- From: Eric van der Vlist <vdv@dyomedea.com>
- Date: Mon, 02 Jul 2001 19:43:24 +0200
- To: "WATKIN-JONES,ADAM (HP-UnitedKingdom,ex1)" <adam_watkin-jones@hp.com>
- Cc: "'xmlschema-dev@w3.org'" <xmlschema-dev@w3.org>
"WATKIN-JONES,ADAM (HP-UnitedKingdom,ex1)" wrote:
>
> Hello,
>
> My head is sore from bashing it against a brick wall so I hope someone can
> help me with this!
I think I can :=) at least to point a couple of issues...
>
> Below is some XML, a schema and the output from XSV when I try to validate.
>
> I am trying to express a key/keyref constraint but I always get an XSV
> error. I have tried altering the position of the xsd:keyref and xsd:key
> elements in case it is some sort of scoping problem but with no success.
Your scoping is good and I suspect a more trivial reason!
.../...
> <xsd:key name="mykey">
> <xsd:selector xpath="k1:keys/k1:key"/>
> <xsd:field xpath="@id"/>
> </xsd:key>
>
> <xsd:keyref name="dummy1" refer="k1:mykey">
If I change this to:
<xsd:keyref name="dummy1" refer="mykey">
XSV doesn't complain any longer.
Now, what's weird is that the rec mentions refer as a QName:
<keyref
id = ID
name = NCName
refer = QName
{any attributes with non-schema namespace . . .}>
Content: (annotation?, (selector, field+))
</keyref>
OTH, since key and keyrefs need to be defined side by side in the same
element, I wonder what is the reason for allowing a QName here since I
don't see how you could refer a key defined for another namespace.
The rec also says that the refer should be QName resolved and my bet
would be that you are right and that it's a SCV bug.
I am still confused, though ;) ...
Hope this helps.
Eric
--
See you at XTech in San Diego.
http://conferences.oreillynet.com/cs/os2001/view/e_spkr/790
------------------------------------------------------------------------
Eric van der Vlist http://xmlfr.org http://dyomedea.com
http://xsltunit.org http://4xt.org http://examplotron.org
------------------------------------------------------------------------
Received on Monday, 2 July 2001 13:43:51 UTC