From: | Jean-Michel POURE <jm(dot)poure(at)freesurf(dot)fr> |
---|---|
To: | postgres(at)vrane(dot)com |
Cc: | pgsql-general(at)postgresql(dot)org |
Subject: | Re: HOWTO - Random character generation for primary key |
Date: | 2002-05-02 11:01:22 |
Message-ID: | 200205021301.22897.jm.poure@freesurf.fr |
Views: | Raw Message | Whole Thread | Download mbox | Resend email |
Thread: | |
Lists: | pgsql-admin pgsql-general |
Le Jeudi 2 Mai 2002 12:13, postgres(at)vrane(dot)com a écrit :
> > Le Jeudi 2 Mai 2002 11:30, postgres(at)vrane(dot)com a écrit :
> > > Untrue. I am using email addresses
> > > as primary keys in a table. Where
> > > did you get that information?
> >
> > Sorry, I did not mean that of course. I posted this small howto in reply
> > to a mail. It shows how to create a VACHAR PRIMARY KEY with unique random
> > values.
> > On 10 character lenght, there is approximatively one chance out of 26^10
> > to have a similar value. If you add a string timestamp, like
> > "20020501_1153_" in front of the random value, it is very likely that
> > this value will be UNIQUE in the word:
>
> In the real world "very likely" is not good enough. Can you gurantee
> 'uniqueness'?
A timestamp (ex: 20020501_1258) followed by PLpgSQL function
random_string(200) value should provide a unique signature. What do you
think? Alternatively, could we use the crypto package to generate a unique
signature?
Cheers,
Jean-Michel POURE
From | Date | Subject | |
---|---|---|---|
Next Message | Judy Jecelin | 2002-05-02 13:19:57 | unsubscribe |
Previous Message | Jean-Michel POURE | 2002-05-02 09:58:53 | Re: HOWTO - Random character generation for primary key |
From | Date | Subject | |
---|---|---|---|
Next Message | Thierry Besancon | 2002-05-02 11:28:13 | hexadecimal values |
Previous Message | Jean-Michel POURE | 2002-05-02 09:58:53 | Re: HOWTO - Random character generation for primary key |