The server side implementation class that gets generated has no dependence on Axis or SOAP. It is meant to contain your business implementation of the function. The skeleton class is Axis aware and calls the implementation class. So the implementation class can be tested without any dependence on Axis or SOAP or HTTP.
If you already have an implementation class, then it may be of no use (the skeleton class could be made to call the existing implementation) but it could also be used to modify the interface of an existing implementation of the function you're trying to publish via a web service. For example, if the output of the existing service provides too much data, this could be cut down by the implementation class that is called by the skeleton.
Hope this helps.
Tony
"Carlos Valiente" <[EMAIL PROTECTED]> wrote on 26/01/2005 17:27:52:
> Hi! Just a silly question: What's the benefit of generating the
> server-side skeleton class for a service and deploying it instead of
> the implementation class? Apart from having a cleaner deploy.wsdd
> descriptor, I cannot see much gain
>
> Cheers,
>
> Carlos
>
- Server-side skeletons - What for? Carlos Valiente
- tony . q . weddle
