RE: [xml-dev] If I want to use catalogs for XSDs, must document instance

["contact" <contact@xml-tools.com>]

Hi Graham,

> <uri name="http://www.atlassian.com/schema/confluence/4/ac/";
uri="confluence.xsd"/>

in XML ValidatorBuddy you have a single command to add any W3C schema with a
target namespace to the XML catalog to create such an <uri> entry for any
following validation. There is also no need to provide a schemaLocation
attribute in the XML instance. The schema mapping takes place when the
namespace is used.

I guess this is close to how you would expect it.

Kind regards
Clemens Uhlenhut
http://www.xml-tools.com

 

________________________________________
From: Graham Hannington [mailto:graham_hannington@fundi.com.au] 
Sent: Freitag, 20. April 2012 09:33
To: xml-dev@lists.xml.org
Subject: Re: [xml-dev] If I want to use catalogs for XSDs, must document
instances contain a schemaLocation?

Liam,

Thank you very much for your quick and informative response, it's much
appreciated.

Regarding your statement:

> doing it based on namespace URIs would not be a good way

I hope you don't think I'm splitting hairs; I sincerely want to understand
this...

Here's how I think it works. When an XML document instance contains the
following schemaLocation attribute:

xsi:schemaLocation="http://www.atlassian.com/schema/confluence/4/ac/confluen
ce.xsd"

my understanding is that catalog-aware XML applications (such as XMLSpy)
look at the first part of that attribute value*, match it with the value of
the name attribute of the following uri element in the catalog:

<uri name="http://www.atlassian.com/schema/confluence/4/ac/";
uri="confluence.xsd"/>

and then load the XSD referred to by the uri attribute
(uri="confluence.xsd").

* From the "XML Schema Part 0: Primer Second Edition"
(http://www.w3.org/TR/xmlschema-0/#schemaLocation)

> The schemaLocation attribute value consists of one or more pairs of URI
references, separated by white space. The first member of each pair is a
namespace name

That is, although an XML application isn't doing it based on the value of
xmlns: attributes (as you say: "Neither should it."), it seems to me that it
*is* doing it based on namespace URIs (specifically, the first member of
each pair of URI references in a schemaLocation attribute value).

Am I just splitting hairs, or am I (as I suspect) missing a deeper point
here?

Also this:

> There is no direct mapping between namespace URIs and Schema documents

Now - with apologies, I do not mean to try your patience or waste your time;
I do read and re-read the related W3C documents, and some of it has sunk in
- I really think I am missing a deeper point here, because, as described
above, I think a direct mapping between a namespace URI and a schema
document is exactly the process that I've just described above, and is
exactly what that uri element in a catalog does:

<uri name="http://www.atlassian.com/schema/confluence/4/ac/";
uri="confluence.xsd"/>

If this uri element does not directly map a namespace URI to a schema
document, what does it do?

Finally:

> the relationship is that Schema documents may define elements associated
with one or more namespace URIs.

I'm not saying that you're asserting otherwise, but I want to confirm that
my current understanding in this area is correct. I think that one schema
document (XSD; a single .xsd file) can define elements associated with only
one namespace URI (its target namespace). An XSD (again, I mean,
specifically, a single .xsd file) can use <import> elements to import
element definitions associated with other namespace URIs, but that is not
quite the same as one XSD defining elements associated with multiple
namespace URIs. An XML vocabulary that involves more than one namespace
cannot be defined in a single .xsd file. Am I correct? If not, could you
please point me to, or provide, an example of a single XSD that defines
elements associated with more than one namepace URI? (The XSD mentioned in
my examples, confluence.xsd - which I wrote - uses import elements to refer
to two other XSDs.)

I hope you don't think that I'm challenging you. That's not my intention. I
am sincerely grateful for your initial reply. Having gained your attention
(and noting your signature line), I am, without wanting to take up too much
of your time, hoping to learn more. And I am prepared to publicly expose my
ignorance to do so :-).

Graham Hannington
Perth, Western Australia
Fundi Software Pty Ltd 2011 ABN 89 009 120 290
=
This message has been scanned for malware by Websense. www.websense.com