What is a document?

[Mark Birbeck]

Don Park wrote
> You quoted me out of context.  The paragraph you quoted was a
> reply to the following paragraph which clearly points out that
> the context of the discussion was the need for a fragment to
> impose constraint on its container:

I suspected as much - that's why I also asked the question which you
answer below.

> Written by somebody, I forget who (sorry):
> >I wouldn't impose such a limitation myself. Why should the fragment
> >'know' anything about its container? (If I understand you correctly -
> >and apologies if I don't.) Provided it has its namespace declaration
> >then it can in theory exist in any container that will have it.

That was me.

> Imagine a repository of XSLT templates which is used by XSL editors
> to construct a XSLT document in a top-down fashion.  Select an element
> in the source tree and you will see a list of templates that can be
> applied.  Most likely not all templates will be appropriate for
> insertion.  Note that each XSLT templates are, in fact, XML fragments.
> This is an example of constraints an XML fragment might impose on its
> container.
> 
> Another example is an e-wallet which holds private information.  When
> a service requires some bits of information and the owner of 
> the wallet
> approves, wallet sends the information in XML.  Requesting service
> defines the format it wants dynamically using a format similar to a
> form and the wallet injects necessary information as if it was filling
> out the form.  If a wire-transfer directions must be sent, it can not
> be just dropped into any location in the outgoing message.  While this
> problem can be solved in other ways, suppose I find it ideal to have
> this 'constraint' encoded into each item in the wallet for
> extensibility.  Again, this is an example of constraints an 
> XML fragment
> might impose on its container.
> 
> Are we in sync?

Yes.

I think both of these are very good examples, but I think that the
information you desire could not be in the fragment itself. The reason I
used the example of returning one node in my previous message was
because it raises the question "where would you put these restrictions?"
I think both of your examples are application specific, and so would
require a specific solution. The discussion about fragments and so on is
a general solution that could be applied across all XML.

(As it happens I think you could do the XSL editor one with a dynamic
schema that your editor understood. As you edit the document and add
nodes you can modify the schema. As you get to each node you can present
a list of valid children - from the schema.)

Regards,

Mark

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)