enumeration and defaults

[Tim Bray]

At 04:34 PM 9/10/99 -0400, Simon St.Laurent wrote:
>I've found myself in a somewhat odd situation, where I'd like to be able to
>include empty (i.e., no value) as a choice in a list of enumerated
>possibilities.  It doesn't seem possible.  (Empty is not a token.)

Right - and I agree that this does seem like something reasonable to want.

>>From 3.3.2
>> #IMPLIED [means] that no default value is provided.

This is a little different from SGML.  The wording in the SGML standard
that describes what #IMPLIED means is really hard to understand, something
about the attribute being "implied" by the software.  In SGML parsers,
the effect was that the parser told you about missing #IMPLIED attributes
(this element could have had a FOO attribute but doesn't), and in early 
drafts of the XML spec we had language requiring XML processors to do the 
same thing.  On reflection, we took it out, and decided that #IMPLIED 
really means that if the attribute isn't there, it just isn't there, 
that's all.  (#REQUIRED says it has to be there, and the presence of a 
default value covers the missing case in another way).  Of course, an 
attribute being absent is obviously different from it being present with 
an empty-string value.

So I'd say that XML DTDs more or less can't do what I think Simon is
trying to do here.  If there are some strong use-cases, someone should
bombard the Schema WG with them so that next-gen schemas have this 
hole filled.  -Tim

xml-dev: A list for W3C XML Developers. To post, mailto:xml-dev at ic.ac.uk
Archived as: http://www.lists.ic.ac.uk/hypermail/xml-dev/ and on CD-ROM/ISBN 981-02-3594-1
To (un)subscribe, mailto:majordomo at ic.ac.uk the following message;
(un)subscribe xml-dev
To subscribe to the digests, mailto:majordomo at ic.ac.uk the following message;
subscribe xml-dev-digest
List coordinator, Henry Rzepa (mailto:rzepa at ic.ac.uk)