Hi,
I had a check on these two classes in core:
AbstractInboundInvocationHandler.java
AbstractOutboundInvocationHandler.java
It seems to me that they are fit for the SPI to provide the logic we
require.
Thanks,
Raymond
----- Original Message -----
From: "Ignacio Silva-Lepe" <[EMAIL PROTECTED]>
To: <[email protected]>
Sent: Thursday, October 05, 2006 11:10 AM
Subject: Re: axis2 binding "compile" dependency on core
Hmm, not sure I follow.
Btw, I'd like to clarify that when I say "we basically agree" I did not
mean that as a forgone conclusion, apologies if I gave the impression that
I was trying to go around the process. I was basically reporting our
conversation on this topic (which happened to come up on a bug-related
discussion). My intent with my reply was to put our, let's call it
preliminary, agreement out for consideration.
So what is the thought, would promoting the
AbstractOutboundInvocationHandler function (not necessarily the class line
by line) to the spi work?
----- Original Message -----
From: "Jim Marino" <[EMAIL PROTECTED]>
To: <[email protected]>
Sent: Thursday, October 05, 2006 1:52 PM
Subject: Re: axis2 binding "compile" dependency on core
O.K. We'll need to move the testcases down to SPI as well then.
Jim
On Oct 5, 2006, at 10:31 AM, Ignacio Silva-Lepe wrote:
Hi Jim,
I had a quick conversation with Raymond about this. We basically agree
that we need to promote the function performed by
AbstractOutboundInvocationHandler to the spi as an extension point,
which is also option 2 in my response to him. This is also what his
statement means, ie, AbstractOutboundInvocationHandler is "performing
an invocation over an interceptor chain", to quote him, and his
proposal is to extract that to the spi.
Thoughts?
----- Original Message ----- From: "Jim Marino"
<[EMAIL PROTECTED]>
To: <[email protected]>
Sent: Thursday, October 05, 2006 1:01 AM
Subject: Re: axis2 binding "compile" dependency on core
On Oct 4, 2006, at 9:58 PM, Raymond Feng wrote:
Hi, Ignacio.
I just found one of your patches added a "compile" dependency on
"core" to "binding.axis2". It's from
"Axis2CallbackInvocationHandler extends
AbstractOutboundInvocationHandler". I don't think it's appropriate to
have binding.axis2 depend on core. Can we fix it?
We had some discussions before to extract the invocation over
interceptor chain into SPI. It seems to be the right way to go.
Hi Raymond, could you explain what you mean by this?
Thanks,
Jim
Thanks,
Raymond
---------------------------------------------------------------------
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
---------------------------------------------------------------------
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
