[Top][All Lists]
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: [Gnumed-devel] database abstraction
From: |
Horst Herb |
Subject: |
Re: [Gnumed-devel] database abstraction |
Date: |
Sat, 26 Oct 2002 10:33:08 +1000 |
User-agent: |
KMail/1.4.7 |
On Sat, 26 Oct 2002 03:32, Hilmar Berger wrote:
> (There is another case in which PgObject will be problematic: if SELECT's
> using LIKE or regular expressions are used, indexes to an object entry
> ala drug.['digi.*'] might fail.)
gmPgObject is an abstraction layer of data, not of database capabilities. It
is not designed for *searching*.
You perform your search by any means you fancy, and once you have identified
the row you wish to work on, you tell gmPgObject.
I don't think that it is realistic to create highly abstract wrappers around
such search functionality; I think SQL is abstract enough ;-)))
The reason is that in searches we are potentially handling large amounts of
data, from which most will never be accessed. Would be a humongous waste of
performance and ressources building complex object hierarchies around such
queries.
That said, you still can do it - gmPgObject would then just be the lowest base
class of the rows you abstract.
Horst
- [Gnumed-devel] database abstraction, Horst Herb, 2002/10/23
- Re: [Gnumed-devel] database abstraction, Hilmar Berger, 2002/10/23
- Re: [Gnumed-devel] database abstraction, Karsten Hilbert, 2002/10/24
- Re: [Gnumed-devel] database abstraction, Horst Herb, 2002/10/24
- Re: [Gnumed-devel] database abstraction, Karsten Hilbert, 2002/10/24
- Re: [Gnumed-devel] database abstraction, Hilmar Berger, 2002/10/24
- Re: [Gnumed-devel] database abstraction, Horst Herb, 2002/10/24
- Re: [Gnumed-devel] database abstraction, Karsten Hilbert, 2002/10/24
- Re: [Gnumed-devel] database abstraction, Hilmar Berger, 2002/10/25
- Re: [Gnumed-devel] database abstraction,
Horst Herb <=
Re: [Gnumed-devel] database abstraction, Hilmar Berger, 2002/10/24