W3C home > Mailing lists > Public > public-xsd-databinding@w3.org > May 2006

RE: ISSUE-58: anonymous complexType

From: <jon.calladine@bt.com>
Date: Tue, 30 May 2006 10:10:08 +0100
Message-ID: <657E2B34242ACC46BDD6774A9728F9E3469CDC@E03MVZ3-UKDY.domain1.systemhost.net>
To: <peter.hendry@capeclear.com>, <public-xsd-databinding@w3.org>

Pete,

>> I would be surprised if any actively developed SOAP implementations
do not fully support anonymous types.

Maybe we were unlucky but this didn't work with our primary commercial
vendors implementation.

>>The message part of the request and response point each point to a
different element where that element contains an anonymous complexType
with a sequence of elements as its child.

Understood, but the first global element could equally point to a named
global complex type that describes the same sequence. I appreciate this
will have some implications from a databinding viewpoint but is
equivalent as far as the message goes (and the venetian blind model IS
well supported in tools).

>> This is a very widely used pattern and disallowing (or even
recommending against it) would, in my view, be unwise as vendors that
support wrapped document/literal will not be moving away from it.

If the pattern is well supported it will be in the BP doc if not then it
cannot be, we are describing what works in the BP not what should
work..... We will soon be entering a phase of more practical and formal
testing which will identify just how niche these problems are.

JonC

-----Original Message-----
From: public-xsd-databinding-request@w3.org
[mailto:public-xsd-databinding-request@w3.org] On Behalf Of Peter Hendry
Sent: 29 May 2006 13:12
To: Databinding WG
Subject: Re: ISSUE-58: anonymous complexType


The wrapped document/literal style favoured by .NET and now JAX-WS when
doing code-2-wsdl generation requires the use of anonymous complex types
to represent the arguments. The message part of the request and response
point each point to a different element where that element contains an
anonymous complexType with a sequence of elements as its child.

This is a very widely used pattern and disallowing (or even recommending
against it) would, in my view, be unwise as vendors that support wrapped
document/literal will not be moving away from it.

This applies to ISSUE-18 also.

I would be surprised if any actively developed SOAP implementations do
not fully support anonymous types.

Pete

Databinding Issue Tracker wrote:
> ISSUE-58: anonymous complexType
>
> http://www.w3.org/2005/06/tracker/databinding/issues/58
>
> Raised by: George Cowe
> On product: Advanced
>
> We use the following pattern in our schemas:
>
> <xs:element name="foo">
>   <xs:complexType>
>     <xs:sequence>
>       <xs:element .../>
>       <xs:element .../>
>     </xs:sequence>
>   <xs:complexType>
> </xs:element>
>
>
>
>
>
> possibly related to ISSUE-18: Schema Authoring Styles
>
>
>
>
>
>
>   
Received on Tuesday, 30 May 2006 09:10:21 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Saturday, 18 December 2010 18:20:37 GMT