XOM API for Processing XML with Java

[Adam Constabaris <adamc AT unc.edu>]

Dirk Bergstrom wrote:
> John Cowan was heard to exclaim, On 01/20/05 05:18:
>
> > The idea of xml:id is that it is an attribute which is always of type
> > ID even in the absence of a DTD declaration.  By using it, documents can
> > gain the advantages of IDs without having to carry the burden of a DTD;
>
>
> I've looked over the W3C recommendation, and a couple other pages I found on
> google, but I still don't understand the *purpose* of xml:id.  All I can find
> is discussion about implementation details.  Can someone provide a quick
> explanation of why I might want to use xml:id, perhaps with a semi-real-world
> example?

xml:id is designed to give you the benefits of having ID attributes in 
your documents without specifically referring to a DTD or XML Schema 
that identifies a particular attribute as being of type ID (which is the 
only way to do it now).  An ID attribute's value must be unique within a 
document; XPath, e.g. provides the id() function to allow fetching a 
node with a specific ID.   An ID attribute is thus kind of like a 
database primary key.

A semi-real world example is on the working draft of the spec at

http://www.w3.org/TR/xml-id/

=====================================
However, some specifications, notably [SOAP], forbid an internal subset, 
and processing the external subset is optional for conformant XML 
processors, leaving no guarantee that all consumers of the XML document 
will be able to successfully recognize the identifiers.

IOW, you can't guarantee that any DTD you supply in a SOAP message will 
be honoured, but you might nevertheless want to use ID-type attributes 
in such a message.

AC