"P. A. Bagyenda" <bagyenda 'at' dsmagic.com> writes:

> Automatic OIDs on each table row were removed from PostgreSQL a while back
> (v7.3?), so it is time they got removed from Kannel's code. You can still of

No, it changed in 8.1.

http://www.postgresql.org/docs/8.3/static/release-8-1.html#AEN89790

> course use OIDs if you create your tables with a syntax something like:
>
> CREATE TABLE (
> ....
>
> ) WITH OIDS;

Yes, and it works properly with old PGs (verified on 7.3.4 and
7.4.5) and new PGs (verified on 8.2.5 and 8.3.1), so it's
definitely the way to go; unless of course, we come up with a
better solution than using the OIDS, which should be avoided as
per PG documentation - but that's been an unreachable quest in
the past, and it will probably say so if we want to stay
compatible with multiple different PG generations.

I'd say this is the best way to go for the moment, and we already
have a separate CREATE TABLE documentation for PG anyway, since
the datatype "INT(10)" doesn't mean something in PG; so we can
just amend it and be happy:

http://www.mnc.ch/~gc/dlr-sdb-postgresql-example-with-oids.patch

Now we're at it, from experience I'd also include an index
creation in the example; with some non negligible traffic, this
index became priceless to keep the disk I/Os in reasonable
values:

http://www.mnc.ch/~gc/dlr-sdb-postgresql-example-with-oids-and-index.patch

-- 
Guillaume Cottenceau, MNC Mobile News Channel SA, an Alcatel-Lucent Company
Av. de la Gare 10, 1003 Lausanne, Switzerland - direct +41 21 317 50 36

Reply via email to