Ken Hofsass wrote:
> Bobby, No, support for that hasn't been added. My recollection is
> that there no firm decision on how this type of test requirement should
> be handled. The discussion needs to be revived, so here's how I
> remember the earlier discussions...
Yes.
> I think that Kohsuke was encouraging the use of the replaceable
> transport mechanism (via re-coding) for MEX runtime & wsimport which
> would eliminate the need to flag tests to run only for specific
> transports since most/all of the rest of JAX-WS & WSIT already uses the
> transport/pipe stuff.
Yes. I thought one of the points of MEX is its transport independence.
So I think it's unfortunate if our MEX implementation doesn't take
advantage of that.
I thought a natural implementation of MEX would use the JAX-WS runtime
as the transport mechanism, hence it would automatically gets the
transport independence from there.
And if the JAX-WS team can't even eat our own dogfood for some technical
reasons, I think we should better learn why.
> However, as Bobby just pointed out in another
> email, Jitu was also saying something about certain tests requiring HTTP.
Yes, even if we port MEX over to the JAX-WS runtime, that does not
eliminate a need for requiring certain transport for some tests.
> Jitu, Kohsuke, can we get your input on this?
What if we have
<descriptor transport="http"> ...
? We can internally have each ApplicationContainer interface declare its
transport type and match against this.
If I remember correctly, Bobby had another requirement that he needed to
deploy a servlet that he uses to emulate a bogus response.
--
Kohsuke Kawaguchi
Sun Microsystems kohsuke.kawaguchi_at_sun.com