Thank you it was pretty much something like that. I need to make a sale
table but the sale code is a random number. Since it is going to be used in
all queries and it is unique, it might as well be my table key (index). I
could make a index and other field to be the sale code but then I would have
2 different indexes, the table index and the sale code that would be used
for all queries.
But my only question would be, in the example quoted would id be really the
table index and is it unique?
>Bernardo de Barros Franco writes:
> > I wanted to index a table by a random key. Exemplifying, when a insert
> > made, the id value is automatically filled with a random number between
> > 10000 and 99999.
>=> create table test (id int default random() * 89999 + 10000, content
>=> insert into test (content) values ('hi');
>INSERT 36163 1
>=> insert into test (content) values ('there');
>INSERT 36164 1
>=> insert into test (content) values ('blah');
>INSERT 36165 1
>=> select * from test;
> id | content
> 61616 | hi
> 72605 | there
> 83469 | blah
Get Your Private, Free E-mail from MSN Hotmail at http://www.hotmail.com.
pgsql-sql by date
|Next:||From: Peter Eisentraut||Date: 2001-04-16 14:03:54|
|Subject: Re: Now for the VARDATA, VARSIZE, and VARHDRSZ stuff|
|Previous:||From: Peter Eisentraut||Date: 2001-04-16 13:55:46|
|Subject: Re: Using Random Sequence as Key|