RE: [dsml] DSMLv2-draft11

[Shon Vella <SVELLA@novell.com>]

I just tried validating:

<batchRequest xmlns:xsd="http://www.w3.org/2001/XMLSchema"
                             xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance"
                        xmlns="urn:oasis:names:tc:DSML:2:0:core">
  <modifyRequest dn="CN=Bob Rush,OU=Dev,DC=Example,DC=COM">
     <attr name="telephoneNumber" operation="replace">
       <value xsi:type="xsd:base64Binary">U2VhcmNoIFJlcXVlc3QgRXhhbXBsZQ==</value>
       <value>234 212 4534</value> 
     </attr>
     <attr name="sn" operation="replace">
        <value>Rush</value>
     </attr>
     <attr name="directReport" operation="add">
       <value>CN=John Smith, DC=microsoft, DC=com</value>
     </attr>
  </modifyRequest>
</batchRequest>

againset XSV (available at ftp://ftp.cogsci.ed.ac.uk/pub/XSV/XSV12.EXE)

and it seemed to like it.

The XML is a composition of a couple of the examples in the spec.

I had to tweak the DSML namespace identifier from the spec because it did not match what was in the schema file. Either the example or the schema file is incorrect and needs to be, but I haven't delved into which.

In any case, at least one validator seems to like the xsi:type, but I haven't tried any others.

Shon Vella
Software Engineer, Consultant
svella@novell.com
Novell, Inc., the leading provider of Net services software
www.novell.com 

>>> Jeff Parham <jeffparh@windows.microsoft.com> 11/12/01 03:45PM >>>
You're right, the schema does seem to indicate that -- I missed that
example.  However, I can't find a parser (MS or non-MS) that will accept
the example, either.  Does *anyone* have a parser that successfully
validates this usage?

-J

-----Original Message-----
From: Shon Vella [mailto:SVELLA@novell.com] 
Sent: Monday, November 12, 2001 12:31 PM
To: dsml@lists.oasis-open.org; Jeff Parham
Subject: Re: [dsml] DSMLv2-draft11


Use of xsi:type as a union discriminator is specifically described in
the schema specification. Either there is something wrong with how the
union is defined in our schema, how it is being applied in the instance
documents, or the validators you are trying are not fully compliant.

See http://www.w3.org/TR/xmlschema-2/#union-datatypes and look at the
second example box.


Shon Vella
Software Engineer, Consultant
svella@novell.com 
Novell, Inc., the leading provider of Net services software
www.novell.com 

>>> Jeff Parham <jeffparh@windows.microsoft.com> 11/12/01 01:06PM >>>
1. Per exchange between Rob and myself on the alias, incorporated Rob's
RegEx pattern for AttributeDescriptionValue.

2. Per implementors, made the "name" attribute required for the
AttributeValueAssertion, AttributeDescription, and SubstringFilter
elements.

3. Changed the "code" attribute to be required for the ResultCode
element.

4. Per implementors, the spec & examples appeared to alternate between
using "notAttempted" and "malformedRequest" as error types for syntax
errors.  Standardized on "malformedRequest".

5. Added the OASIS statements required by
http://www.oasis-open.org/who/intellectualproperty.shtml.

6. Per Christine's 11/9 mail issue #1, changed the name of the
"controlType" attibute in the spec to "type" to match the post-baseline
schema.

7. Per implementors and Christine's 11/9 mail issue #2, changed
lingering reference to "extendedReq" to "extendedRequest" in the schema.

8. Per Shon's 10/10 mail, changed xsi:type="base64binary" to
xsi:type="xsd:base64Binary" in spec examples.  (I missed this in the
prior revs.)

9. Per implementors, DsmlANY allows complex types (nested elements) but
does not allow simple types (xsd:string, xsd:base64Binary, etc.).
Therefore removed definition of DsmlANY and replaced references to it
with references to xsd:anyType, which allows both simple and complex
types.

### UNRESOLVED ###
Use of xsi:type as a union discriminator for instances of DsmlValue does
not appear to be allowed -- we've tried 3 different validators and this
is consistently rejected.  XML Schema appears to disallow use of
xsi:type as a union discriminator, though it works fine for refining
xsd:anyType instances and for specifying derived types.  One way or
another we must have a union discriminator for DsmlValue, even if it
requires going back to the "manual" discrimination used in the baseline
spec.  Andy is trying to get to the bottom of this, but if anyone has
any info to contribute please do.

Thanks,
-J


----------------------------------------------------------------
To subscribe or unsubscribe from this elist use the subscription
manager: <http://lists.oasis-open.org/ob/adm.pl>

----------------------------------------------------------------
To subscribe or unsubscribe from this elist use the subscription
manager: <http://lists.oasis-open.org/ob/adm.pl>