[
https://issues.apache.org/jira/browse/WICKET-1279?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Eelco Hillenius resolved WICKET-1279.
-------------------------------------
Resolution: Won't Fix
Fix Version/s: 1.3.1
Assignee: Eelco Hillenius
I've looked into it a bit, and since it is not a very common case to do this, I
don't think it is worth fixing it. Besides the alternative that I gave on the
mailing list (initialize wicket-spring in newConverterLocator), you could
alternatively override internalInit, initialize wicket-spring first, and then
call super.internalinit.
> newConverterLocater created before Spring Injection
> ---------------------------------------------------
>
> Key: WICKET-1279
> URL: https://issues.apache.org/jira/browse/WICKET-1279
> Project: Wicket
> Issue Type: Task
> Affects Versions: 1.3.0-final
> Environment: n/a
> Reporter: Karen Schaper
> Assignee: Eelco Hillenius
> Priority: Minor
> Fix For: 1.3.1
>
>
> When upgrading to wicket 1.3 I had to use a new way of registering
> converters. The issue is that the newConverterLocater method is called
> before any Spring injection takes place. I couldn't call
> InjectorHolder.getInjector().inject( this ) in my converter class.
> Eelco asked me to enter this issue and stated in the mail list that:
> "We should probably create
> the converterLocator and possibly the session store lazily instead of
> in Application#internalInit."
> My workaround was to put the
> addComponentInstantiationListener( new SpringComponentInjector( this ) );
> into the newConverterLocator which at least allows my server to start up.
> Haven't been able to test further since I have any issues now going on.
> My first JIRA entry.
> Karen
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
