You are viewing a plain text version of this content. The canonical link for it is here.
Posted to java-dev@axis.apache.org by Michael Rheinheimer <ro...@us.ibm.com> on 2008/04/11 00:11:53 UTC
[Axis2] arbitrary prefix for anon operations?
Hi,
The sandesha team has requested that a custom prefix be included in the
QNames of the anonymous operations (anonOutInOp, etc) to differentiate
axis2-originated anon operations from customer-defined operations. Is it
simply a matter of adding a prefix to the QName definitions in kernel
code: org.apache.axis2.client.ServiceClient? Doing so caused no errors
in my testing, but I am certainly capable of overlooking something, and
this is core kernel stuff, so please help me out! See related Jira:
https://issues.apache.org/jira/browse/AXIS2-3730
Thank you much!
mike
Re: [Axis2] arbitrary prefix for anon operations?
Posted by R J Scheuerle Jr <sc...@us.ibm.com>.
I will be displaying a great deal of ignorance here. Please read
carefully.
Currently the anonOutInOp is defined as an unqualified QName. In general,
this seems like the wrong approach for anything that is flowed in a
message.
My suggestion is that the anonOutInOp have a real namespace (perhaps
http://org.apache.axis2).
And that the code be changed to send use this new QName but also be smart
enough to understand messages that have the old unqualified qname.
To answer Dustin's question: I believe Sandesha needs a way to recongize
the anonOutInOp that is flowed in a message and determine that it is not a
user-defined operation.
If we always flow the anonOutInOp as a qualified QName, this is an easy
check.
David Parsons is the developer that requested this change
Rich Scheuerle
IBM Web Services
Apache Axis2 (scheu@apache.org)
512-838-5115 (IBM TL 678-5115)
"David Illsley" <da...@gmail.com> wrote on 04/11/2008 02:29:34 AM:
> Mike,
> In general I'm against using bits of a QName for anything other than
> bits of a 'real' QName... it just feels likely to confuse. A parameter
> sounds feasible and clean option, perhaps there are others.
> Cheers,
> David
>
> On Thu, Apr 10, 2008 at 11:16 PM, Dustin Amrhein <da...@yahoo.com>
wrote:
> > Have we explored storing a specific Parameter on the AxisOperationthat
told
> > them an operation was not a user-defined operation? I know there is
very
> > little chance of a collision with a user-defined operation, but there
is
> > still a small chance. Is storing a parameter on the AxisOperation
instead a
> > feasible option?
> >
> >
> >
> > Michael Rheinheimer <ro...@us.ibm.com> wrote:
> >
> > Hi,
> >
> > The sandesha team has requested that a custom prefix be included in the
> > QNames of the anonymous operations (anonOutInOp, etc) to differentiate
> > axis2-originated anon operations from customer-defined operations. Is
it
> > simply a matter of adding a prefix to the QName definitions in kernel
code:
> > org.apache.axis2.client.ServiceClient? Doing so caused no errors in my
> > testing, but I am certainly capable of overlooking something, and this
is
> > core kernel stuff, so please help me out! See related Jira:
> >
> > https://issues.apache.org/jira/browse/AXIS2-3730
> >
> > Thank you much!
> > mike
> >
> >
> > __________________________________________________
> > Do You Yahoo!?
> > Tired of spam? Yahoo! Mail has the best spam protection around
> > http://mail.yahoo.com
>
>
>
> --
> David Illsley - IBM Web Services Development
>
> ---------------------------------------------------------------------
> To unsubscribe, e-mail: axis-dev-unsubscribe@ws.apache.org
> For additional commands, e-mail: axis-dev-help@ws.apache.org
>
Re: [Axis2] arbitrary prefix for anon operations?
Posted by David Illsley <da...@gmail.com>.
Mike,
In general I'm against using bits of a QName for anything other than
bits of a 'real' QName... it just feels likely to confuse. A parameter
sounds feasible and clean option, perhaps there are others.
Cheers,
David
On Thu, Apr 10, 2008 at 11:16 PM, Dustin Amrhein <da...@yahoo.com> wrote:
> Have we explored storing a specific Parameter on the AxisOperation that told
> them an operation was not a user-defined operation? I know there is very
> little chance of a collision with a user-defined operation, but there is
> still a small chance. Is storing a parameter on the AxisOperation instead a
> feasible option?
>
>
>
> Michael Rheinheimer <ro...@us.ibm.com> wrote:
>
> Hi,
>
> The sandesha team has requested that a custom prefix be included in the
> QNames of the anonymous operations (anonOutInOp, etc) to differentiate
> axis2-originated anon operations from customer-defined operations. Is it
> simply a matter of adding a prefix to the QName definitions in kernel code:
> org.apache.axis2.client.ServiceClient? Doing so caused no errors in my
> testing, but I am certainly capable of overlooking something, and this is
> core kernel stuff, so please help me out! See related Jira:
>
> https://issues.apache.org/jira/browse/AXIS2-3730
>
> Thank you much!
> mike
>
>
> __________________________________________________
> Do You Yahoo!?
> Tired of spam? Yahoo! Mail has the best spam protection around
> http://mail.yahoo.com
--
David Illsley - IBM Web Services Development
---------------------------------------------------------------------
To unsubscribe, e-mail: axis-dev-unsubscribe@ws.apache.org
For additional commands, e-mail: axis-dev-help@ws.apache.org
Re: [Axis2] arbitrary prefix for anon operations?
Posted by Dustin Amrhein <da...@yahoo.com>.
Have we explored storing a specific Parameter on the AxisOperation that told them an operation was not a user-defined operation? I know there is very little chance of a collision with a user-defined operation, but there is still a small chance. Is storing a parameter on the AxisOperation instead a feasible option?
Michael Rheinheimer <ro...@us.ibm.com> wrote:
Hi,
The sandesha team has requested that a custom prefix be included in the QNames of the anonymous operations (anonOutInOp, etc) to differentiate axis2-originated anon operations from customer-defined operations. Is it simply a matter of adding a prefix to the QName definitions in kernel code: org.apache.axis2.client.ServiceClient? Doing so caused no errors in my testing, but I am certainly capable of overlooking something, and this is core kernel stuff, so please help me out! See related Jira:
https://issues.apache.org/jira/browse/AXIS2-3730
Thank you much!
mike
__________________________________________________
Do You Yahoo!?
Tired of spam? Yahoo! Mail has the best spam protection around
http://mail.yahoo.com