[jira] [Commented] (UIMA-2347) StackOverflowError by greedy TextMarker rule elements

["Peter Klügl (JIRA)" <de...@uima.apache.org>]

    [ https://issues.apache.org/jira/browse/UIMA-2347?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13431946#comment-13431946 ] 

Peter Klügl commented on UIMA-2347:
-----------------------------------

I fixed this problem now also for composed rule elements like (ANY ANY)+. However, there is a problem with additional matching alternatives during the greedy match, which cannot be performed correctly. Therefore, I introduce a new parameter "simpleGreedyForComposed" for the analysis engine. The default value is false: inference behaves as before with the possibility of an StackOverflow caused by "silly" rules. When set to true, no StackOverflow could be observed, but there is the rare possibility to miss some rule match completely. In 99.9% of the use cases, the user does not have to change anything right now.
                
> StackOverflowError by greedy TextMarker rule elements
> -----------------------------------------------------
>
>                 Key: UIMA-2347
>                 URL: https://issues.apache.org/jira/browse/UIMA-2347
>             Project: UIMA
>          Issue Type: Bug
>          Components: TextMarker
>            Reporter: Peter Klügl
>            Assignee: Peter Klügl
>
> The new implementation of the TextMarker rule inference is now based on recursion. Greedy rule elements continue their match on additional annotations therefore with a call of continueMatch(). This causes a quite big call stack for rules like ANY+ on large documents. Investigate the error. Maybe introduce a new method continueOwnMatch that is based on a loop.

--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators: https://issues.apache.org/jira/secure/ContactAdministrators!default.jspa
For more information on JIRA, see: http://www.atlassian.com/software/jira