Re: [ubl] More code list questions about NDR, and about genericode

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

At 2005-10-17 07:09 +0100, Anthony B. Coates wrote:
>Ken, apologies I didn't reply earlier.

No apologies needed, Tony ... it took me three weeks to make the time 
to work on this.

>On Tue, 11 Oct 2005 22:27:20 +0100, G. Ken Holman
><gkholman@CraneSoftwrights.com> wrote:
>
>>    genericode ShortName  ==  UBL xxxxxCodeListID
>>    genericode LongName == UBL xxxxxCodeListName
>>    genericode Version == UBL xxxxxCodeListVersionID
>>    genericode CanonicalUri == UBL xxxxxCodeListURI
>>    genericode CanonicalVersionUri == UBL xxxxxCodeListSchemeURI
>
>Not quite.  Have a look at the last set of conversion stylesheets I sent
>out.

I have been.  I believe the package is 
genericode-and-xsd-and-sch-v2.zzz that I downloaded or received by 
email September 20:

   http://lists.oasis-open.org/archives/ubl/200509/msg00093.html

>UBL doesn't have an equivalent for "CanonicalUri", which is a URI
>that uniquely identifies the code list as an entity, independently of any
>specific version.  In the stylesheets, I have to create this by chopping
>the version off the version-specific URN for the code list.

So noted ... thank you I'll do the same.

The UBL-CodeList-CurrencyCode-1.0.xsd file in that ZIP package of 
yours appears not to match the one found in UBL 1.0 CD2:

   http://docs.oasis-open.org/ubl/cd-UBL-1.0/xsd/codelist/UBL-CodeList-CurrencyCode-1.0.xsd

>>Tony, can you confirm this is all of the version checking that I need to
>>do when using a genericode file to express the values in a UBL instance?
>
>I believe so; again, see the stylesheets I did, as I had everything there
>that was needed to match Marty's example Schemas.
>
>>(1) for @xxxxxID remove "ID" and suffix with "Code..." from genericode
>>table above as appropriate
>>(2) for @xxxxxCode remove "Code" and suffix with "Code..." from
>>genericode table above as appropriate
>>(3) for <xxxxx> suffix with "code..." from genericode table above as
>>appropriate
>
>I also did a Schematron Schema to check these, bundled in with the
>stylesheets.

Fine .....

>Did you see that?

What I see is that your scripts produced your output from your 
version of the code list schema expression, but when I run your 
scripts from the UBL-1.0-CD2 version of the currency code list schema 
expression I do not get the same results.  The resulting Schematron 
script does no checking of the code list meta data information of 
what I'm looking for ... only value checking against the enumeration.

I'm trying to work from the published code list expressions.

Whereas the genericode expression from your version of the currency 
code list has a bunch of application information embedded, the 
genericode expression I create from the published version of the 
currency code list does not have any application information embedded.

I then see that your GC->SCH scripts are relying on the application 
information in order to synthesize the business rule checks of the 
code list metadata.

>I don't remember needing to do this,
>though.  The only open issue I had was if someone used a code list type to
>create a differently named element.  That would require either (i)
>Schematron using a Schema-aware version of XPath 2.0 that checks types, or
>(ii) the user would have to create their own Schematron rule based on the
>supplied one (not a big job).

I agree ... not a big one, and my concept of the XML instance agreed 
upon between trading partners accommodates this as it enumerates all 
of the information items that use a particular code list.

I'm going to try and follow through with the munging that I planned, 
modified by your comment about the canonical URI:

    genericode ShortName  ==  UBL xxxxxCodeListID
    genericode LongName == UBL xxxxxCodeListName
    genericode Version == UBL xxxxxCodeListVersionID
    genericode CanonicalUri == UBL xxxxxCodeListSchemeURI substring 
with no version
    genericode CanonicalVersionUri == UBL xxxxxCodeListSchemeURI

but I think I might just abandon the substring and go with the 
version URI only of the two URIs since the URI is being checked 
already in its entirety.

In the above "xxxxx" comes from attributes and elements as noted in 
my earlier post:

(1) for @xxxxxID remove "ID" and suffix with "Code..." from 
genericode table above as appropriate
(2) for @xxxxxCode remove "Code" and suffix with "Code..." from 
genericode table above as appropriate
(3) for <xxxxx> suffix with "code..." from genericode table above as 
appropriate

However, I'll be working from the UBL 1.0 CD2 code list expressions 
since I don't know where you got your code list expression.

I do note that the genericode <ShortName> element created by your 
scripts is empty when running the XSD to genericode scripts on the 
UBL 1.0 CD2 code list expressions, so we'll have to check this when 
it comes to doing all this again with UBL 2.0.

Thanks, Tony!

. . . . . . . . . Ken



--
World-wide on-site corporate, govt. & user group XML/XSL training.
G. Ken Holman                 mailto:gkholman@CraneSoftwrights.com
Crane Softwrights Ltd.          http://www.CraneSoftwrights.com/o/
Box 266, Kars, Ontario CANADA K0A-2E0    +1(613)489-0999 (F:-0995)
Male Cancer Awareness Aug'05  http://www.CraneSoftwrights.com/o/bc
Legal business disclaimers:  http://www.CraneSoftwrights.com/legal