[SML] Whether to support Attribute or not?

[Rick Jelliffe]

From: Pavel Velikhov <pvelikho at cs.ucsd.edu>

 >Rick Jelliffe wrote:
>>
>>
>> From: Pavel Velikhov <pvelikho at cs.ucsd.edu>
>>
>> >If you don't know whether the object you are picking is an
>> >attribute or an element, you will have to write a disjunction.
>>
>> When don't you know whether an object you are picking is an
>> attribute or an element?
>>
>> Rick Jelliffe
>
>When you are querying a collection that does not have a DTD for
>example. You may know that homes have an id, but you don't know
>whether the id is an attribute or a element. Another case is a
>relational->XML wrapper, it export relational attributes as
>XML attributes, but might as well export them as elements.

Isn't your complaint really that there is no standard mapping from
a relational schema to a notional DTD/Schema?  Let alone for
round-tripping.   A policy that said "1 field equals 1 element"
would meet your requirements even in XML ; SML reduces
the choices in one area, but not enough to make querying work
automatically.

Without such a mapping policy, you cannot assume anything about
what is returned by a query.  If the database uses case-sensitive names,
which case is returned for the generic identifier? If the data is stored
as a date, what date format is returned?  If a text field is used, can
that
text field contain markup?

So it would be a standard mapping policy that would fix the problem,
in any markup language.

Rick Jelliffe



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 unsubscribe, mailto:majordomo at ic.ac.uk the following message;
unsubscribe 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)