From: | Wolfgang Keller <feliphil(at)gmx(dot)net> |
---|---|
To: | pgsql-general(at)postgresql(dot)org |
Subject: | Re: database design best pratice help |
Date: | 2013-01-30 17:43:42 |
Message-ID: | 20130130184342.939c6f9a1559c119603874fa@gmx.net |
Views: | Raw Message | Whole Thread | Download mbox | Resend email |
Thread: | |
Lists: | pgsql-general |
> In my db I have about one hundred tables like this:
>
> code
> description
>
> To avoid to have a so great number of similar tables in the db
> I wonder if it is a good idea to unify all these tables in one big
> table like this:
>
> id
> code
> table_ name
> description
Bad idea.
E.g. how do you want to warrant (integrity!) that records in
the table <table_x> point only to codes that are valid for this table.
To implement this as a constraint and at object level would create an
obfuscated mess.
Besides, that "id" column is a bad idea as well. The primary key of such
a table would have to be composed of "code" and "table_name", of course.
> 2. only one controller to manage the table
With any reasonable database framework that I know of, it's
ridiculously trivial to declare these codes as de-normalises "one to
many"-attributes at the object level. No need to implement *any*
"controllers" at all.
Sincerely,
Wolfgang
From | Date | Subject | |
---|---|---|---|
Next Message | Bret Stern | 2013-01-30 17:45:12 | Windows Phone/Postgresql |
Previous Message | John R Pierce | 2013-01-30 17:18:50 | Re: pg_Restore |