From: | Vincenzo Romano <vincenzo(dot)romano(at)notorand(dot)it> |
---|---|
To: | Dimitri Fontaine <dfontaine(at)hi-media(dot)com> |
Cc: | Pavel Stehule <pavel(dot)stehule(at)gmail(dot)com>, PostgreSQL General <pgsql-general(at)postgresql(dot)org> |
Subject: | Re: Converting BYTEA from/to BIGINT |
Date: | 2010-07-27 09:34:35 |
Message-ID: | AANLkTingQ=Co21qTGUqd0XHyP37pp2qApiSZZHXeiN15@mail.gmail.com |
Views: | Raw Message | Whole Thread | Download mbox | Resend email |
Thread: | |
Lists: | pgsql-general |
2010/7/27 Dimitri Fontaine <dfontaine(at)hi-media(dot)com>:
> Vincenzo Romano <vincenzo(dot)romano(at)notorand(dot)it> writes:
>> Now, why doing this?
>> I am using a plain SEQUENCE to create a (kind of) "session ID". That
>> is simple but predictable.
>> The idea is to use this function in conjunction with encrypt (from
>> pgcrypto) and the blowfish algorithm
>> to make that sequence numbers somehow unpredictable.
>> I'm pretty sure there are better (or at least easier) solutions out
>> there, but there needs to be also some fun
>> in here.
>
> I think you'd be interested into the following:
>
> http://wiki.postgresql.org/wiki/Pseudo_encrypt
That solution has a limit I knew about: it only generates 31-bit values.
I could "easily" trick it to handle 62-bit. Then I decided to play that game ...
--
Vincenzo Romano
NotOrAnd Information Technologies
NON QVIETIS MARIBVS NAVTA PERITVS
From | Date | Subject | |
---|---|---|---|
Next Message | Ravi Katkar | 2010-07-27 10:23:05 | Re: ODBC and configure |
Previous Message | Dimitri Fontaine | 2010-07-27 09:28:57 | Re: Integral PG DB replication |