Re: problematic view definition
От | Karsten Hilbert |
---|---|
Тема | Re: problematic view definition |
Дата | |
Msg-id | 20110221070633.GB2610@hermes.hilbert.loc обсуждение исходный текст |
Ответ на | Re: problematic view definition (Tom Lane <tgl@sss.pgh.pa.us>) |
Список | pgsql-general |
On Sun, Feb 20, 2011 at 02:31:46PM -0500, Tom Lane wrote: > Karsten Hilbert <Karsten.Hilbert@gmx.net> writes: > > On Wed, Feb 09, 2011 at 11:12:01PM +0100, Karsten Hilbert wrote: > >> Unfortunately I do not understand why PostgreSQL says > >> > >> psql:xx.sql:14: ERROR: could not implement UNION > >> DETAIL: Some of the datatypes only support hashing, while others only support sorting. > > > The solution lies in these bits: > > >> pk_context | integer[] | > > > This data type can only be hashed. > > >> xmin_message_inbox | xid | > > > This data type can only be sorted. > > ITYM the other way round, right? Indeed :-) > As of 9.1 there will be support for hashing arrays, so this particular > problem should go away without hacks. Great. PG is getting better by the day :-) Anyway, I consider the "explicit cast" not really that bad of a hack. Karsten -- GPG key ID E4071346 @ gpg-keyserver.de E167 67FD A291 2BEA 73BD 4537 78B9 A9F9 E407 1346
В списке pgsql-general по дате отправления: