Re: Is it possible to build a purely relational database on top of SQL?

From: Peter Koch Larsen <pklspam_at_mailme.dk>
Date: Wed, 7 Jul 2004 17:21:58 +0200
Message-ID: <pyUGc.23101$Vf.1176447_at_news000.worldonline.dk>


Hi Leandro

As you see, I get lots of weird characters. Perhaps you should consider using a simple codepage when responding in english?

"Leandro Guimaraens Faria Corsetti Dutra" <leandro_at_dutra.fastmail.fm> skrev i en meddelelse news:pan.2004.07.05.23.17.04.217302_at_dutra.fastmail.fm...
> Em Mon, 05 Jul 2004 15:58:32 +0200, Peter Koch Larsen escreveu:
>
> > How far has Alphora come in terms of conformance with D?
>
> As far as I know it is totally conformant, only not
> implementing some extras like hierarchies of types – they did, but in

I would like an explanation what – means ;-)

> the version 2 βs decided it was not the right thing to do.
>
>
> > And if they are
> > 100% conformant, do they allow SQL access to the same data?
>
> It’s up to the SysAdmins to block the SQL server behind a
> Dataphor installation. But it even provides an SQL lookalike,
> RealSQL, to avoid the need to.

I am naturally talking in terms of migration here... if you would transfer a project from some SQL-dialect to Alphoras language (D4?), then surely it must be important with a high degree of SQL compliance?

>
>
> > I believe it could give some problems querying a database with user
> > defined datatypes in it
>
> I don’t see why.
>
Well.... if the relation uses some polymorph type - say SHAPE to use a TTM example - and the actual type is say a polygon with houndreds of vertices, then how will the data be represented to the SQL user?

/Peter Received on Wed Jul 07 2004 - 17:21:58 CEST

Original text of this message