[prev in list] [next in list] [prev in thread] [next in thread] 

List:       kde-pim
Subject:    Re: Need for a general purpose database in KDE (maybe kab)
From:       Rik Hemsley <rik () rikkus ! demon ! co ! uk>
Date:       1999-07-12 9:56:31
[Download RAW message or body]


On 12-Jul-99 Mirko Sucker wrote:
> dsanders@cch.com.au wrote:
> 
>> Hmm I don't quite understand this. It is definitely desirable for an address
>> book entry object to have a method which returns all fields defined for that
>> object, and to be able to add/update fields/values at runtime.
> 
> I am not sure about this. Of course, it is desirable, but this limits the
> backends to be registry-type databases. But most DB engines are table
> based, and tables need a specified width. The only way to join both aspect is
> to use a table that provides some reserve fields.

Are you saying that it would be excessively hard work to create an extensible
addressbook format if storing the addressbook within a database ?

> Sorry, guys, but most of what you discuss here has been discussed over a
> long time. See the "What database" thread in the mailing list archives and
> much more. Please do not expect this job to be standard database development,
> much stuff is a matter of taste, and sometimes it is hard really code what all
> people want.

I did follow that thread. I don't think the design of KDE's addressbook should
follow the as yet unwritten design of a general KDE database storage backend.

> What I want to say is:
> Lets start the project kde-pim, but do not estimate a early time when it
> should
> be finished. Go create a REAL GOOD DATABASE DESIGN. And implement it.

kde-pim has been going a long time now. It started when I began writing Empath
last November. We've now got a mail client that uses Maildir and will be
easy to map a CORBA interface to. We also have (two!) excellent rfc822 message
parsing libraries, a vCard library and an LDIF library.

As for designing a database, I presume you mean designing an API for a database
backend, or do you mean designing a structure for kab records ? I'm trying
to design a record structure that allows for extension records and value types
right now.

> And touch the kdelibs only if you finished it and know that it works.

Heh, not yet ;)

> I will join the development, and I am sure Markus Wuebben also will.

Good. Can you help out with thinking about the addressbook record structure ?
I have some BNF + diagrams I'm working on that I'll post once they look like
being right.

Cheers,
Rik



--
KDE - Colour outside the lines  : http://www.kde.org
[[without]] - software for KDE  : http://without.netpedia.net

[prev in list] [next in list] [prev in thread] [next in thread]