RE: [ubl-dev] Re: Code List Value Validation

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

At 2006-04-11 07:46 -0700, David RR Webber \(XML\) wrote:
>Ok - understood - I'll post some suggestions and see what people make of
>them.  I'm not seeing anything  catastrophic here - mostly imcremental
>enhancements!

Well, I stand by my suggestion that giving enhancements to Tony will 
be less productive than giving requirements to Tony ... in the last 
post and this post you've been talking about designing your own 
changes without indicating the requirements not being met that need 
changes to fix.

As I said, Tony owns genericode and it is his considered work and 
effort in meeting code list requirements for a number of projects, 
not just UBL.  When I identified a requirement I had that wasn't 
being met with genericode, I had to convince him it was a 
sufficiently useful requirement to risk tinkering with any part of 
the genericode vocabulary.

>I saw your note that the parsing in xslt was slow - and you had to
>switch to SAX/python - so I thought I'd suggest some cosmetic changes
>that have worked before for me to improve speed - and have side benefit
>of improving the human side - this is mostly just re-jiggering - very
>easy to do with an xslt globally.

But I didn't say that the speed issue had anything to do with 
genericode files.  You'll see from the context of my comment that I 
was dealing with the XSD schema files for UBL 2.0:

At 2006-04-10 22:03 -0400, I wrote:
>Absolutely ... I did (though straightforward it was a bit of a challenge):
>
>    http://www.oasis-open.org/archives/ubl/200602/msg00069.html
>
>The "Garden of Eden" approach to the UBL NDR requires every element 
>and data type be global ... because of that I was able to use XSLT 
>to process the schema expressions.  I've successfully generated the 
>XPath files for all of the document types, and I used the XPath file 
>results to synthesize the above single "default" context association 
>file for UBL 2.
>
>BTW, XSLT was so slow at some of the processes, I ended up rewriting 
>some of the steps in Python/SAX.

The speed issue was dealing with 350Mb intermediate XPath files 
producing 600Mb HTML files ... XSLT was (expectedly) slow, but I had 
to prove to myself that the algorithms were correct (the file sizes 
floored me!).  A SAX process works without a footprint of the input 
file size, and the end result ran in a fraction of the time, and 
proved to me that the file sizes were, indeed, as large as produced 
by the slow XSLT.  The end result was 9.6Gb of text and HTML:

   http://www.oasis-open.org/archives/ubl/200601/msg00065.html

I certainly didn't say there were any issues of XML vocabulary design 
for genericode that XSLT could not handle.  No speed tweaking is 
needed for those files as a result of the work I've been doing.

At 2006-04-11 07:46 -0700, David RR Webber \(XML\) wrote:
>Also - we'd had a thread on the possibly applicability of registry - so
>wanted to make sure if we need to add a tag or two to support use of
>registry - we explore that now - before this train is getting too far
>down the rails and then someone says - why did you not consider that?

Then I will leave that to your expertise as I don't know enough of 
registry to see any role it has in the steps of the methodology I've 
proposed ... I based my methodology on transformation and ISO/IEC 
19757-3 Schematron.

I look forward to your input, David,

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

--
Registration open for XSLT/XSL-FO training: Wash.,DC 2006-06-12/16
Also for XML/XSLT/XSL-FO training:Birmingham,England 2006-05-22/25
Also for XSLT/XSL-FO training:    Copenhagen,Denmark 2006-05-08/11
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/u/
Box 266, Kars, Ontario CANADA K0A-2E0    +1(613)489-0999 (F:-0995)
Male Cancer Awareness Aug'05  http://www.CraneSoftwrights.com/u/bc
Legal business disclaimers:  http://www.CraneSoftwrights.com/legal