Re: [xml-dev] IDs considered harmful or why keys might be better than IDs...

[John Cowan <jcowan@reutershealth.com>]

Jonathan Borden wrote:


> The only reason that I see the need for something really lightweight like
> "xml:id" is that internal subsets are not well handled by common software
> (e.g. SAX),


Actually, SAX is able to report the types of attributes.  There is
nothing in XML 1.0 requiring XML processors to make this information
available, however.

It would be interesting to know which XML processors don't report
attribute type information for attributes declared in the internal
subset.  Ignoring ATTLIST declarations in the internal subset
altogether is forbidden, because defaulted attribute values
must be properly reported.

-- 
Not to perambulate             || John Cowan <jcowan@reutershealth.com>
    the corridors               || http://www.reutershealth.com
during the hours of repose     || http://www.ccil.org/~cowan
    in the boots of ascension.  \\ Sign in Austrian ski-resort hotel