[
Lists Home |
Date Index |
Thread Index
]
- To: <xml-dev@lists.xml.org>
- Subject: Assigning meaning to an element if attributes are present
- From: "Young Matthew" <matthew.young@forsakringskassan.se>
- Date: Fri, 11 Feb 2005 08:10:02 +0100
- Thread-index: AcUQCLTZ6VKVJ7BoR5Sixn5DWJiGUA==
- Thread-topic: Assigning meaning to an element if attributes are present
Hej,
Developing a XML grammer for configuring J2EE resources (via JMX) which
is translated into an Ant build file. Want to keep the set of XML
elements to a minimum and look into the idea of assigning
meaning/context to elements if attributes are present or not (or in
certain combinations). For example, the major J2EE managed objects are
"resource", "deployable", and "properties" elements designated as
children to either "domain" or "subdomain" nodes. All managed object
declarations which are children of "subdomain" nodes are restricted to
reference (via IDREF attribute) a managed object with a "domain" parent
(containing an ID attribute). This pairing of "referencing objects" to
unique managed objects (i.e. "pattern objects") is done to administrate
the definition of managed objects from the domain only.
Anyway the domain nodes could also have "complete" managed objects
without ID or IDREF attributes and thus representing a complete
definition of a managed object. Depending on which attributes are
present and which parent an element has determines the
context/intention/meaning of the element. Is this alright? Or should
XML elements solely inherit their meaning from element type
declarations. Read in the "W3C Recommendation" that quote "An element
type declaration constrains the element's content".
/ Matthew Young
|