[ https://issues.apache.org/jira/browse/TINKERPOP-1948?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16445868#comment-16445868 ]
Daniel Kuppitz commented on TINKERPOP-1948: ------------------------------------------- Hmm, in theory, we could allow {{+}} and {{-}} for all {{TemporalAmount}} values, but this would be much uglier than what I did to support {{min()}} and {{max()}} for all comparable types. Allowing {{Comparable}} was rather easy because it's a base interface of {{Number}}, but {{TemporalAmount}} has nothing in common with {{Number}}, hence {{sum()}} would have to accept arbitrary {{Object}} s. I would prefer to have that handled by lambdas: {noformat} gremlin> d1 = java.time.Duration.ofDays(1) ==>PT24H gremlin> d2 = java.time.Duration.ofDays(2) ==>PT48H gremlin> __.inject(d1, d2).fold(java.time.Duration.ZERO, {a, b -> a + b}) ==>PT72H {noformat} Maybe we could provide predefined lambdas for some common cases..? {noformat} gremlin> __.inject(1, 2).fold(0, sum) ==>3 gremlin> __.inject(d1, d2).fold(java.time.Duration.ZERO, TimeUtil.sum) // this is made up ==>PT72H {noformat} The benefit of this is that predefined lambdas are not treated like custom lambdas (e.g. in DSE Graph you won't have to allow lambdas to use them). > Sum-step should work on properties of type 'Duration'. > ------------------------------------------------------ > > Key: TINKERPOP-1948 > URL: https://issues.apache.org/jira/browse/TINKERPOP-1948 > Project: TinkerPop > Issue Type: Improvement > Components: process > Affects Versions: 3.3.2 > Reporter: Daniel C. Weber > Priority: Minor > > Consider the following query: > {code:java} > g.inject(_P1, _P2).sum() > {code} > where _P1 and _P2 are bound to values of type 'gx:Duration' (Graphson). > This will fail with the following error message: > {code:java} > [WARN] AbstractEvalOpProcessor - Exception processing a script on request > [RequestMessage{, requestId=c3e3b3c1-a0dd-4029-b0b6-f7650f6866b3, op='eval', > processor='', args={gremlin=g.inject(_P1, _P2).sum(), bindings={_P1=PT1S, > _P2=PT2S}}}]. > java.lang.ClassCastException: java.time.Duration cannot be cast to > java.lang.Number > at > org.apache.tinkerpop.gremlin.process.traversal.step.map.SumGlobalStep.projectTraverser(SumGlobalStep.java:52) > at > org.apache.tinkerpop.gremlin.process.traversal.step.map.SumGlobalStep.projectTraverser(SumGlobalStep.java:37) > at > org.apache.tinkerpop.gremlin.process.traversal.step.util.ReducingBarrierStep.processAllStarts(ReducingBarrierStep.java:84) > at > org.apache.tinkerpop.gremlin.process.traversal.step.util.ReducingBarrierStep.processNextStart(ReducingBarrierStep.java:113) > at > org.apache.tinkerpop.gremlin.process.traversal.step.util.AbstractStep.hasNext(AbstractStep.java:143) > at > org.apache.tinkerpop.gremlin.process.traversal.util.DefaultTraversal.hasNext(DefaultTraversal.java:192) > at > org.apache.tinkerpop.gremlin.server.op.AbstractOpProcessor.handleIterator(AbstractOpProcessor.java:89) > at > org.apache.tinkerpop.gremlin.server.op.AbstractEvalOpProcessor.lambda$evalOpInternal$5(AbstractEvalOpProcessor.java:252) > at > org.apache.tinkerpop.gremlin.groovy.engine.GremlinExecutor.lambda$eval$0(GremlinExecutor.java:274) > at java.util.concurrent.FutureTask.run(FutureTask.java:266) > at > java.util.concurrent.Executors$RunnableAdapter.call(Executors.java:511) > at java.util.concurrent.FutureTask.run(FutureTask.java:266) > at > java.util.concurrent.ThreadPoolExecutor.runWorker(ThreadPoolExecutor.java:1142) > at > java.util.concurrent.ThreadPoolExecutor$Worker.run(ThreadPoolExecutor.java:617) > at java.lang.Thread.run(Thread.java:745) > {code} > This is obviously due to the fact that the Sum-steps expects its arguments to > be of type Number, which Duration obviously is not. Since summation of > Durations is easily available through [this > method|[https://docs.oracle.com/javase/8/docs/api/java/time/Duration.html#plus-java.time.Duration-]] > it's worth consideration. The feature would save us an ugly roundtrip by > saving longs since in some of our queries, we use sum(). > > -- This message was sent by Atlassian JIRA (v7.6.3#76005)