[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

RE: Indexing XML documents





> -----Original Message-----
> From: Falko Braeutigam [mailto:falko@softwarebuero.de]
> Anyway, why do you want to store Contact Information in
> XML?

The contact information was an example. Contact Information is
different around the world. Names, Addresses, etc. are used
differently around the world. One way to do this to create a vCard
standard and store data in a structured format. Trouble with this
method is trying to get everybody to agree on a standard is a very
difficult proposition. Another way is let every one store the
information in XML and provide transformation routines to common
formats. This would especially true if data is stored in a distributed
fashion.

> An dedicated object model seems to be much better suited. Such a
simple
> object model would
> allow you to implement as access paths (indexes, trees,
> whatever) to the data
> as you like.

Could you explain further on a "dedicated object model". My view is
that DOM/XPath provides such a mechanism and does not preclude indexes
to be built for faster access of data.

> ozone/XML just provided the persistent DOM implementation. You can
use any DOM
> based tool on top of it

Ok I am a bit confused on the Ozone architecture. Who handles actual
query process. Finally, I assume that Ozone handles the query that
tells it to gets all the book nodes which meets a certain criteria. If
that the case, that Ozone knows about the nodes in the XML document
and can optimize the processes.

>
> However, of course it is no problem to implement any kind
> of data access path
> in your application. Just put your data in a hashmap... or
> a treemap... or an
> array or whatever meets the needs of your application.
>

I like the solution that Lars proposed in a follow-up mail (will send
a more detailed reply) and this may be possible thru a middleware with
sits between XT and Ozone-Db.

Regards,
Ashvil