Re: modelling history in a database

From: stu <smcgouga_at_nospam.co.uk>
Date: Fri, 14 Feb 2003 11:39:19 -0000
Message-ID: <b2ikjv$4rt$1$8302bc10_at_news.demon.co.uk>


I actually did a simple version of this a while back. archive all fields to csv format and store in a memo field:
name, age, height, weight,....

This was fine until ppl started changing field layouts/names! Might be worth looking at xml for this. you would dump data out like: <name>stuart</name> <Age>23</Age> ..... or if it is a longer term archive you could even consider dumping csv files into a zip/rar file to archive off. you can control zip/rar with vb.

just make sure your restore function is flexible.

cheers
Stu

"Richard Wheeldon" <richard_at_rswheeldon.com> wrote in message news:3E4CD01D.4487_at_rswheeldon.com...
> stu wrote:
> > > Second, you could create a table Change that has fields for the table,
> > > field, old data, new data and date of change.
> > I like that idea! Have you tried it in practice? Does anybody else
have
> > any comments on this suggestion?
>
> I was considering doing this for a project I'm involved in. It's a
> coursework marking system, where coursework and question definitions
> are spread across several tables. I was considering a version history
> using a serialized LOB (i.e serialized all the objects, then save
> them to a blob) giving only one backup/version table covering data
> in several tables. This may be a good way if there's lots of complex
> data changes and the only retrieval likely is simple reads,
>
> In this case the schema would be
> table change(
> id_of_thing_being_changed references id_of_current_thing
> date_of_version date
> definition blob
> );
>
> Worth considering ?
>
> Richard
Received on Fri Feb 14 2003 - 12:39:19 CET

Original text of this message