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

 


Help: OASIS Mailing Lists Help | MarkMail Help

ubl-lcsc message

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


Subject: [ubl-lcsc] [Fwd: Re: [ubl] UML diagrams of xCBL OrderDetail]



here is an update from dave carlson about the UML diagrams he developed (and an additional one for Team One).

can i ask Gunther (who raised the question) if he could liaise directly with dave (dcarlson@ontogenics.com) on the matter of  discrepencies with the model and xCBL3.0

the matter of "implied logical relationships" that may be missing should be agenda item for next week's call.

-------- Original Message --------
Hi Tim,
First of all, I've attached the OrderSummary model.  It is a very managable size and relatively quick to produce and format.  You had not previously requested this one.  I'll push the OrderHeader models up on my priority list (I'm trying to refocus on activities that will generate revenue).
 
Can you provide at least one specific example of "implied logical relationships" that are missing?  I am using version 3.0 of the SOX schemas for xCBL.  When I produced the models for Catalog and TradingPartner, I used my UML model to generate XSD schemas, which I then used to successfully validate the test documents that ship with xCBL 3.0.
 
I reviewed the OrderSummary diagram against the xCBL html docs and saw one item that was not on the diagram (although it was in the underlying UML model).  The "number" superclass was not on the diagram, representing the simple content type (in XSDL terms) of three elements.  I added it to the diagram.  In other more complex diagrams, they would get very cluttered to show all simple type classes (e.g. string, number, etc.).
 
I'm aware of one other subtle difference in the logica model: if a SOX element has a named association to a ListOfXXX element, I don't show both the role name and the ListOfXXX when linked to the target class.  I do, however, show the "1..*" or "0..*" multiplicity, so one would argue that the *logical* model is still accurate.  This situation is pretty rare in xCBL.
 
Because the xCBL schemas are not split into logical modules (e.g. Core, Order, etc.) I created several of the module files to assist the reverse engineering process.  So I might have missed a small number of misc elements, but I'm not aware of any.  So, examples of missing elements are appreciated.
 
 
- 
regards
tim mcgrath
fremantle  western australia 6160
phone: +618 93352228  fax: +618 93352142 

OrderSummary.pdf



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


Powered by eList eXpress LLC