RE: [xml-dev] Re: determining ID-ness in XML

[Leigh Dodds <ldodds@ingenta.com>]

> -----Original Message-----
> From: David Carlisle [mailto:davidc@nag.co.uk]
> Sent: 06 November 2001 09:49
> To: ldodds@ingenta.com
> Cc: xml-dev@lists.xml.org
> Subject: Re: [xml-dev] Re: determining ID-ness in XML
>
>
>
>
> > The 'XML processor' will be checking uniqueness of the identifiers, so
> > it'd be a snap to check that cross-references are correct as well.
>
> It's not clear that checking uniqueness is desirable. It makes sense for
> validating parsers, but as a run time linking facility which appears to be
> the main use case here, just taking the first available
> node with that identifier seems preferable.

That does seem reasonable, but the preceding discussion seems
to have been primarily about unique identifiers: IDs (or things that
are not IDs but are unique).

[But I'll still (doggedly) assert that being able to define an attribute
as being a cross-reference (to a unique id or no) is as useful a
feature as being able to define a unique id. If the only use case
we're considering here is linking, then I'll accept that it's a step
too far. But is that the only use case?]

> (This is like the xpath id() function, which does not require unique ids.)

I thought it did require uniqueness:

"The id function selects elements by their unique ID"
http://www.w3.org/TR/xpath#function-id

> if you connect to http:ksvdbsldbsl#foo and ksvdbsldbsl has two nodes
> labeled "foo" then going to the first of them is (or might be)
> preferable to being given an error message.

Yep.

So why not this: http://www.foo.com#foo=bar  (Assuming the '=' is escaped)
Lets call it the 'attribute selector' proposal because I lifted it from the
workaround CSS suggests for the lack of available unique ids.

Cheers,

L.