OASIS Mailing List ArchivesView the OASIS mailing list archive below
or browse/search using MarkMail.


Help: OASIS Mailing Lists Help | MarkMail Help



   RE: need for defining standard APIs for xml storage

[ Lists Home | Date Index | Thread Index ]
  • From: "Didier PH Martin" <martind@netfolder.com>
  • To: "KenNorth" <KenNorth@email.msn.com>, "Dongwook Shin" <dwshin@nlm.nih.gov>, <xml-dev@xml.org>
  • Date: Mon, 3 Apr 2000 20:25:10 -0400

Hi Ken,

Ken said:
 You use objects for the interface to the engine, not to traverse or operate
directly on the data.

Didier replies:
This is precisely what I mean. The SelectNodes (QueryType, expression) takes
as parameter a query language type, and a query expression whatever the
query language type is and whatever the query expression is. Therefore the
interface member is invariant even if the information set engine may support
a plethora of query language. The node set composition and information
retrieval is opaque and what it passed to the interface member is two
strings and a node-set returned. We should also mention that the node-set
interface is itself an abstraction and a proposed object model to data that
may be structured in a manner proper to the information set.

If the requested information is totally unstructured then a single node is
returned and the data content a whole document. Or, if the engine is
sufficiently intelligent, nodes could be created even if the document has no
XML structures. But for the moment, let's take a chunk at a time ;-)

And yes, nothing prevent the information set designer to have a fragment of
the information set to be created dynamically from a SQL query or whatever.
So, some nodes of the information set could be the equivalent of the
xlinclude:include element, be resolved internally and integrated in the
information set. The name space or information space presented to the client
is therefore independent of any way you obtain the data. Again, this is a
concept we got with Grove but as we are now used in the XML world to have
cats renamed felines (or minou :-), let's rename Grove  information set and
let's transfer some of its important concepts :-)

Didier PH Martin
Email: martind@netfolder.com
Conferences: Web Chicago(http://www.mfweb.com)
             XML Europe (http://www.gca.org)
Book: XML Professional (http://www.wrox.com)
column: Style Matters (http://www.xml.com)
Products: http://www.netfolder.com

This is xml-dev, the mailing list for XML developers.
To unsubscribe, mailto:majordomo@xml.org&BODY=unsubscribe%20xml-dev
List archives are available at http://xml.org/archives/xml-dev/


News | XML in Industry | Calendar | XML Registry
Marketplace | Resources | MyXML.org | Sponsors | Privacy Statement

Copyright 2001 XML.org. This site is hosted by OASIS