Re: "Unvisible" duplicates in SELECT [...] HAVING count(attrib) > 1

18.05.2001, 11:12:27, Tom Lane wrote:

> Krzysztof Gajdemski <songo(at)vi(dot)pl> writes:
> > PostgreSQL version (example: PostgreSQL-7.1): PostgreSQL-7.1rc4
>
> > query doesn't produce reliable results for some specific circumstances
>
> I tried your example in current sources and get reasonable-looking
> behavior:
>
> regression=# SELECT nick FROM user_id GROUP BY nick HAVING count(nick) > 1;
> nick
> --------------------
> dr dre\$\$\$\$\$\$
> sexylady\@
> (2 rows)

[ cut ]

> It could be that there's a locale dependency here (I did this in a
> non-locale-enabled compilation),

I've used binary PostgreSQL package from Debian/unstable distribution,
so could it be, that this is Debian specific problem if you can't
reproduce it.

> but I think it's more likely that
> you've run into one of the complex-query bugs we corrected since rc4.
> Please update to 7.1.1 (or better 7.1.2, which should be out very soon)
> and see if you still see a problem.

Yes, it is a good idea. I've to try compile it directly from source
(with locale support enabled).

k.
--
- - - - - - Krzysztof Gajdemski | songo @ vi.pl | KG4751-RIPE
Registered Linux User # 133457 | BLUG Registered Member # 0005
PGP public keys at: keyserver.linux.bydg.org * KeyID: D336705B
,,Szanuję was wszystkich, którzy pozostajecie w cieniu'' SNERG