Re: [xml] validation fails for embedded RELAX NG schema

[Michael Ludwig <mlu as-guides com>]

Jerzy Jalocha N schrieb:
> On Mon, Aug 3, 2009 at 8:34 AM, Michael Ludwig<mlu as-guides com>
> wrote:
> > I'd be surprised to discover that embedding RNG schemas in arbitrary
> > documents is actually supported. What makes you think it is?
>
> I am actually absolutely not sure about this. [...] I just took the
> following information, and put it together:
>
>  * It is valid from the XML point of view (as you agree with me).

Strictly speaking, it is well-formed, but not valid - validity requires
validation against a DTD, or another schema. But I know what you mean.

>  * And xmllint copes perfectly with the opposite case, of arbitrary
> content embedded in the RELAX NG schema:
>
> <?xml version='1.0' encoding='UTF-8'?>
> <grammar xmlns='http://relaxng.org/ns/structure/1.0'>
>    <embedded xmlns='http://namespace.net/something'>
>       <!-- Arbitrary content embedded here. -->
>    </embedded>
>    <start xmlns='http://relaxng.org/ns/structure/1.0'>

Haven't checked, but that is probably permitted by the Relax NG
specification.

> Adding these two pieces of information, It seems reasonable, or at
> least probable, to embed the whole schema in an arbitrary XML
> document.

Consider that you could embed C source code in an XML document. Would it
be reasonable to assume that doing so would automagically invoke the
compiler?

On a sidenote, XSLT has a mechanism to convey an XSLT meaning to a
document other than <xsl:stylesheet> :

http://www.w3.org/TR/xslt#result-element-stylesheet

So your assumption about Relax NG is not absurd in principle; any such
auto-mechanism just typically has (and probably needs) a more thorough
definition.

(I seem to remember, though, that the Perl interpreter has a flag for
allowing execution of Perl code from text such as an email. Go figure.
In eight years, I think I've never used that -x flag.)

Michael Ludwig