[XML-DEV Mailing List Archive Home] [By Thread] [By Date] [Recent Entries] [Reply To This Message]

Re: Groves (was re: bunch of other stuff...)

• To: xml-dev@l...
• Subject: Re: Groves (was re: bunch of other stuff...)
• From: Joe English <jenglish@f...>
• Date: Fri, 27 Sep 2002 18:14:59 -0700
• In-reply-to: <200209272030.QAA29376@m...>
• References: <200209272030.QAA29376@m...>

John Cowan wrote:

> What I'd like to understand about groves is what the valid datatypes for
> property values can be.  I assume they include at least:
>
> 	node
> 	ordered list of nodes
> 	set of nodes
> 	integer
> 	string
> 	boolean
> 	enumeration of tokens
>
> What else can exist?

See <URL: http://www.ornl.gov/sgml/wg8/docs/n1920/html/clause-A.4.1.html>,
clause A.4.1.2, "Classes and properties."

You've got nodes, node lists, and named node lists (the nodal types);
characters, strings, integers, booleans, and "component names"
(the primitive data types); and integer lists, string lists,
and "component name" lists.

The most interesting one is the "named node list" type:
it's a node list with the extra condition that each member
of the list has a name property  which distinguishes it from
the other members of the list.  (This is used to model
the 'attributes' property of 'element' nodes, for example).

There is no "node set" data type; in fact even named node lists
are considered to be ordered.

A "component name" is a reference into the grove plan, if I understand
it correctly.  (Actually I'm not entirely sure what these are for.)


--Joe English

  jenglish@f...

• References:
  • Re: Groves (was re: bunch of other stuff...)
    • From: John Cowan <jcowan@r...>

• Prev by Date: RE: Rethinking namespaces, attribute remapping (was Re: TAG on HLink)
• Next by Date: Re: limits of the generic
• Previous by thread: Re: Groves (was re: bunch of other stuff...)
• Next by thread: SOAP patented
• Index(es):
  • Date
  • Thread