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

[Elliotte Rusty Harold <elharo@metalab.unc.edu>]

At 12:38 PM +1100 11/6/01, Marcus Carr wrote:


>Replace "in the first place" with "at some point in their lifecycle" and you
>have a problem. An instance may be used as well-formed or valid any number of
>times in any order - if you add attributes, you risk breaking the data for a
>downstream player who wants to validate. If that player doesn't control the
>DTD, a preprocess will be required to strip the attributes out. How could
>this not be so?

It can not be so, because four years ago the XML Working group 
decided to let documents be invalid. Validity is optional in XML. It 
is not required. If you're adding or stripping attributes willy-nilly 
during a document's lifecycle without paying attention to what the 
DTD says, then the document may become invalid. This is true whether 
the attributes are named xml:id, xlink:label, id, or shazbot. Nothing 
we're doing changes any of that.

If you want your documents to be valid, you declare all possible 
attributes in the DTD including xml:id/xlink:label/id/shazbot. If you 
want your documents to be valid with or without certain attributes, 
you make those attributes #IMPLIED. If you don't care whether your 
documents are valid or not, you ignore DTDs.
-- 

+-----------------------+------------------------+-------------------+
| Elliotte Rusty Harold | elharo@metalab.unc.edu | Writer/Programmer |
+-----------------------+------------------------+-------------------+
|          The XML Bible, 2nd Edition (Hungry Minds, 2001)           |
|              http://www.ibiblio.org/xml/books/bible2/              |
|   http://www.amazon.com/exec/obidos/ISBN=0764547607/cafeaulaitA/   |
+----------------------------------+---------------------------------+
|  Read Cafe au Lait for Java News:  http://www.cafeaulait.org/      |
|  Read Cafe con Leche for XML News: http://www.ibiblio.org/xml/     |
+----------------------------------+---------------------------------+