Re: [ebxml-bp] Regarding the alleged Bad V2 Schema Woes

["David RR Webber" <david@drrw.info>]

Dale,

Oh crap!  I see - its an ephemereal structure - just add bits
to replace others as you go along ; -)

Yeah - Spy should really have an alert warning system for
denoting substitution groups!

However - its still flakey that the schema parser itself throws
an error and complains that the ReceiptAcknowledge are
missing *required* elements - when in fact its a substitution group!?

Gasp.

So - am I right in thinking that these elements are required - if you state
that ReceiptAcknowledge is "true" - but otherwise they are not-required?

That would make more sense here...

Thanks, DW

----- Original Message ----- 
From: "Dale Moberg" <dmoberg@cyclonecommerce.com>
To: "David RR Webber" <david@drrw.info>; <martin.me.roberts@bt.com>;
<ebxml-bp@lists.oasis-open.org>
Sent: Friday, May 21, 2004 1:02 PM
Subject: RE: [ebxml-bp] Regarding the alleged Bad V2 Schema Woes



DRRW says:
Then when I realized this - I went back into Spy to see what was
*really* going on.  The big problem I had was that the visualization
Spy draws is missing key aspects of the structure!  Specifically
the ReceiptAcknowledgement stuff - so by looking at the raw
XSD I could see the ReceiptAcknolwedgement elements - but
since Spy did not display them I was puzzled - are they needed,
are they attributes or elements, was it a cut and paste error
in editing the schema?


Dale>

OK I think I can help here, David. The ReceiptAcknowledgement element
does not have global visibility. [I often work in text mode on this
stuff so I wasn't grokking what worried you.]

What is available is a type definition for "ReceiptAcknowledgementType"

Then in Martin's specialized new improved (!! my emphasis)
BusinessTransaction variants, such as "CommercialTransaction" you will
find a local element declaration:

<xsd:element name="ReceiptAcknowledgement"
type="ReceiptAcknowledgementType"/>

That is OK and actually has advantages rather than using, for example, a
ref="global element". Another substitution group element for
BusinessTransaction could have a ReceiptAcknowledgement local element
and it could be of a derived or even a brand spanking new type.

I know you hanker for a rng schema version of all this, freed from the
gratuitous complexity and nine-ways-to-do-anything of xsd schemas, but I
am afraid this is just typical for xsd.

But so far I have not found a mistake. You won't be able to see the
ReceiptAcknowledgement in the graphical list because it is not a global
element.
You should see a named ReceiptAcknowledgementType in your list though.

Hope that helps on that bit.