I don't have a problem with the concept of well-formedness, only with
the assumption, or requirement, that a document/object must be fully
validated every time it is imported into a program before use. XML 1.1
encoded as text must be parsed and a side effect is that if you are
parsing, you can and should fully validate well-formedness when
parsing. The issue is that esXML is designed to avoid parsing. That
means that the library must be the primary driver to enforce
well-formedness and that a receiver of esXML must either accept:
- assumption of overall well-formedness and validation of only
those elements visited during manipulation (i.e. the paths through the
object to the elements used, but potentially not the whole structure),
or
- overhead of a full validation step before processing.
I think that this choice of whether to fully validate or to accept
exception handling for detecting corruption but not assuring that there
is no corruption even in areas not 'visited' is the valid situation.
The opposing argument is that there is a strict requirement to fully
validate an XML-equivalent object for well-formedness after
reading/importing and before any processing is allowed. While good
practice in certain application situations, this is not necessary or
appropriate for many applications and is not a valid strict requirement
in my view. It should be detectable when processing, but only when
encountered, and otherwise optional just as validation with a
schema/DTD is optional, hence my comparison.
esXML does have a complete equivalent to XML 1.1 well formedness and to
canonicalization, as I suspect that most XML-equivalent data structures
would. It may even be very efficient to validate, but 'friendly' sets
of applications and those that are exception based should be able to
live with exceptions thrown when corruption or any violation of
well-formedness is found in the course of processing.
If well-formedness validation is pedantically considered a requirement
of any XML 1.1 processing, then that will have to be an additional
semantic relaxed slightly to "well-formedness validated when
encountered during manipulation or full on demand", along with loss of
text structuring basis.
sdw
Thomas B. Passin wrote:
Elliotte
Rusty Harold wrote:
You're mixing apples and oranges. Schema
validation (even DTD validation) is explicitly optional in XML.
Furthermore you can process an invalid document. That is definitely not
true of a malformed document. Failing to check well-formedness is not
an option.
But if the xml "alternative" is not a text format, well-formedness may
not have an equivalent.
Cheers,
Tom P
-----------------------------------------------------------------
The xml-dev list is sponsored by XML.org <http://www.xml.org>, an
initiative of OASIS <http://www.oasis-open.org>
The list archives are at http://lists.xml.org/archives/xml-dev/
To subscribe or unsubscribe from this list use the subscription
manager: <http://www.oasis-open.org/mlmanage/index.php>
--
swilliams@hpti.com http://www.hpti.com Personal: sdw@lig.net http://sdw.st
Stephen D. Williams 703-724-0118W 703-995-0407Fax 20147-4622 AIM: sdw
|
