OASIS Mailing List ArchivesView the OASIS mailing list archive below
or browse/search using MarkMail.

 


Help: OASIS Mailing Lists Help | MarkMail Help

ebxml-core message

[Date Prev] | [Thread Prev] | [Thread Next] | [Date Next] -- [Date Index] | [Thread Index] | [Elist Home]


Subject: RE: Tag Languages, UID's etc.


Robert Miller wrote:

When we
> exchange documents, his computer will reference his Schema 
> (stored in an
> ebXML repository next to mine perhaps), and my computer will 
> use his Schema
> to get at the UID's that tell my computer what to do.

That was one of my points, you have to go fetch some other resource to determine the interpretation of the document, you have to trust that the author of the resource isn't trying to attack your
system and you have to trust that resource isn't being changed to redefine what initially appeared to be an order to be a quotation or vice-versa.

If XML Schema or DTD validation will be used (assuming no prevalidation transformation), then the namespace qualified tag name is going to be the key to associate an element with the appropriate
schema resource.  To let the tag name be a meaningless comment and have an attribute value be the definitive key to the interpretation of the element means that you have to reinvent your own
validation means or do some prevalidation transform.

If there are going to be an unlimited number of synonyms that could be used in an incoming message, the first thing that would typically be done is to convert the message into a canonical form and
then have the interior code work on that canonical form.

For example, if I got something like

<FinancialAccountIdentifier UID="1234567"/> and
<GermanFinancialAccountIdentifier UID="1234567"/>

My first action would be to convert that into some canonical representation and then performance validation disgarding the original tag name since it was just a meaningless comment.  The result of
this transform for both of these fragments could be:

<_1234567 ..../>

or

<FinancialAccountIdentifierInOneSpecificHumanLanguage.../>

However, since it is very easy for a analyst to apply a transform to create a locale-specific format from a canonical format, why the insistance to pass around localized messages from system to
system?

Anyway, I'm going skiing and said my two cents.


[Date Prev] | [Thread Prev] | [Thread Next] | [Date Next] -- [Date Index] | [Thread Index] | [Elist Home]

Search: Match: Sort by:
Words: | Help


Powered by eList eXpress LLC