[
https://issues.apache.org/jira/browse/TINKERPOP-2938?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Yang Xia closed TINKERPOP-2938.
-------------------------------
Fix Version/s: 3.7.0
3.6.5
Assignee: Yang Xia
Resolution: Fixed
I've added failure message assertions for appropriate cases where possible. For
mergeV(), errors in the argument map will throw during traversal construction,
instead of during execution, so the top level error output will be different at
toList(), which I think is expected.
> Revisit merge step feature tests
> --------------------------------
>
> Key: TINKERPOP-2938
> URL: https://issues.apache.org/jira/browse/TINKERPOP-2938
> Project: TinkerPop
> Issue Type: Improvement
> Components: process
> Affects Versions: 3.6.2
> Reporter: Yang Xia
> Assignee: Yang Xia
> Priority: Critical
> Fix For: 3.7.0, 3.6.5
>
>
> I've added a set of tests for
> [https://github.com/apache/tinkerpop/pull/2036,] with the CTR here
> [https://github.com/apache/tinkerpop/commit/2c9e5300d6901c72e565f89e50ab0e4844ccd56e].
>
> Ideally I want the Feature test to check that the exception fails with the
> intended error message, however, there seems to be a problem with the
> exception falling through when you have an empty graph and invalid maps to
> add just to g.V(['~label', vertex]). The intended errors appear to be null
> when you check with the step, but then gets printed after in the terminal
> output, which I wonder if it's not being caught in the toList() step, or if
> there is some race going on.
--
This message was sent by Atlassian Jira
(v8.20.10#820010)
