XML Schema EDC, UPA and substitution groups

[Soren Kuula]

Hi,

Does anyone happen to know whether this is legal:

<complexType name= ...>
<choice>
<element ref="foo"/>
<element ref="bar"/>
</choice>
</complexType>

<!-- nice enough so far... -->

<element name="foo" type=... substitutionGroup="bar"/>

<element name="bar" type=.../>

Problem: If bar is susbsted by foo, then EDC is violated.... hehehe --

Wonder is this is explicitly prohibited somehow?

Also, does anyone know why they required that the type of foo be derived
from that of bar, and is it true that it must be in ONE or more steps
(not zero or more?)

Soren

[Priscilla Walmsley]

Hi,

Yes, this is prohibited, by UPC and not EDC. It is discussed in:

http://www.w3.org/TR/xmlschema-1/#non-ambig

and

http://www.w3.org/TR/xmlschema-1/#cos-nonambig
It is _not_ true that it has to be at least one step. foo and bar can
have the same type.

Hope that helps,
Priscilla

----------------------------------
Priscilla Walmsley
Author, Definitive XML Schema
Definitive XQuery
http://www.datypic.com
----------------------------------

*** Sent via Developersdex http://www.developersdex.com ***

[Soren Kuula]

Hi, thanks Priscilla

Priscilla Walmsley wrote:

Yes, this is prohibited, by UPC and not EDC. It is discussed in:
Whoops, yes there is no EDC issue with that. UPC -- is that UPA ;) ?
It is _not_ true that it has to be at least one step. foo and bar can
have the same type.

Thought so .. ok I have a correction for my author friend on that one...

Soren