RE: Proposed resolution: issues 78, 16

Matt, the first case, the first child _not_ marked as a root,
would mean that there can be an element later that is marked as a
root. The "start" attribute is to be used on the element _above_
what it concerns, i.e. on the Body element for RPC element
recognition, or on Header for pointing to the first header to
process.

This brings me to another reason why we cannot remove "root"
attribute if we add "start" - RPC needs to point to the RPC
element while an encoding wants to mark serialization root(s). If
we remodel RPC to be encoding-independent, IMHO it would
naturally not be a serialization root, either, then. Every
parameter would be its serialization root.

                            Jacek Kopecky

                            Idoox
                            http://www.idoox.com/




On Thu, 2 Aug 2001, Matt Long wrote:

 > Would not the first immediate child of header|body either not marked with
 > root attribute or marked with SOAP-ENC:root="1" be equivalent to the
 > serialization root marked with the "start" attribute?
 >
 > -Matt
 >
 > > -----Original Message-----
 > > From: xml-dist-app-request@w3.org
 > > [mailto:xml-dist-app-request@w3.org]On
 > > Behalf Of Jacek Kopecky
 > > Sent: Thursday, August 02, 2001 11:31 AM
 > > To: Doug Davis
 > > Cc: xml-dist-app@w3.org
 > > Subject: Re: Proposed resolution: issues 78, 16
 > >
 > >
 > >  Doug,
 > >  I don't think we could remove "root" from section 5 because in
 > > general you can have multiple serialization roots and using
 > > "start" for marking these would be a hack.
 > >
 > >                             Jacek Kopecky
 > >
 > >                             Idoox
 > >                             http://www.idoox.com/
 > >
 > >
 > >
 > >
 > > On Thu, 2 Aug 2001, Doug Davis wrote:
 > >
 > >  > ok- #1 is a good reason.  8-)
 > >  > If we do go with "start" we can remove "root" right?
 > >  > -Dug
 > >  >
 > >  >
 > >  > Jacek Kopecky <jacek@idoox.com>@w3.org on 08/02/2001 11:42:18 AM
 > >  >
 > >  > Sent by:  xml-dist-app-request@w3.org
 > >  >
 > >  >
 > >  > To:   Doug Davis/Raleigh/IBM@IBMUS
 > >  > cc:   <xml-dist-app@w3.org>
 > >  > Subject:  Re: Proposed resolution: issues 78, 16
 > >  >
 > >  >
 > >  >
 > >  >  Doug,
 > >  >  There are couple of reasons for "start":
 > >  >  1) with "root", you have to parse the whole body to see if the
 > >  > attribute was not used in case it really was not, while with
 > >  > "start" you either know it's the first element (if "start"'s not
 > >  > present) or you only have to parse up to the element pointed to
 > >  > by "start",
 > >  >  2) "root" belongs to encodings and if we move it to the core,
 > >  > the multiref notion would be moved as well (because "root" is for
 > >  > use with multirefs) and we would practically mandate a particular
 > >  > way of referencing data inside the payload,
 > >  >  3) as has been pointed out a few times, "start" has more uses,
 > >  > like e.g. it could point to the first header to be processed.
 > >  >  Kind regards
 > >  >
 > >  >                             Jacek Kopecky
 > >  >
 > >  >                             Idoox
 > >  >                             http://www.idoox.com/
 > >  >
 > >  >
 > >  >
 > >  >
 > >  > On Thu, 2 Aug 2001, Doug Davis wrote:
 > >  >
 > >  >  > Perhaps someone could explain something to me.
 > >  >  > The proposal is for a "start" attribute that
 > >  >  > refers to the top-most element in the body.  Right
 > >  >  > now SOAP has the notion of a "root" attribute
 > >  >  > (granted its in the encoding section, but it can
 > >  >  > be moved).  How is the "start" attribute any
 > >  >  > better than the "root" attribute?  In both cases
 > >  >  > we need to read/parse at least the first XML element
 > >  >  > of each to determine either if the "name" matches
 > >  >  > the one on the "start" attribute, or if the
 > >  >  > "root" attribute is there.  I don't see the
 > >  >  > benefit of "start". Am I missing something?
 > >  >  > -Dug
 > >  >  >
 > >  >
 > >  >
 > >  >
 >

Received on Thursday, 2 August 2001 13:05:55 UTC