uima-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Peter Klügl (JIRA) <...@uima.apache.org>
Subject [jira] [Commented] (UIMA-6057) Avoid falsely switching classloader
Date Tue, 02 Jul 2019 13:24:00 GMT

    [ https://issues.apache.org/jira/browse/UIMA-6057?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16876963#comment-16876963
] 

Peter Klügl commented on UIMA-6057:
-----------------------------------

Maybe I do not completely understand the description of your use case. I have many question
...

In my understanding, we have exactly 1)-4) with the problem in 4) and the patch would fix
that. Do I miss something?

Our problem is that process() switches back to the wrong class loader when it should not switch
at all. We are maybe missing the logic of the aggregate analysis engine since we are in a
primitive one. Should it not be possible to represent the classloader situation with more
than a field and an instance comparison? How would that extension break other use cases?

How can I run a pipeline with the same class loader if a PEAR is involved? How can I intentionally
share a classloader with the PEAR?

Would it be helpful if we create a failing test that reproduces our problem?



> Avoid falsely switching classloader
> -----------------------------------
>
>                 Key: UIMA-6057
>                 URL: https://issues.apache.org/jira/browse/UIMA-6057
>             Project: UIMA
>          Issue Type: Bug
>          Components: Core Java Framework
>            Reporter: Matthias Koch
>            Priority: Major
>         Attachments: UIMA-6057.diff
>
>
> In some cases the classloader is switched back, although it hasn't be switched before
processing.



--
This message was sent by Atlassian JIRA
(v7.6.3#76005)

Mime
View raw message