Hi Glen, It's all up for grabs as far as I'm concerned.. though I have the following requirements in mind... 1. Must be possible to deploy support for only one of the supported addressing versions 2. Must be possible to disable the validation 3. Avoid churn for its own sake 4. Some of the validation has to happen after operation dispatch
I'm pretty sure there are people who would object to including "WS-Addressing" specific logic in the "Addressing"Dispatcher so there should probably be: 5. No WS-Addressing spec type logic in the AddressingDispatcher Brian is right that I'm mid refactor of the dispatchers (and hope to complete that this week) but I don't think that's relevant. I'm not aware of any performance numbers suggesting it's important to minimise the chain length... so I'm not sure if it's worth the churn. Do you have anything in particular you are considering? David On 20/03/07, Brian De Pradine <[EMAIL PROTECTED]> wrote:
Hello Glen, See my response below. Cheers Brian DePradine Web Services Development IBM Hursley External +44 (0) 1962 816319 Internal 246319 If you can't find the time to do it right the first time, where will you find the time to do it again? Glen Daniels <[EMAIL PROTECTED]> wrote on 20/03/2007 17:11:58: > Hi folks: > > Why do we have a separate AddressingValidationHandler? Wouldn't it make > more sense simply to do any addressing-specific validation right in the > AddressingInHandler or the AddressingBasedDispatcher? Is it possible to > simplify this? I believe that historically the AddressingBasedDispatcher is not meant to be tied to strongly to ws-addressing. Hence it is not a part of the Addressing module :-(. This has meant the need to have a separate AddressingValidation handler. However, I believe that David is planning some refactoring of the dispatchers which may mean that we can revisit this. > Also, rather than having separate Handlers for the different versions of > addressing, wouldn't it make more sense to have just one > AddressingInHandler and then let that do version-specific work with its > own utility classes? A separate handler seems like overkill. I had considered refactoring the inbound processing into one handler, but elected not to do it in the end. At the time, I was taking the approach of "if it ain't broke, don't fix it." Now, I have no objections to doing it if folks feel that this is the best way to go. > > Thanks, > --Glen > > --------------------------------------------------------------------- > To unsubscribe, e-mail: [EMAIL PROTECTED] > For additional commands, e-mail: [EMAIL PROTECTED] > ________________________________ Unless stated otherwise above: IBM United Kingdom Limited - Registered in England and Wales with number 741598. Registered office: PO Box 41, North Harbour, Portsmouth, Hampshire PO6 3AU
-- David Illsley - IBM Web Services Development --------------------------------------------------------------------- To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
