[ubl-dev] Test Assertions for UBL Calculation Model?

[Stephen Green <stephengreenubl@gmail.com>]

---------- Forwarded message ----------
From: Stephen Green <stephen.green@documentengineeringservices.com>
Date: 2009/7/1
Subject: Re: [ubl-dev] Test Assertions for UBL Calculation Model?
To: ubl-dev@lists.oasis-open.org, ubl-comment@lists.oasis-open.org,
"G. Ken Holman" <gkholman@cranesoftwrights.com>


One thing we notably lack with these Test Assertions
for a proposed UBL 2 Calculation Model is a set of
references to some approved rules for the calculations.

I thought we could put something on the Wiki in such a
way that we can make at least provisional references to
individual, atomic rules (perhaps like those which have
placeholders or vague references in these TAs). Then I
noticed that the same approach has been put forward in
OIC where they are undertaking a similar task with TAs,
albeit on a higher scale of magnitude. So maybe, while
we seek an ultimately TC-approved (or maybe even just
HISC-approved) set of calculation rules for UBL, we
could start to formulate such rules on the wiki in a way
which can already be referenced in draft test assertions,
then publish such a draft set of test assertions before
finally publishing the rules in a more permanent fashion
and publishing the corresponding test assertions too
with references to the fixed rules once they are agreed.

All this would take time and any software changes required
could be introduced over the medium to long term (compare
this with something like the introduction of the WS-I profiles
for web services or the ACID tests for web browsers).

Best regards

Stephen D Green



2009/6/26 Stephen Green <stephengreenubl@gmail.com>:
> Here is a more mature/complete calculation model
> represented in unapproved draft (0.4) OASIS Test
> Assertions Guidelines TC (TAG TC) test assertions
> markup.
>
> Hopefully the calculation 'rules' can be extracted in
> a more human-readable format (hopefully on wiki for
> HISC Calculation Model within a short while, time
> permitting).
>
> http://wiki.oasis-open.org/ubl/Example_Calculation_Models/Invoice_Tax_Total
> http://lists.oasis-open.org/archives/ubl-dev/200906/msg00065.html
>
> I do hope this can eventually mature to the point it
> can be formally approved by the UBL TC.
>
>
> I'll attach the XML file as an attachment too (but I'm
> not sure whether the list server will allow it through)
>
> Note: the 'permitted' value for 'PrescriptionLevel' reflects
> same degree of ambiguity within the definitions and between
> those and the examples/accepted practice; plus the
> conformance definition being in terms of schema-validity
> such that there may be several conformant calculation
> models, though I am here seeking the best practise
> one which I believe is implicit in the UBL Invoice model.
>

ubl-ta-draft-0-4.zip