Re: [ubl-ndrsc] range of valid values for cc types

["Stephen Green" <stephen_green@seventhproject.co.uk>]

Hi

Please may I add to Anne's question the matter of:

Would there be any need to specify *how* (leading
sign, trailing sign, etc) a negative amount is
specified as a separate BIE to ensure correct
interpretation of the negative aspect? Is the
expression of negativity controlled enough so as to
be clear and unambiguous?

This is similar to
another issue - Do we need precision BIE(s) to
allow unambiguous expression of a) decimals
(separator = . or , or ...etc) b) 1000s etc separators
- either adding separate BIE(s) for decimal places
(might have to instead find a way to associate this
kind of data with each amount or currency), etc.

(Behind this is the wide range of uses worldwide of
different ways to express decimals, thousands or
tens of thousands and positive/negative sign.)

Perhaps this value metadata should be attributes of an
amount (particularly decimal places and/or separator
and pre-decimal number of digits and/or thousand (etc)
separator and ,as my original question asks, whether
negative or positive and/or leading or trailing sign (etc))

then

Is it catered for by either
a) something in the CCT Spec
to limit how the above are expressed or indicated and/or
b) something in the CCT supplementary components or
xsd base type?

If not, can we add supplementary components for these of
our own?
Or should we be thinking of adding BIEs?
Or should we just leave it for implementers? (???)

I guess it applies too to cct:Quantity and cct:Measure as
well as to cct:Amount.

Thanks

Steve
----- Original Message -----
From: "Anne Hendry" <anne.hendry@sun.com>
To: <ubl-ndrsc@lists.oasis-open.org>
Sent: Tuesday, February 03, 2004 7:08 AM
Subject: [ubl-ndrsc] range of valid values for cc types


> Hi,
>
> In a recent Libarary Content meeting there was a question as to whether
> negative values could be valid values for some entities, as in the case
> of an adjustment amount.  In looking at
> UBL-CoreComponentTypes-1.0-beta.xsd and
> UBL-CoreComponentParameters-1.0-beta.xsd (from UBL 1.0 Beta), I noticed
> these types defined:
>
> cct:AmountType based on xsd:decimal
> cct::DateTimeType based on xsd:dateTime
> cct:IdentifierType based on xsd:decimal
> cct:MeasureType based on xsd:decimal
> cct:NumericType baed on xsd:decimal
> cct:QuantityType based on xsd:decimal
>
> rt:DateTimeType based on cct:DateTimeType
> rt:DateType based on xsd:date
> rt:TimeType based on xsd:time
> rt:MeasureType based on cctMeasureType
> rtNumericType based on cct:NumericType
> rt:ValueType based on cct:NumericType
> rt:RateType based on cct:NumericType
> rt:QuantityType based on cct:QuantityType
>
> Since all the base types used by cct noted above are xs primitive types
> that allow negative values, and I can't see anywhere in either of the
> above files where there are any restrictions in terms of
> positive/negative value ranges (eg. no use of derived types such as
> PositiveInteger, nonNegativeInteger, unsgined*, etc), is it safe to
> assume the range of valid values for UBL entities based on these CC
> types includes the full range of values as allowed by the w3c schema
> primitive types on which they're based?
>
>
> Thanks,
> Anne
>
>
> To unsubscribe from this mailing list (and be removed from the roster of
the OASIS TC), go to
http://www.oasis-open.org/apps/org/workgroup/ubl-ndrsc/members/leave_workgro
up.php.
>
>