I have used custom soap processors in most of my projects. Apart from the well known advantages they have over the default soap processors, they also gave me the ability to create a “front-end” component to several services. With a well written custom soap processor, one could also eliminate hard-wiring service invocations (by plucking the QName from SOAP input and locating the service to be invoked).
I read with consternation yesterday that custom soap processors have been deprecated in webM 7.1. From what I have read in the documentation, it looks like I would have to create a default soap processor “target” for each service I intend to expose as a web service. Each such target would have the logic for extracting the soap request payload, service invocation and composing the soap response.
It sounds quite counter-intuitive to me that we need to create a default soap processor target for each service that I need to expose. This would also lead to copying/duplicating all the code that handles soap data. One could argue that soap data handling can be written in a separate utility service to be invoked from the default soap processor targets, but that doesnt take away the fact that there would be lot of services which perform very little function apart from just acting as soap receivers.
Am I missing something very important in webM 7.1 that makes my fears/concerns unfounded?