RE: New issue: We need a 'default default' action for faults

A predefined fault action URI might help in the WSDL-less service
scenario, but also in the case where I have a WSDL which defines the
action values for all my normal messages and for many or most of my
faults, but there are some extra faults not defined in the WSDL that
still need action URIs.

So I put forth as one justification for the proposal that there are more
use cases for a predefined fault action than a predefined message
action.

> -----Original Message-----
> From: Marc Hadley [mailto:Marc.Hadley@Sun.COM]
> Sent: Monday, February 07, 2005 10:30 AM
> To: Jonathan Marsh
> Cc: Martin Gudgin; public-ws-addressing@w3.org
> Subject: Re: New issue: We need a 'default default' action for faults
> 
> Hmmm, so do we also need a 'default default' action for normal
> messages
> when there is no WSDL ? Seems like this would be the ultimate result
> of
> this train of logic.
> 
> Marc.
> 
> On Feb 4, 2005, at 2:43 PM, Jonathan Marsh wrote:
> 
> >
> > All good, though it is strange to have something in the WSDL binding
> > spec that is specifically non-WSDL.  Perhaps the predefined URI
> should
> > go in the core spec somewhere instead?
> >
> >> -----Original Message-----
> >> From: Martin Gudgin
> >> Sent: Friday, February 04, 2005 8:36 AM
> >> To: public-ws-addressing@w3.org
> >> Cc: Jonathan Marsh
> >> Subject: RE: New issue: We need a 'default default' action for
> faults
> >>
> >> Mark pointed out to me that we have a 'template' for raising
> issues.
> >> So here's the issue using that template which will hopefully be
> more
> >> complete and easier for people to get their heads around;
> >>
> >> Title: 'default default' action URI for fault messages.
> >>
> >> Description: The resolution for Issue 35[1] removed the fixed
> default
> >> action URI for fault messages and replaced it with a algorithm
> similar
> >> to that for non-fault messages. However, I believe we need a fixed
> URI
> >> for people to use when returning a fault that is NOT described in
> >> WSDL. If a fault isn't described in WSDL, then people used to be
> able
> >> to use the fixed URI. That option is no longer available to them.
> >>
> >> Justification: If we don't add a 'default default' then faults not
> >> described in WSDL will have to define a specific action, whereas
> they
> >> might wish to utilize a default fixed value. And they may have
> nowhere
> >> to define that specific action because the fault is NOT described
> in
> >> WSDL; they can't use explicit association (wsa:Action) because
> there
> >> is no where to hang the attribute.
> >>
> >> Target: wsdl
> >>
> >> Proposal:  Add text along the following lines to section 3
> >>
> >> 	3.4 Default Action URI for faults NOT listed in WSDL
> >>
> >> 	In some systems not all faults can or will be listed in a WSDL
> >> description.
> >> 	Such faults still need an action URI. Faults not listed in a
> >> WSDL
> >> 	description MAY use the following
> >> 	action URI; http://www.w3.org/ws/2005/02/addressing/fault
> >>
> >>
> >> cheers
> >>
> >> Gudge
> >>
> >> [1] http://www.w3.org/2002/ws/addr/wd-issues/#i035
> >>
> >>> -----Original Message-----
> >>> From: public-ws-addressing-request@w3.org
> >>> [mailto:public-ws-addressing-request@w3.org] On Behalf Of
> >>> Martin Gudgin
> >>> Sent: 01 February 2005 09:27
> >>> To: public-ws-addressing@w3.org
> >>> Cc: Jonathan Marsh
> >>> Subject: New issue: We need a 'default default' action for faults
> >>>
> >>>
> >>> We've got rid of the fixed default action URI for fault messages
> and
> >>> replaced it with a algorithm similar to that for non-fault
> messages.
> >>> However, I believe we need a fixed URI for people to use when
> >>> returning
> >>> a fault that is NOT described in WSDL. If my fault isn't described
> >> in
> >>> WSDL, I used to be able to use the fixed URI and now I can't
> because
> >>> it's gone.
> >>>
> >>> Please can we put the fixed URI (or one like it) back, indicating
> >> that
> >>> it is intended ONLY for use with faults NOT listed in the WSDL.
> >>>
> >>> Note, I am NOT trying to re-open issue 35[1], I just think we
> missed
> >> a
> >>> case, that's all.
> >>>
> >>> Thanks
> >>>
> >>> Gudge
> >>>
> >>> [1] http://www.w3.org/2002/ws/addr/wd-issues/#i035
> >>>
> >>>
> >
> >
> ---
> Marc Hadley <marc.hadley at sun.com>
> Web Technologies and Standards, Sun Microsystems.

Received on Tuesday, 8 February 2005 00:07:43 UTC