Getting a spell to work...

Stephen McConnell mcconnell at dpml.net
Fri Dec 3 05:46:12 EST 2004



> -----Original Message-----
> From: support-dpml-bounces at lists.ibiblio.org [mailto:support-dpml-
> bounces at lists.ibiblio.org] On Behalf Of David Leangen
> Sent: 03 December 2004 11:33
> To: DPML Support
> Subject: RE: Getting a spell to work...
> 
> 
> > This is an ant error - and I'm guessing that something in the
> > HibernateDocletTask is invoking a method to construct an implicit
> > nested <hibernate> element - and whatever that element contains -
> > it's probably referencing a class that is not included in the plugin
> > descriptor.
> 
> 
> Ok, you're right, about one thing: the classes are indeed being loaded
> correctly. So, I can conclude that the general approach is correct and
I
> don't need to declear an antlib as Niclas suggested... perhaps.

Creating an antlib will not solve your problem - but it is helpful if
you planning on automating plugin loading (but we'll cover that another
day).


> The class is indeed in the plugin descriptor, so in theory should be
> loaded
> by the classloader. So, what I suspect is that the new transit does
not
> have
> support for subtasks.
> 
> The reason I say this is because it appears that the nested
<hibernate>
> element that is reported by this error is delared as a subtask. (Ok,
maybe
> all nested elements are... I don't know off hand.). My guess is that
the
> subtask may not be getting declared.
> 
> 
> Is this possible?

No idea because I have no idea what a subtask is in this context.
Basically ant has tasks and some tasks can contain tasks - but at the
end of the day they are just tasks.

When you run with -v .... 
What does the stack trace say (the one originating in your task)?

> 
> Keep in mind that I have changed nothing from the previous version
except
> what you've told me to, and it was working before. So, I think I can
> safely assume that the problem is not with the nested elements or the
> attributes I am using.

My hunch is it may be something to do with the dependencies declared in
the index (or more specifically a dependency that is not required at
build time but shows up at runtime).

Steve.


> Regards,
> Dave
> 
> 
> _______________________________________________
> support-dpml mailing list
> support-dpml at lists.ibiblio.org
> http://lists.ibiblio.org/mailman/listinfo/support-dpml




More information about the support-dpml mailing list