- From: Eric van der Vlist <vdv@dyomedea.com>
- Date: Wed, 06 Feb 2002 09:34:41 +0100
- To: "Henry S. Thompson" <ht@cogsci.ed.ac.uk>
- Cc: Jeni Tennison <jeni@jenitennison.com>, xmlschema-dev@w3.org
Hi Henry,
Can we come back on this old email where we have different appreciations
on you answer ;=) ?
Henry S. Thompson wrote:
> Eric van der Vlist <vdv@dyomedea.com> writes:
>
> <snip/>
>
>>but what about:
>>
>><xs:redefine schemaLocation="bar.xsd">
>> <xs:group name="foo">
>> <xs:sequence>
>> <xs:sequence minOccurs="0">
>> <xs:group ref="foo"/>
>> </xs:sequence>
>> <xs:element ref="bar"/>
>> </xs:sequence>
>> </xs:group>
>></xs:redefine>
>>
>>which is equivalent?
>>
>
> Good point -- worth a clarification.
>
>
>>And even your:
>>
>>
>>
>>><xs:redefine schemaLocation="...">
>>><xs:group name="foo">
>>>
>>> <xs:choice>
>>> <xs:element ref="C" minOccurs="0" />
>>> <xs:group ref="foo" />
>>> </xs:choice>
>>></xs:group>
>>></xs:redefine>
>>>
>>has the effect to allow a content where none of the content of "foo" is
>>mandatory which seems contrary to (what I have understood of) the philosophy
>>of the redefinition which is to avoid that the applications are "too much
>>surprised" by the redefinition.
>>
>
> Also true.
Does your answer (Also true) mean that my interpretation is right and
that this should be avoided if nor forbidden or does it mean that it's
true that the schema rec allows them?
Thanks
Eric
>
> ht
>
--
Rendez-vous a Paris pour mes formations XML/XSLT.
http://dyomedea.com/formation/
------------------------------------------------------------------------
Eric van der Vlist http://xmlfr.org http://dyomedea.com
http://xsltunit.org http://4xt.org http://examplotron.org
------------------------------------------------------------------------
Received on Wednesday, 6 February 2002 03:35:21 UTC