Re: Schema-Specification normative preference wasRE:[ebxml-msg]Issue73:http://schemas.xmlsoap.org/soap/envelopenamespace

[Duane Nickull <duane@xmlglobal.com>]

Comments inline:

Arvola Chan wrote:
> 
> Duane:
> 
> The reason why the ebxml-msg TC has to define its own namespace is because
> the ebXML message header elements are defined as extensions to the SOAP
> Header and SOAP body elements. SOAP requires extension elements to be
> namespace qualified.
>>>>>>>

Arvola:

Yes - I am very aware of namespaces themselves.  My question revolved
around a comment that someone made to this thread complaining that they
could not import the definitions from the namespace schemaLocation
target.  I wondered why it would ever be a requirement to import the
actual definitions (the specification says you must be able to do so via
the schemalocation attribute as you quote below.).  If it is a
requirement, then I personally was curious of the driver for that
requirement.

It seems like it may not be a formal requirement and that perhaps we are
simply using the namespace values to avoid conflicts (since you cite the
SOAP example as the driver).  If this is true,  then there is no
problem.

> 
> The schemaLocation attributes in the schema and in instance documents only
> serve as hints. Schema processors are not required to use these hints. The
> application that invokes a schema processor can direct the latter not to
> dereference the value specified for the schemaLocation attribute and instead
> use some cached version of the schema.
>>>>>

Local caching still requires you resolve it at least once.
  
> The following is an excerpt from http://www.w3.org/TR/xmlschema-0/
> 
> "
> 5.6 schemaLocation
> XML Schema uses the schemaLocation and xsi:schemaLocation attributes in
> three circumstances.
> 
> 1. In an instance document, the attribute xsi:schemaLocation provides hints
> from the author to a processor regarding the location of schema documents.
> The author warrants that these schema documents are relevant to checking the
> validity of the document content, on a namespace by namespace basis. For
> example, we can indicate the location of the Report schema to a processor of
> the Quarterly Report:
<SNIP>/.....

My interpretation of the above:  This refers to importing definitions in
order to perform a validating parse on an "instance document" - as the
example states.  If someone wishes to check the validity of a namespace
qualified element to see if its' content matches (ie - datatyping
constaints or content model), then they may have to resolve the
schemalocation.

What I am asking is "Is the requirement importing definitions OR is it
to avoid naming conflicts?"

If anyone knows the answer, can they please state the answer and if it
is definition importing as a strict requirement,  please state why.

Thanks

Duane Nickull
-- 
CTO, XML Global Technologies
****************************
Transformation - http://www.xmlglobal.com/prod/foundation/
ebXML Central - http://www.xmlglobal.com/prod/central/