> We can achieve desired result through creating a separate partitioned table
> and making the DETACH/ATTACH manipulation, though. But IMO it's not flexible
> case.
I think it would be great to allow end user to decide. If user is
not interested in subpartitions he or she can use syntax like 'CREATE
TABLE ... PARTITION BY HAHS(i) PARTITIONS 3 CREATE AUTOMATICALLY;' or
maybe a build-in procedure for this. Otherwise there is also
ATTACH/DETACH syntax available.
Anyway all of this is something that could be discussed infinitely and
not necessarily should be included in this concrete patch. We could
probably agree that 3 or 4 separately discussed, reviewed and tested
patches are better than one huge patch that will be moved to the next
commitfest because of disagreements regarding a syntax.
On Wed, Mar 01, 2017 at 05:10:34PM +0300, Maksim Milyutin wrote:
> On 01.03.2017 05:14, Amit Langote wrote:
> > Nagata-san,
> >
> > > A partition table can be create as bellow;
> > >
> > > CREATE TABLE h1 PARTITION OF h;
> > > CREATE TABLE h2 PARTITION OF h;
> > > CREATE TABLE h3 PARTITION OF h;
> > >
> > > FOR VALUES clause cannot be used, and the partition bound is
> > > calclulated automatically as partition index of single integer value.
> > >
> > > When trying create partitions more than the number specified
> > > by PARTITIONS, it gets an error.
> > >
> > > postgres=# create table h4 partition of h;
> > > ERROR: cannot create hash partition more than 3 for h
> >
> > Instead of having to create each partition individually, wouldn't it be
> > better if the following command
> >
> > CREATE TABLE h (i int) PARTITION BY HASH (i) PARTITIONS 3;
> >
> > created the partitions *automatically*?
>
> It's a good idea but in this case we can't create hash-partition that is
> also partitioned table, and as a consequence we are unable to create
> subpartitions. My understanding is that the table can be partitioned only
> using CREATE TABLE statement, not ALTER TABLE. For this reason the new
> created partitions are only regular tables.
>
> We can achieve desired result through creating a separate partitioned table
> and making the DETACH/ATTACH manipulation, though. But IMO it's not flexible
> case.
>
> It would be a good thing if a regular table could be partitioned through
> separate command. Then your idea would not be restrictive.
>
>
> --
> Maksim Milyutin
> Postgres Professional: http://www.postgrespro.com
> Russian Postgres Company
>
>
> --
> Sent via pgsql-hackers mailing list (pgsql-hackers@postgresql.org)
> To make changes to your subscription:
> http://www.postgresql.org/mailpref/pgsql-hackers
--
Best regards,
Aleksander Alekseev
