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


Help: OASIS Mailing Lists Help | MarkMail Help

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [xml-dev] problem in using schema for schemas


Takuki Kamiya wrote:

> Hi,
> I have encountered a problem in using schema for schemas for XML
> schema pre-processing.
> I am using schema for schemas for pre-processing an XML schema
> for syntax checking and possibe default-value contribution
> before turing it over to my downstream schema processor.
> The problem is that schema for schemas always contributes some
> attribute values to certain elements (if none was present) which XML
> Schema specification explicitly prohibits in some context.
> For example, <element> under <complexType> or <group> may have
> "ref" attribute. In such a case, for example, "nillable" property
> is not allowed to be present as per section 3.3.3. However, schema
> for schemas apparently not only tolerates it (which is actually ok),
> but also suggests to contribute default value "false" if none was
> sepcified.

This is a good catch and an illustration of the limitations of what you 
can describe with W3C XML Schema.

Here, a language which would allow the definition of two elements 
"element" with different content models under "group" would have been 
definitely useful.

Unfortunately, it is not possible with W3C XML Schema and they've done 
their best with what they had at hand!

The other alternative (better for refs but much worse for anonymous 
declarations) would have been to remove the default value for nillable...

> I am not sure whether schema for schemas is intended for possible
> use for pre-processing schemas (which I think is useful), but it
> would be better if the PSVI can be correctly used for downstream
> schema processing.
> I could modify schema for schemas by myself locally, but noticed
> it is normative. I think those default values that clashes with
> explicit constraints had better be taken out from the schema for
> schemas if there is any revision planned.

Again, I don't think removing the default value would be any better.

What about using a RELAX NG schema for W3C XML Schema instead ;) ?



> -Takuki Kamiya

Rendez-vous  Paris pour le Forum XML.
Eric van der Vlist       http://xmlfr.org            http://dyomedea.com
http://xsltunit.org      http://4xt.org           http://examplotron.org