+1 Glen. Let's do it. -- dims
On 5/8/07, Glen Daniels <[EMAIL PROTECTED]> wrote:
Hi Dennis! Dennis Sosnoski wrote: > Rather than changing the existing methods, I suggest just adding a > MessageContext field and get/set methods. That way the engine can call > setMessageContext(ctx) before calling setComplete(), and any clients > that want to access the context can use the getMessageContext() method > for access. It's not pretty, but it at least avoids breaking > compatibility with existing client code. Ugh. :( Here's my take on this. Axis2 is still pretty brandy-new, and right now there are many fewer people using it than we hope to have in a year, or two years, or five. There are some places in our stack of APIs that are... "suboptimal" comes to mind but that's really a euphemism. :) We can either say "well that's life" and continue trying to build a vast global user base for a stack we know will continue to be flawed forever (until we get fed up enough to do Axis3!), or we can try to fix these kinds of cleanliness/design issues ASAP with as much care as we can - and look forward to an even bigger vast global user base because our APIs will be easy, clean, and make sense. I'd much rather fix what problems we can now. Yes, it will affect our current user base, but we can be as friendly about it as possible using deprecation, and the Axis2 1.2 users are still pretty much early adopters so I don't think it'll be too egregious for them. So the net is - your solution would work, but I would MUCH rather do the following. Deprecate the current Callback class, but still support it in old code for at least one full release cycle (I haven't looked deeply into this but I assume it's doable). Add a new class AxisCallback with the API that we end up with on this thread, and use that into the future. Thoughts? --Glen --------------------------------------------------------------------- To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
-- Davanum Srinivas :: http://davanum.wordpress.com --------------------------------------------------------------------- To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
