I actually pulled the RNG schemas out of the UBL 2.2 release as we
had no evidence that they were ever used in UBL 2.1:
http://docs.oasis-open.org/ubl/os-UBL-2.1/UBL-2.1.html#S-UBL-2.1-RELAX-NG-SCHEMAS
This was consistent with the concept of publishing multiple
alternative non-normative representations of the normative semantic
library and normative XML schemas. The semantic library is described
using the Core Component Technical Specification (CCTS) 2.01 from
UN/CEFACT. UBL uses the OASIS Naming and Design Rules to create
various schema semantic expressions of the document constraints.
And so I also pulled the ASN.1 schemas out of UBL 2.2, but we still
have demand for those, and so those are now separately published (and
currently under review):
https://lists.oasis-open.org/archives/ubl/201902/msg00020.html
As there has been zero demand for us to publish UBL 2.2 in RELAX-NG,
I think I was justified in removing it. We can always publish a new
committee note if needed.
This semantic library in CCTS approach also allows us to publish a
JSON alternative representation of the UBL semantics, and this also
has been used and we have had good feedback ... the latest for
Draft07 of JSON is under review:
https://lists.oasis-open.org/archives/ubl-ndrsc/201811/msg00000.html
What is different about NVDL is that I could include the one NVDL
fragment in each of the xsd/ and xsdrt/ subdirectories of the
distribution in order to despatch the validation to the normative XSD
schemas. That "feels" different to me than offering an alternative
set of validation semantics.
. . . . . Ken
At 2019-03-02 17:45 +0100, Imsieke, Gerrit, le-tex wrote:
>Content-Transfer-Encoding: 7bit
>
>On 02.03.2019 17:29, G. Ken Holman wrote:
>>Thank you!
>>Your thinking of alternatives also leads me to consider including
>>an NVDL schema in the distribution. Then users can use that one
>>schema to validate any XML document.
>>A quick search, though, doesn't help me find any examples of
>>command-line NVDL processors that I could include as a JAR file in
>>the UBL distribution. The 20151127 jing that I found appears to
>>take in only an RELAX-NG argument as input.
>
>Yes. I believe the jing version that you mentioned and maybe a more
>up-to-date version that is also maintained/patched by the oXygen
>folks is the only viable NVDL validator available. I think it does
>neither XSD nor ISO Schematron validations. Of course you can try to
>convert all XSDs to RNGs. We are using some
>msv-rngconverter-2013.2.jar for it, but it doesn't always work out of the box.
>
>Gerrit
--
Contact info, blog, articles, etc. http://www.CraneSoftwrights.com/x/ |
Check our site for free XML, XSLT, XSL-FO and UBL developer resources |
Streaming hands-on XSLT/XPath 2 training class @ US$45 (5 hours free) |
