Re: [ubl-dev] CCTS vs UDT

["G. Ken Holman" <gkholman@CraneSoftwrights.com>]

At 2011-12-06 13:37 +0100, ericdes wrote:
> Now I would like to understand the differences between the "CCTS 
> Core Component Type Schema Module" and the "UDT Unqualified Data 
> Type Schema Module".

The answers differ in UBL 2.0 and UBL 2.1.

In UBL 2.0 the CCTS Core Component Type Schema Module is present for 
information only, it is not included in the schema import hierarchy, 
and it merely informed UN/CEFACT in their creation of the UDT 
Unqualified Data Type schema that was included.

In UBL 2.1 the CCTS Core Component Type Schema Module is included in 
the schema import hierarchy, imported by the UBL Unqualified Data 
Type schema ... the UN/CEFACT UDT fragment is no longer being used.

> Why is e.g. "AmountType" defined in both those namespaces?

Because the unqualified data types are based on the core component 
types.  In UBL 2.0 this basis is merely informative.  In UBL 2.1 the 
basis is direct schema inheritance.  The basis of interoperability 
between different XML vocabularies is that vocabularies use at the 
leaf level of the tree content that is always based on a core 
component type (element types) and its supplementary components (attributes).

> Moreover it may have 2 optional attibutes "currencyID" and 
> "currencyCodeListVersionID" in CCTS, while it may contain only one 
> attribute, a required "currencyID" in UDT?

That was true in UBL 2.0 UDT, but it was recognized that the approach 
taken in UBL 2.0 was insufficient for long-term use in UBL.  In UBL 
2.1 UDT both attributes can be specified on amounts.

It was recognized the use of hardwired code lists in UBL 2.0 
unqualified data types inherited from UN/CEFACT does not meet 
real-world requirements of being able to accommodate changes in code 
lists or trading partners conventions of using code lists.  A lengthy 
discussion of currency codes took place on this list a couple of years ago:

  http://lists.oasis-open.org/archives/ubl-dev/200905/msg00002.html

So, in UBL 2.1, the currencyID attribute is simply a normalized 
string, and it is up to value validation to confirm that the 
currencyID attribute (in concert with the currencyCodeListVersionID) 
is meaningful in a particular trading partner relationship (without 
touching the schema).

The same is true for MIME types, units of measure, etc.

I hope this helps.

. . . . . . . . . . . . Ken

--
Contact us for world-wide XML consulting and instructor-led training
Free 5-hour video lecture: XSLT/XPath 1.0 & 2.0 http://ude.my/t37DVX
Crane Softwrights Ltd.            http://www.CraneSoftwrights.com/u/
G. Ken Holman                   mailto:gkholman@CraneSoftwrights.com
Google+ profile: https://plus.google.com/116832879756988317389/about
Legal business disclaimers:    http://www.CraneSoftwrights.com/legal