Hi Alban,
On 7/3/06, Alban Hertroys <[EMAIL PROTECTED]> wrote:
Alexander Farber wrote:
> punbb=> select username, md5('deadbeef' || password) from users where id
> = 4;
> username | md5
> --+--
> Vasja| dcde745cc304742e26d62e683a9ecb0a
>
Alexander Farber wrote:
I wonder, what is faster: fetching 2 columns - the
username and the md5-result and then comparing the
md5 string against the argument in my app, like here:
punbb=> select username, md5('deadbeef' || password) from users where id
= 4;
username | md5
---
Yes, you're probably right.
I'm just trying to ensure, that the slow md5() function isn't
called for every row in the table. If that's not the case, then
the other tiny speed differences are not that important for me.
Your query works too, thanks for the hint.
punbb=> select username, md5('dead
On Mon, Jul 03, 2006 at 03:13:15PM +0200, Alexander Farber wrote:
> Hello,
>
> in my application I'm trying to authenticate users
> against a table called "users". The integer column
> "id" should match, but also an md5 hash of the
> "password" column (salted with a string) should match.
> My auth
Hello,
in my application I'm trying to authenticate users
against a table called "users". The integer column
"id" should match, but also an md5 hash of the
"password" column (salted with a string) should match.
My authentication function (written in C, using libpq)
should return a "username" (is
