You have to start from somewhere ..in our case a definition that is both complete and syntactically correct But..If the wsdl you provide (either its content or style) is hosed there is no way for the program to correct it However.. If the program is *somehow* able to intuit from the services operations/methods and parameter(s) what the generated operation and the resulting request and response parameters should be *then* that feature should definitely be supported +1 M- ********************************************************************* This email message and any files transmitted with it contain confidential information intended only for the person(s) to whom this email message is addressed. If you have received this email message in error, please notify the sender immediately by telephone or email and destroy the original message without making a copy. Thank you.
----- Original Message ----- From: "D.Kreft" <[EMAIL PROTECTED]> To: <[email protected]>; "Martin Gainty" <[EMAIL PROTECTED]> Sent: Thursday, September 21, 2006 5:52 PM Subject: Re: Axis2 is ignoring my WSDL > On 9/21/06, Martin Gainty <[EMAIL PROTECTED]> wrote: > >> What I found is.. if you use any style other than RPC the wsdl's wont be >> created under AXIS2 > > ...which I maintain is an unnecessary "feature". One should be able to > specify his own WSDL regardless of what the messageReceiver is set to > be for that service *and* be able to drop it in the appropriate place. > Altering the behavior of the WSDL generation based upon the seemingly > decision that RPC* methods cannot use a pre-generated WSDL does > nothing more than apply arbitrary rules of engagement and remove power > from the developer. > > No software developer can possibly anticipate all the "crazy" ways > that other developers will want to use his products or services, so > it's not a good idea for him to pretend like he knows exactly how > everyone is going to use them. Software should have sensible, > intuitive default behavior and allow for developers to deviate from > the "norm" (whatever that is) by flipping a few bits. One should not > have to "flip bits" to achieve sensible behavior. > > -dan >
