| From: | Fernando Schapachnik <fschapachnik(at)vianetworks(dot)com(dot)ar> |
|---|---|
| To: | power2themacs <power2themacs(at)yahoo(dot)com> |
| Cc: | pgsql-general(at)postgresql(dot)org |
| Subject: | Re: ISOLATION LEVEL SERIALIZABLE |
| Date: | 2002-03-26 20:06:26 |
| Message-ID: | 20020326170626.J47521@ns1.via-net-works.net.ar |
| Views: | Whole Thread | Raw Message | Download mbox | Resend email |
| Thread: | |
| Lists: | pgsql-general |
En un mensaje anterior, power2themacs escribió:
> >In table two you would not create a serial instead you would create an
> >INTEGER because serial is a counter and the values in table 2 may not be
> >in table 1.
> >
> >Use a transaction like as follows
> >
> >BEGIN;
> >INSERT INTO TABLE1 VALUES (Whatever values);
> >var = SELECT CURRVAL('sequence_name');
> >INSERT INTO TABLE2 VALUES (var,whatever else);
> >COMMIT;
> >
>
> But this is the race condition I am trying to avoid. Someone can
If you are in a trasaction, the RDBMS guarantees that your statements
executes as if their where the only ones (ie, as in a serial
execution). No race condition there.
Regards.
Fernando P. Schapachnik
fschapachnik(at)vianetworks(dot)com(dot)ar
| From | Date | Subject | |
|---|---|---|---|
| Next Message | Stephan Szabo | 2002-03-26 20:09:41 | Re: ISOLATION LEVEL SERIALIZABLE |
| Previous Message | Doug McNaught | 2002-03-26 19:58:47 | Re: ISOLATION LEVEL SERIALIZABLE |