Re: RPC/Encoding dependency

 Gudge, 
 I see what you mean, thank you. I think that the RPC section
should not talk about elements anymore because the resolution was
to base it on the Data Model, not the particular in-spec Encoding
of it.

                   Jacek Kopecky

                   Senior Architect, Systinet (formerly Idoox)
                   http://www.systinet.com/



On Tue, 12 Feb 2002, Martin Gudgin wrote:

 > The fact that the RPC mapping spells out that you have to use elements with
 > names derived a certain way. It does not allow attributes, or other element
 > names to be used. Seems pretty clear cut to me.
 > 
 > Gudge
 > 
 > 
 > ----- Original Message -----
 > From: "Jacek Kopecky" <jacek@systinet.com>
 > To: "Martin Gudgin" <marting@develop.com>
 > Cc: "XML Protocol Discussion" <xml-dist-app@w3.org>
 > Sent: Tuesday, February 12, 2002 10:19 AM
 > Subject: Re: RPC/Encoding dependency
 > 
 > 
 > > Gudge,
 > >  it's unclear to me what makes you think that specifying an other
 > > encoding (of our data model) is prohibited on the RPC request or
 > > response structs.
 > >  Kind regards,
 > >
 > >                    Jacek Kopecky
 > >
 > >                    Senior Architect, Systinet (formerly Idoox)
 > >                    http://www.systinet.com/
 > >
 > >
 > >
 > > On Fri, 8 Feb 2002, Martin Gudgin wrote:
 > >
 > >  > In the editors copy of Part 2[1], section 4.1[2] states;
 > >  >
 > >  > 'As noted above, RPC invocation and response structs can be encoded
 > >  > according to the rules in 3 SOAP Encoding, or other encodings can be
 > >  > specified using the encodingStyle attribute'
 > >  >
 > >  > While I grant that strictly speaking for each parameter accessor I
 > could use
 > >  > a different encoding, the encoding of the procedure/method signature
 > is, by
 > >  > definition, the same as Section 3[3] encoding. For example, given the
 > >  > following COM IDL method signature
 > >  >
 > >  >     void Add ( [in] long x, [in] long y, [out] long* sum );
 > >  >
 > >  > I could not encode the request as;
 > >  >
 > >  >     <Add x='10' y='20'
 > xmlns:soap='http://www.w3.org/2001/12/soap-envelope'
 > >  >          soap:encodingStyle='urn:example-org:attrenc' />
 > >  >
 > >  > in fact, I could ONLY encode the request as;
 > >  >
 > >  >     <Add xmlns:soap='http://www.w3.org/2001/12/soap-envelope'
 > >  >          soap:encodingStyle='http://www.w3.org/2001/12/soap-encoding' >
 > >  >       <X>10</x>
 > >  >       <y>20</y>
 > >  >     </Add>
 > >  >
 > >  > Given the following IDL
 > >  >
 > >  >     struct Point
 > >  >     {
 > >  >         long x,y;
 > >  >     }
 > >  >
 > >  >     void Add ( [in] struct Point pt1,
 > >  >                [in] struct Point pt2,
 > >  >                [out] struct Point* ptret );
 > >  >
 > >  > I *could* encode each point using something other than section 3;
 > >  >
 > >  >     <Add xmlns:soap='http://www.w3.org/2001/12/soap-envelope'
 > >  >          soap:encodingStyle='http://www.w3.org/2001/12/soap-encoding' >
 > >  >       <pt1 x='10' y='20' soap:encoding='urn:example-org:attrenc' />
 > >  >       <pt2 y='10' x='20' soap:encoding='urn:example-org:attrenc' />
 > >  >     </Add>
 > >  >
 > >  > but the request element is always serialized according to Section 3.
 > >  >
 > >  > Gudge
 > >  >
 > >  > [1] http://www.w3.org/2000/xp/Group/1/10/11/soap12-part2.xml
 > >  > [2] http://www.w3.org/2000/xp/Group/1/10/11/soap12-part2.xml#IDAGG5CF
 > >  > [3] http://www.w3.org/2000/xp/Group/1/10/11/soap12-part2.xml#soapenc
 > >  >
 > >
 >

Received on Monday, 18 February 2002 07:02:52 UTC