I'm not sure if I understood you right, but it
sounds similar to a case which I faced recently. Why not to use an
inverse approach? In other words: trying to find those registries which
hasn't got at least one value which differs from which you want to look
for. How? Using the EXISTS function with some kind of subquery I guess.
On 21/09/10 10:11, Michele Petrazzo - Unipex wrote:
Oliveiros
d'Azevedo Cristina ha scritto:
Hello again, Michele,
Ciao,
I haven't open my mailbox during weekend so I
couldn't follow up your
question.
No problem!
It would help if you explain a little better
the background of the
problem you're trying to solve.
You want to find all the user IDs which have the same value on a
given time interval?
Is my understanding correct?
Yes.
Re-reading my post I saw that I could explain better!
id_user | value | datetime
1 | 1 | xxx
1 | 2 | xxx+1
1 | -1 | xxx+2
2 | -1 | xxx
2 | -1 | xxx+1
3 | 4 | xxx
3 | 10 | xxx+1
3 | 4 | xxx+2
4 | 3 | xxx
4 | 3 | xxx+1
So, the new question: how I can find which id_user has _all_ the
"value"
that I'm looking for? Say -1 as 3 and I want a id_user=2 for the first
and for the latter id_user=4
Thanks,
Michele
Best, Oliver
----- Original Message ----- From: "Michele Petrazzo - Unipex"
<michele.petrazzo@unipex.it> To: "Oliveiros d'Azevedo Cristina"
<oliveiros.cristina@marktest.pt> Cc:
<pgsql-sql@postgresql.org> Sent:
Friday, September 17, 2010 4:45 PM Subject: Re: [SQL] all the table
values equal
Oliveiros d'Azevedo Cristina ha scritto:
Howdy , Michele,
Give this a try
SELECT id_user FROM t_your_table WHERE datetime BETWEEN A --
plug here beginning of interval AND B -- and end here GROUP BY
id_user HAVING COUNT(*) = -SUM(value)
Then tell me if it gives you what you want
Thanks, it works, but... it's really a trickle that exploits the
value -1 if I understand how its work. If there is another value
where look for? Example 13?
Thanks
-- Sent via pgsql-sql mailing list (pgsql-sql@postgresql.org) To
make changes to your subscription:
http://www.postgresql.org/mailpref/pgsql-sql