W3C home > Mailing lists > Public > public-ws-chor@w3.org > September 2005

Re: Checking the CDL schema

From: Nickolas Kavantzas <nickolas.kavantzas@oracle.com>
Date: Thu, 15 Sep 2005 19:27:43 -0700
Message-ID: <009101c5ba66$39175c70$538e1990@us.oracle.com>
To: "Tony Fletcher" <tony_fletcher@btopenworld.com>
Cc: <public-ws-chor@w3.org>
MessageWhat exactly XMLSpy gives you as an error?

I've just run 'Check Well Formedness' and 'Validate' under the XML tab on this schema 
and things are OK.

--
Nick
  ----- Original Message ----- 
  From: Tony Fletcher 
  To: public-ws-chor@w3.org 
  Sent: Thursday, September 15, 2005 6:01 PM
  Subject: Checking the CDL schema


  Dear CDLers,

  Greetings to you all - and keep up the good work.  Just to show that I am still lurking (/taking an interest!) and will try to be helpful when I can....

  I have extracted the full CDL schema from the 15 Sept 05 editors' version and run it through XMLSpy, Stylus Studio and Oxygen.  Unfortunately they all agree that there is a somewhat subtle 'technical' problem with the schema in its current form.

  The XML Schema 'any' construct is often used as 'catchall' extension point, but often is a source of problems as schema validators will throw a 'wobbly' if the can not always tell when something in an instance document corresponds to this particular 'any' in the schema or something else.  We have currently fallen into this trap!

  In your case all your elements are built on   tExtensibleElements  which contains  <any namespace="##other" processContents="lax" minOccurs="0" maxOccurs="unbounded"/>

  In particular this 'any' of tExtensibleElements used as the basis for tSequence means that a validator can not tell, in all cases, whether an element in an instance document corresponds to this any or the one in the sequence allowed by activity which also includes an 'any' in its choice.

  One solution that works is to wrap the 'any' in tExtensibleElements in another element as below:

  <complexType name="tExtensibleElements">
    <annotation>
     <documentation> This type is extended by other WS-CDL component types to allow elements
      and attributes from other namespaces to be added. This type also contains the
      optional description element that is applied to all WS-CDL constructs.
     </documentation>
    </annotation>
    <sequence>
     <element name="description" minOccurs="0">
      <complexType mixed="true">
       <sequence minOccurs="0" maxOccurs="unbounded">
        <any processContents="lax"/>
       </sequence>
       <attribute name="type" type="cdl:tDescriptionType" use="optional"
        default="documentation"/>
      </complexType>
     </element>
     <element name="CDLExtension" type="cdl:CDLExtensionType" minOccurs="0"
      maxOccurs="unbounded"/>
    </sequence>
    <anyAttribute namespace="##other" processContents="lax"/>
   </complexType>

   <complexType name="CDLExtensionType">
    <sequence>
     <any namespace="##other" processContents="lax" minOccurs="0" maxOccurs="unbounded"/>
    </sequence>
   </complexType>

  Actually this technique would allow the namespace attribute for the any in the CDLExtensionType to be changed to  namespace="##any"  if desired.
  However, I am sure that there are other solutions as well (and not everyone regards this type of problem as serious).

  I have also tried checking the two complete examples against my modified schema (and the original) and they check out with the exception that the third informationType (name="badPOAckType") has an attribute of exceptionType="true" which is (no longer?) allowed by the schema.  Fine with that attribute removed.


  Best Regards     Tony
  A M Fletcher
  Tel: +44 (0) 1473 729537   Mobile: +44 (0) 7801 948219
   tony.fletcher@choreology.com    amfletcher@iee.org       (also tony_fletcher@btopenworld.com)
Received on Friday, 16 September 2005 02:27:58 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Saturday, 18 December 2010 01:01:41 GMT