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

RE: OOQuery Language

Hi Don,

good luck with your implementation, I am looking forward for a running
demo.  We have been working on object oriented data modelling and data
integration from heterogenous sources for some time now. We have not 
used Ozone yet as such, mostly digging into SQL and Spatial
(Arc/Info) sources. (There is not single partner of ours who could
provide us with OO data yet  ;o(  
We hvae at:    http://nzdis.otago.ac.nz/projects 
an OQL parser with full OQL EBNF grammar available for download (which is
based around SableCC parser generator) which you may find useful (the
quasi grammar from the ODMG OQL 3 book is not correct as you probably
already noticed). There is little documentation there, but we are happy to
answer all the questions ;o)  We have also a simple Arc/Info gate which
enables us to ask OQL queries to spatial file formats (it is in its
infancy yet though, and we do not redistribute it) and it is an OQL
interpreter based on the AST generated by the parser, after simplification
and tree reorganization. 

I would be interested in sharing experience in writting an OQL
interpreter - our current one is not modular and is specialized for
ArcInfo file formats, however we do have a need for a modular one, to
which we can plug different "native" drivers, like JDBC-based,
Arc/Info-based, Ozone-based etc.  Maybe we could exchange ideas and built
common API for the interpreter to make it more pluggable, and then you can
finish it up with Ozone "driver" and we could make JDBC or/and Arc/Info
"drivers" for it, what do you think?  (JDBC I mean a "driver" which takes
requests from the interpreter as a method calls via uniform API and
translates them into JDBC calls on some SQL-ba`sed data, wraps results
into objects/sets/bags/collections and returns back to the calling
interpreter).  What do others think? 


regards
Mariusz




On Fri, 15 Sep 2000, Don Berendsen wrote:

> I'm starting a crude implementation of OQL queries for ozone. I'm using it 
> at the beginning of an XML based report generation engine. The query defines 
> the data content of a report and subsequent XML/XSL/XSLFO documents define 
> the structure and display format of the output. This enables me to define a 
> report through a series of XML/XSL/XSLFO documents without writing any code 
> and then publish the report as XML, PDF or on screen as well as printing it 
> out.
> 
> I've chosen OQL since it is reasonably well-known and well-defined, it is 
> much simpler to fit to an OODBMS than SQL, and it has grammers available.
> 
> Although I think the query-kernal is ultimately the way to go, my need to 
> get a product out the door and my limited programming skills preclude me 
> from taking on the implementation of a general solution such as that.
> 
> My crude implementation will focus on the subset of OQL I need and be 
> oriented to the specific structure of our application. For example our query 
> results will be hierarchical composites rather than simple sets or bags.
> 
> Still this work may be useful to others as an example and a basis for their 
> solutions and a more comprehensive one and I'd be glad to pass the work 
> along as an example rather than a tool or framework.
> 
> Although I can understand the desire for a more OO query, my brief survey of 
> the genesis of OQL and other attempts at OO query languages convinced me the 
> definition of a robust new OO query language is not a trivial task. Although 
> S.O.D.A. and other alternatives may look interesting, AFAIK they don't have 
> a complete, rigorous definition to base an implementation upon. This is why 
> I chose not to participate in  the earlier request related to implementing 
> S.O.D.A. OQL may not be elegant, truly OO, or computationally complete but 
> it's servicable.
> 
> don
> 
> >
> >On Thu, 14 Sep 2000, Knapp, Robert \(CAP, CMC\) wrote:
> > > >%_Falko,
> > > 	From what I'm reading here (and I really just skimmed it)
> > > it sounds like you object (and your reasons are sound) to
> > > having OOQL built into Ozone.
> >Do you mean ODMG OQL? I assume yes.
> >
> > > How about building an OOQL to Ozone bridge, that would do all of the
> > > by hand work?
> > >
> > > Here's my example.  I'm working on a Laboratory Information Management
> > > System, one of the basic requirements for most labs is a SQL-like 
> >language
> > > so that they can print reports and such.
> >I understand this requirement. What about a XML interface for reports and 
> >such?
> >
> > > If someone were to build a bridge(separate from ozone) that accepted
> > > queries and spit back collections, would that be an acceptable solution? 
> >To
> > > both you and the people in need of ooql?
> >In general, even OQL is "acceptable" for me. I just said that am not going 
> >to
> >work on OQL support for ozone now or later.
> >
> > >
> > > I've been following this list for a while, and it looks like
> > > Ozone suits 75%+ of our needs(which to date seems to
> > > be more than anyone else!),
> >:)
> >
> > > so I'd like to try and figure
> > > out some way to make everyone happy here.
> >Thank you, Rob. These are very good points.
> >
> >Conclusion:
> >- descriptive query lang(s) is(are) needed
> >
> >- an independent query-kernel seems a good way to implement the pure query
> >functionality
> >
> >- different fron-ends can use this kernel to provide the different APIs 
> >(OQL,
> >SODA, 'XQL')
> >
> >
> >Falko
> _________________________________________________________________________
> Get Your Private, Free E-mail from MSN Hotmail at http://www.hotmail.com.
> 
> Share information about yourself, create your own public profile at 
> http://profiles.msn.com.
> 

-- 
  _  _   _  _   _    _
 / `'/ /-/ / ' / \/ / 
~~~~~~~~~~~~~~~~~~~~~~
--
Mariusz Nowostawski    Email: mariusz@rakiura.org   Web: www.rakiura.org
Rakiura - The Glowing Sky, Evolution Research & Software Agents at Large  
Phone work: +64 3 479 8317  home: +64 3 473 8693     Fax: +64 3 479 8311