W3C home > Mailing lists > Public > xmlschema-dev@w3.org > October 2004

Re: "Re: xs:include changing targetNamespace"'

From: Kasimier Buchcik <kbuchcik@4commerce.de>
Date: Wed, 06 Oct 2004 12:17:34 +0200
To: <xmlschema-dev@w3.org>
Message-ID: <4163C63E.3000806@4commerce.de>

Hi,

ht@inf.ed.ac.uk wrote:
> "Michael Kay" <mike@saxonica.com> writes:
> 
> 
>>>Does this mean as well, that if a schema document (A) with an
>>>existing target namespace includes a schema document (B) with no
>>>target namespace, which, in turn, includes a schema document (C),
>>>the target namespace of (B) will still be 'absent' when the
>>>constraints for including of (C) are applied? IOW, does this
>>>chameleon-effect touch the first level of includes only?
>>>
>>
>>My reading is that the components derived from C are copied into the schema
>>corresponding to B with no change in namespace, but when the components are
>>then copied into A, the namespace is changed regardless whether they
>>originated from B or C.
> 
> 
> That's my reading also.

I see. Are those schema construction stages and copy operations needed 
to ensure some of the constraints I'm possibly not aware of, or could 
you imagine an on-the-fly including mechanism? If the resulting schema, 
constructed on-the-fly, would be correct, would the sub-stage schemata 
still be needed?

Thanks & regards,

Kasimier

PS: Hmm, the address of Henry Thompson was somehow mangled in my last mail.
Received on Wednesday, 6 October 2004 10:18:13 UTC

This archive was generated by hypermail 2.3.1 : Wednesday, 5 February 2014 07:15:11 UTC