[
https://issues.apache.org/jira/browse/WICKET-6477?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Martin Grigorov updated WICKET-6477:
------------------------------------
Fix Version/s: (was: 7.10.0)
(was: 8.0.0)
8.0.0-M9
> Component.getDefaultModelObject() wraps in RuntimeException instead of
> WicketRuntimeException
> ---------------------------------------------------------------------------------------------
>
> Key: WICKET-6477
> URL: https://issues.apache.org/jira/browse/WICKET-6477
> Project: Wicket
> Issue Type: Bug
> Components: wicket
> Affects Versions: 7.9.0
> Reporter: Per Huss
> Assignee: Martin Grigorov
> Priority: Minor
> Fix For: 8.0.0-M9
>
> Attachments: componentexception.zip
>
>
> If a model throws an exception when wicket tries to get the model value
> during rendering using Component.getDefaultModelObject() the original
> exception will be wrapped in a vanilla RuntimeException. This will then be
> wrapped again in a WicketRuntimeException. Perhaps there is a reason for this
> behaviour that I'm not aware of, or perhaps it's just accidental? I would
> expect the getDefaultModelObject() to wrap directly in a
> WicketRuntimeException, which will propagate the wicket stack properly
> without further wrapping.
> My use case is a SessionTimeoutException (originating from a timed out
> session to a third system) that depending on which path wicket has travelled
> when the exception is thrown, it is sometimes wrapped directly in a
> WicketRuntimeException and sometimes wrapped in a RuntimeException wrapped in
> a WicketRuntimeException. This can be handled by the top level exception
> handler of course, but requires more complexity and may offer a few runtime
> WTFs...
> Changing Component.getDefaultModelObject() to throw a WicketRuntimeException
> should be fairly safe since it is a RuntimeException and it will avoid the
> redundant wrapping.
> Unit test quickstart attached.
> Last: Thank you for awesomeness!
--
This message was sent by Atlassian JIRA
(v7.6.3#76005)
