[
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)
