[XML-DEV Mailing List Archive Home] [By Thread] [By Date] [Recent Entries] [Reply To This Message]

Re: If I want to use catalogs for XSDs, must document instance

• From: "Graham Hannington" <graham_hannington@fundi.com.au>
• To: xml-dev@lists.xml.org
• Date: Fri, 20 Apr 2012 21:43:20 +0800

Thanks again for the coherent responses, Liam. Especially at 4 am. I hope 
you got a few hours of sleep before reading this!

> No. In this case they should look at the confluence.xsd part. For
> example, if your XML document was hosted at
> http://www.example.org/docs/argyle.xml, then the XML Catalog would be
> consulted for http://www.example.org/docs/confluence.xsd.
> See http://www.w3.org/TR/2012/REC-xmlschema11-1-20120405/#schema-loc

Ah, "4.3.2 How schema definitions are located on the Web" ("non-Web 
mechanisms for delivering schemas for ·assessment· exist, but are outside 
the scope of this specification").

Sorry, I fear I should have been clearer, earlier: I am using desktop XML 
editors (such as jEdit and XMLSpy) to work with files that are located 
either on my PC's local (C:) drive or on LAN drives (or that I create from 
scratch in the editor). None of the files involved are located on the Web.

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

> As schemas become larger, it is often desirable to divide their content 
among several schema documents [...] Instance documents that conform to 
schema whose definitions span multiple schema documents need only 
reference the 'topmost' document

Also, the Primer refers to "the location of schema documents": that is, it 
refers to schema documents as being persistent objects - with locations - 
as opposed to, say, in-memory objects that might be programmatically 
generated from one or more such persistent objects.

I'm comfortable with the concept that the content of a schema can be 
divided among several schema documents; this describes the Confluence XML 
schema that I've developed.

From your email:

> Second, a Schema Document can be comprised of multiple xsd files.
> Neither I nor the XSD spec uses Schema Document to mean a single xsd 
file.

In the light of your email, I think my problem might be that I've been 
misreading the various W3C documents, and that I've been incorrectly 
thinking: one schema document = one .xsd file = one <schema> element. I 
think you're saying the term "schema document" actually means (please feel 
free to correct me): an .xsd file, *and* any other .xsd files to which it 
refers (via, for example, <import> and <include>). Or are you saying that 
a "schema document" is not just those .xsd files, but a single object that 
is generated (in memory, by an XML processor) from all of those files? (In 
which case, how can the primer refer to the persistent "location" of a 
schema document?)

Finally (cringe, sorry), I'm still not clear on the following issue: in a 
single .xsd file (that does not refer to any other .xsd files, via 
<include>, <import>, or any other means), can I define some elements that 
are associated with one namespace, and other elements that are associated 
with another namespace? I think the answer is "no".

Graham Hannington
Perth, Western Australia

Fundi Software Pty Ltd  2012  ABN 89 009 120 290


This message has been scanned for malware by Websense. www.websense.com

• References:
  • If I want to use catalogs for XSDs, must document instances contain aschemaLocation?
    • From: "Graham Hannington" <graham_hannington@fundi.com.au>
  • Re: If I want to use catalogs for XSDs, must documentinstances contain a schemaLocation?
    • From: Liam R E Quin <liam@w3.org>
  • Re: If I want to use catalogs for XSDs, must document instancescontain a schemaLocation?
    • From: "Graham Hannington" <graham_hannington@fundi.com.au>
  • Re: If I want to use catalogs for XSDs, must documentinstances contain a schemaLocation?
    • From: Liam R E Quin <liam@w3.org>