Re: sequences vs oids as primary keys
| От | Merlin Moncure |
|---|---|
| Тема | Re: sequences vs oids as primary keys |
| Дата | |
| Msg-id | b42b73150607260654t52545014x884cf74388ae01f4@mail.gmail.com обсуждение исходный текст |
| Ответ на | Re: sequences vs oids as primary keys (Kenneth Downs <ken@secdat.com>) |
| Список | pgsql-general |
On 7/25/06, Kenneth Downs <ken@secdat.com> wrote:
> craigp wrote:
>
> >1) does it make sense (and would it be possible) to make a rule which would,
> >say, somehow write into the oid field of a tuple to be returned by lastoid? i'm
> >assuming here that the database would not have oid's enabled.
> >
> >
> >
> We do this in a trigger. We assign the NEXTVAL to a variable, write
> that to the row, then raise its value as a notice. Then we just
> retrieve the notice.
another way to to this is make a dynamic plpgsql function that takes
an insert statement and sequence name as parameters and returns the
currval on the way out. to the op I would suggest that you can inline
currval into insert statements following the original insert e.g.
insert into master default values;
insert into detail(currval('master_id_seq'), foo, bar);
merlin
В списке pgsql-general по дате отправления: