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

RE: Very Urgent !!

On Fri, 19 Jan 2001, Knapp, Robert \(CAP, CMC\) wrote:
> >%_>-----Original Message-----
> >From: Falko Braeutigam [mailto:falko@smb-tec.com]
> >Sent: Friday, January 19, 2001 12:32 PM
> >To: 'ozone-users@ozone-db.org'
> >Cc: 'lee@insight.rr.com'
> >Subject: RE: Very Urgent !!
> 
> 
> >On Fri, 19 Jan 2001, Knapp, Robert \(CAP, CMC\) wrote:
> >> >%_>> 1. to run a xpath query against more than one XMLContainer
> >> >Yes, but you have to do it 'by hand'. Prowler (part of
> >> >http://www.infozone-group.org) provides a way to query over several
> >> document
> >> >(and data sources!) Anyway, I guess this does fit your needs.
> >> 
> >> Several data sources?  Is this limited to ozone only
> >> or any XML repository? 
> >Any kind of data source! 
> 
> >But there are some issues with the XML mapping. Prowler provides an uniform
> XML
> >facade for all the connected data sources but this is intended to be used
> for
> >presenting the data only. People are trying to use this XML facade as the
> basis
> >for their business logic which leads to performance and other problems.
> 
> If I understand what you are saying (it's been a long week, so please bare
> with me),
> it may be exactly what we need.  Our design specifies for a database
> encapsulator
> which will accept queries, translate them and return an XML string(or parsed
> xml string) to the calling agent. (One of the reasons Ozone is our first
> target is that I thought it would be easy to implement on Ozone, and allow
> us to concentrate on the bussiness logic of the app). All we would be doing
> is storing and retrieving these objects.  So this should be easy, but
> there's always a but.... Certain industries require specific databases (for
> example most US government labs require an oracle backend.), so we need
> something that can retrieve and store from OODBMS's and RDBMS's. 
> 
> Do you see any issue using prowler in this way?
Basically not but this depends on what the queries are. If it's SQL, then you
are fast but you will stick with RDBMSs. If you plan to use XPath, then you
are completely independent of the underlying datastore but it's very very
very hard to find a good SQL->XML mapping and tool configuration that performes
well.


Falko

> 
> >> For example SQL Server 2000 
> >> *claims* to be accessible as an xml data store.  
> >It doesn't have to provide a XML interface. There are adaptors for
> JDBC/SQL.
> 
> This may save us a lot of trouble! I'm all excited now! WEEEE-HOOOOO!!!!!
> 
> >Falko
> 
> RobK
> 

----------------------------------------
Content-Type: text/html; name="unnamed"
Content-Transfer-Encoding: quoted-printable
Content-Description: 
----------------------------------------

-- 
______________________________________________________________________
Falko Braeutigam                              mailto:falko@smb-tec.com
SMB GmbH                                        http://www.smb-tec.com