Schema and Tree representation of the document

Hi,

By reading the Schema specification, it seems that the logical representation
of a document instance in a validation process using schema needs to
be a tree :
- schema uses a infoset representation of the document
- a validating processor must scan the document tree by
    validating at first the root node and then process its children nodes (1)

So, I suppose that for beeing conform, a processor need to 
work on the real tree view of the document, physically : 
when it works on a node, it needs a access all its children
and then needs to have the whole subtree loaded.
We need a "tree oriented" processor for using schema validation,
and a "event-driven" processor is not able to do this.

If it is true, how may i work with large documents 
(for example, an aircraft maintenance manual document
is more that 100Mo) with schema. If my software environment
does not include a "doc on disk" mecanism with a tree oriented
representation (i.e. giving node navigation interface), I will
not be able to work with a schema ? 
Or do I misunderstand something ?

Regards,

-- bruno --
-----------------------------------------------------
Bruno Chatel
Tel : (33)[0]4 96 11 14 57
E-mail : bcha@chadocs.com   
http://www.chadocs.com
-----------------------------------------------------



(1) http://www.w3.org/TR/xmlschema-0/#conformance
"Conformance checking can be thought of as proceeding in steps, first checking
 that the root element of the document instance has the right contents, then 
checking that each subelement conforms to its description in a schema, and 
so on until the entire document is verified. "

ps : sorry for my "hesitating" english

Received on Monday, 16 October 2000 10:37:29 UTC