 shortestPath(from, i);
+            } catch(Exception e) {
+                e.printStackTrace();
             } finally {
                 ctxt.stop();
             }

diff --git a/src/main/java/eu/socialsensor/graphdatabases/TitanGraphDatabase.java b/src/main/java/eu/socialsensor/graphdatabases/TitanGraphDatabase.java
@@ -1,14 +1,7 @@
 package eu.socialsensor.graphdatabases;
 
 import java.io.File;
-import java.util.ArrayList;
-import java.util.HashMap;
-import java.util.HashSet;
-import java.util.Iterator;
-import java.util.LinkedList;
-import java.util.List;
-import java.util.Map;
-import java.util.Set;
+import java.util.*;
 import java.util.concurrent.TimeUnit;
 
 import org.apache.commons.configuration.Configuration;
@@ -17,6 +10,7 @@
 import org.apache.tinkerpop.gremlin.process.traversal.dsl.graph.GraphTraversal;
 import org.apache.tinkerpop.gremlin.process.traversal.dsl.graph.GraphTraversalSource;
 import org.apache.tinkerpop.gremlin.process.traversal.dsl.graph.__;
+import org.apache.tinkerpop.gremlin.process.traversal.util.FastNoSuchElementException;
 import org.apache.tinkerpop.gremlin.structure.Direction;
 import org.apache.tinkerpop.gremlin.structure.Edge;
 import org.apache.tinkerpop.gremlin.structure.Property;
@@ -288,12 +282,16 @@ public void shortestPath(final Vertex fromNode, Integer node)
     {
         final GraphTraversalSource g = graph.traversal();
         final Vertex toNode = getVertex(node);
-        //TODO(amcp) how to limit depth to 5?
-        List<Path> paths = g.V(fromNode).repeat(__.both().simplePath()).until(__.is(toNode)).limit(1).path().toList();
-
-        for(Path path : paths) {
-            path.size();
-        }
+        //            repeat the contained traversal
+        //                   map from this vertex to inV on SIMILAR edges without looping
+        //            until you map to the target toNode and the path is six vertices long or less
+        //            only return one path
+        GraphTraversal<?, Path> t =
+        g.V(fromNode).repeat(__.both().simplePath()).until(__.is(toNode).and(__.filter(it -> it.path().size() <= 6)))
+                     .limit(1).path();
+        //when the size of the path in the traverser object is six, that means this traverser made 5 hops from the
+        //fromNode, a total of 6 vertices
+        t.tryNext().ifPresent( it -> it.size());
     }
 
     @Override
@@ -624,7 +622,7 @@ public Vertex nextVertex(Iterator<Vertex> it)
     public Vertex getVertex(Integer i)
     {
         final GraphTraversalSource g = graph.traversal();
-        final Vertex vertex = g.V(T.label, NODE_LABEL).has(NODE_ID, i).next();
+        final Vertex vertex = g.V().has(NODE_ID, i).next();
         return vertex;
     }
 }
diff --git a/src/main/java/eu/socialsensor/insert/TitanMassiveCustomIds.java b/src/main/java/eu/socialsensor/insert/TitanMassiveCustomIds.java
@@ -26,7 +26,7 @@ public Vertex getOrCreate(String value)
     {
         final Long longVal = Long.valueOf(value); //the value used in data files
         //the value used in data files sometimes is zero so add one for the purposes of generating ids
-        final Long longPositiveVal = Long.valueOf(value) + 1;
+        final Long longPositiveVal = longVal + 1;
         //send everything to partition 1 by adding 1
         final long titanVertexId =
                 TitanId.toVertexId((longPositiveVal << 1) + 1 /*move over 1 bit for 2 partitions (2^1 = 2)*/);

diff --git a/src/test/resources/META-INF/input.properties b/src/test/resources/META-INF/input.properties
@@ -1,6 +1,6 @@
 # Choose which data sets you want to include in the benchmark by removing the contents.
 #Enron
-#eu.socialsensor.dataset=data/Email-Enron.txt
+eu.socialsensor.dataset=data/Email-Enron.txt
 #Amazon
 #eu.socialsensor.dataset=data/Amazon0601.txt
 #YouTube
@@ -11,7 +11,7 @@
 #eu.socialsensor.actual-communities=com-lj.all.cmty.txt
 #Synthetic
 #can change the number in the filename of the synthetic datasets to 1000, 5000, 10000, 20000, 30000, 40000, 50000
-eu.socialsensor.dataset=data/network50000.dat
+#eu.socialsensor.dataset=data/network10000.dat
 #eu.socialsensor.actual-communities=data/community50000.dat
 
 eu.socialsensor.database-storage-directory=storage
@@ -32,7 +32,7 @@ eu.socialsensor.databases=ttupl
 #eu.socialsensor.databases=tce
 #eu.socialsensor.databases=ti
 #eu.socialsensor.databases=orient
-eu.socialsensor.databases=neo4j
+#eu.socialsensor.databases=neo4j
 
 # Database specific options
 # Titan options
@@ -82,8 +82,8 @@ eu.socialsensor.benchmarks=MASSIVE_INSERTION
 #eu.socialsensor.benchmarks=SINGLE_INSERTION
 eu.socialsensor.benchmarks=FIND_NEIGHBOURS
 eu.socialsensor.benchmarks=FIND_ADJACENT_NODES
-#eu.socialsensor.benchmarks=FIND_SHORTEST_PATH
-#eu.socialsensor.shortest-path-random-nodes=100
+eu.socialsensor.benchmarks=FIND_SHORTEST_PATH
+eu.socialsensor.shortest-path-random-nodes=100
 
 # The clustering benchmark is not permutable even if eu.socialsensor.permute-benchmarks=true
 #eu.socialsensor.benchmarks=CLUSTERING