OASIS Mailing List ArchivesView the OASIS mailing list archive below
or browse/search using MarkMail.

 


Help: OASIS Mailing Lists Help | MarkMail Help

ubl message

[Date Prev] | [Thread Prev] | [Thread Next] | [Date Next] -- [Date Index] | [Thread Index] | [List Home]


Subject: First draft UBL 1.1 schemas


Hello TC,

please find in the attachement the first draft of UBL 1.1 schemas based on
UBL 1.0 models.


List of requested changes:
==========================

C1 Implementation of DOC8
-------------------------
Please look into the CoreComponentTypes, Unspecialized Data types,
Specialized Data type and Codelist schemas.

Differences between the schemas and the rule:

The rule says there should be an annotation "Name", but in the schemas there
is an annotation "DictionaryEntryName" because it is mandatory in the
CoreComponentParameter schema.

In addition there is an annotation "ComponentType" because it is also
mandantory in the CCP schema.

The rule mentions an optional item "Possible Value(s)" but there are no
possible value in the schemas. To fill this item I need some input from the
TC.

I enhanced the CoreComponentParameter schema by adding the elements
"PrimitiveType" and "PossibleValue".


C3 New Rule CTD20 states:
-------------------------

"For every property identified in the UBL model a named XSD:ComplexType must
be defined"

This was already implemented in the UBL 1.0 schemas. For every combination
"PropertyTerm" + "RepresentationTerm" there is an xsd:complexType and a
corresponding xsd:element.


C5 1.1 Schema layout required by the currently amended rule GSX1
     with added agreements/comments in square brackets and
     with a), b), etc denoting subsections for clarity
----------------------------------------------------------------

This rule is implemented.



Other changes
=============

OC1
---
I added space trimming for the "fixed" attribute of Supplementary
Components.

OC2
---
The rule CTN2 concerning the names of complexTypes for BBIEs is not correct
implemented in the 1.0 schemas. The rule says that the propertyTerm and its
qualifiers and the representation term go into the name of the complexType
for an BBIE. In the UBL 1.0 schemas the name of the complexType is only
based on the propertyTerm and representationTerm without considering the
qualifiers.

So I created 2 sets of schemas. One set "CTN2 correct" and "CTN2 wrong". The
instances won't be affected by this change, but I think the TC should decide
if we use the correct implementation from now on.

OC3
---
In the attachement you can also find the spreadsheets and code list files I
used for the schema generation. The code list files contain now all codes
that were used for the UBL 1.0 schemas.



Problems
========
I imported all spreadsheets beside the ones for the CoreComponentTypes and
the UnspecialiedDataTypes into EDIFIX, but there is no machine readable way
to restrict automatically the available codes for "UBL_ Amount Currency.
Identifier" and to add the information about code list name and agency. So I
had to do this by hand in EDIFIX.



Best Regards,



David




















spreadsheets.zzz

schemas.zzz



[Date Prev] | [Thread Prev] | [Thread Next] | [Date Next] -- [Date Index] | [Thread Index] | [List Home]