Re: [ubl-comment] Duplicated fields in Genericode file

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

On further inspection, the bug is not revealed in the "putative 
release" from September 23 as I said because I opened the wrong file 
on my system!  It turns out that my rewrite of the environment to be 
automated produces cleaner files than what I used in the first public review.

Looking back at that csprd01 distribution (which was implied in your 
original post but I was distracted by my recent development) I find 
the fault you mention here in this file:

http://docs.oasis-open.org/ubl/csprd01-UBL-2.2/cl/gc/default/UnitOfMeasureCode-2.2.gc

The putative package cited earlier has been replaced with a new 
code-lists-ubl-2.2-csd02wd02-20170925-2330z.zip package and so the 
next public review will have these clean files:

https://www.oasis-open.org/committees/document.php?document_id=61646

The code list generation environment I use now has been updated to 
incorporate both W3C Schema and Schematron validation passes on the 
generated code list files.

Thank you, again, David, for bringing to light the need for 
validating the code list files before distribution.

. . . . . . Ken

At 2017-09-25 17:09 -0400, G. Ken Holman wrote:
> Thank you, David, for bringing this to the committee's attention!
>
> At 2017-09-25 21:09 +0100, David Goodenough wrote:
> > In the UBL 2.2 distribution ZIP file there is a file:-
>
> I'm assuming you are referring to the development "putative release" 
> recently posted, because I can find the problem you've reported in 
> the files cited from my post here:
>
>   https://lists.oasis-open.org/archives/ubl/201709/msg00042.html
>
> Indeed, there are a number of entries with multiple description 
> fields.  I can trace this to the source XML because there are 
> multiple paragraphs in the Excel table cell for those Unit of 
> Measure entries.  The code is triggering on the paragraphs, not on the cells.
>
> > The description of Genericode does not specify what should happen when
> > a columnref is duplicated in a row - should the values be concatenated
> > or should the duplicate be ignored?
>
> The duplicate should be reported as a bug to the programmer!  I feel 
> sufficiently chastised, thank you.  You did the right thing by reporting it.
>
> Accordingly, I will incorporate a two-stage validation of the 
> generated genericode files:  the first pass will use the XSD schema, 
> and I'll write a genericode set of Schematron assertions to confirm 
> issues such as duplicate columns and absent required columns.  This 
> should allow reviewers such as yourself to focus on the content and 
> not the structure of the files.
>
> Thank you, again, for your support of our work by taking the time to 
> report this!  And it is timely so that I can get the fix in for our 
> next public review.
>
> . . . . . . . . Ken


--
Contact info, blog, articles, etc. http://www.CraneSoftwrights.com/o/ |
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) |