[
https://issues.apache.org/jira/browse/TINKERPOP-1615?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15871893#comment-15871893
]
Daniel Kuppitz commented on TINKERPOP-1615:
-------------------------------------------
Hmm, even {{times(1000)}} works smoothly for me. [~jeromatron], I have a
suspicion that your test might have been like this.
1. You execute {{g.V().has('foo', 'id',
'4b200757-96e2-463a-a7a0-7b9b75d4dbd3').repeat(__.in('subfolder_of')).times(100).hasNext()}},
the query appears to take forever, but DSE Graph ( ? ) starts to cache all
(relevant) vertices in the background.
2. You try {{g.withoutStrategies(RepeatUnrollStrategy.class).V().has('foo',
'id',
'4b200757-96e2-463a-a7a0-7b9b75d4dbd3').repeat(__.in('subfolder_of')).times(100).hasNext()}}
and DSE Graph makes use of the cached vertices and returns the result quickly.
After that, did you try to execute it again with the strategy included?
> Doing a step many times can lead to unforeseen problems with the
> RepeatUnrollStrategy
> -------------------------------------------------------------------------------------
>
> Key: TINKERPOP-1615
> URL: https://issues.apache.org/jira/browse/TINKERPOP-1615
> Project: TinkerPop
> Issue Type: Bug
> Components: process
> Affects Versions: 3.2.3
> Reporter: Jeremy Hanna
> Assignee: Daniel Kuppitz
>
> The following traversal comes back with an error that appears to be related
> to the {{RepeatUnrollStrategy}}:
> {code}
> g.V().has('foo', 'id',
> '4b200757-96e2-463a-a7a0-7b9b75d4dbd3').repeat(__.in('subfolder_of')).times(100).hasNext()
> {code}
> appears to either error out or take forever. However if I do something like:
> {code}
> g.withoutStrategies(RepeatUnrollStrategy.class).V().has('foo', 'id',
> '4b200757-96e2-463a-a7a0-7b9b75d4dbd3').repeat(__.in('subfolder_of')).times(100).hasNext()
> {code}
> it works subsecond.
> So something funky with {{RepeatUnrollStrategy}} is going on.
--
This message was sent by Atlassian JIRA
(v6.3.15#6346)
