[
https://issues.apache.org/jira/browse/OPENNLP-722?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14174258#comment-14174258
]
Joern Kottmann commented on OPENNLP-722:
----------------------------------------
During the indexing the features are mapped to integer ids. That is done for
example in the TwoPassDataIndexer around line 138.
The following code run differently on Java 8 compared to Java 7:
Set<String> predicateSet = new HashSet<String>();
... some code which adds String to the predicateSet ...
for (Iterator<String> pi=predicateSet.iterator();pi.hasNext();index++) {
String predicate = pi.next();
...
}
The ordering of the elements returned by the Iterator is not identical in the
Java versions.
Anyway, that shouldn't be an issue since the code still assigns one unique id
to each feature.
The "context" later on gets sorted by integer feature id. This is done in
ComparableEvent around line 39.
Now the ordering of the features inside the context is different between the
two versions.
> PerceptronPrepAttachTest fails only on Java 8
> ---------------------------------------------
>
> Key: OPENNLP-722
> URL: https://issues.apache.org/jira/browse/OPENNLP-722
> Project: OpenNLP
> Issue Type: Bug
> Components: Build, Packaging and Test
> Affects Versions: 1.6.0
> Reporter: Joern Kottmann
> Assignee: Joern Kottmann
> Priority: Minor
> Fix For: 1.6.0
>
>
> The test
> PerceptronPrepAttachTest.testPerceptronOnPrepAttachDataWithStepSizeDecrease
> fails if executed on Java 8.
> It would be really nice to track down the cause of that.
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
