From: | messias <messias(at)fw(dot)hu> |
---|---|
To: | sdavis2(at)mail(dot)nih(dot)gov |
Cc: | pgsql-novice(at)postgresql(dot)org |
Subject: | Re: Re: Monitoring new records |
Date: | 2008-02-29 07:06:41 |
Message-ID: | E1JUzKL-0002Jp-00@osiris.swi.hu |
Views: | Raw Message | Whole Thread | Download mbox | Resend email |
Thread: | |
Lists: | pgsql-novice |
--- Sean Davis <sdavis2(at)mail(dot)nih(dot)gov> wrote (2008.02.28 22:56:08):
---
> 2008/2/28 messias <messias(at)fw(dot)hu>:
> > Hello
> >
> >
> >
> > I am new in SQL programing and I have a problem. I would like
to
> >
> > monitor the new records which is inserted into an SQL table.
> >
> > I found some solution which mentioned a queue (log) table which
can
> >
> > be filled by a trigger. The client would read from that queue
table,
> >
> > and delete the readed data.
> >
> > I think this solution can work with only one client.
> >
> > But I would have more than one client at the same time, and I
do
not
> >
> > know how I can delete the data from the queue table.
> >
> > I can make queue table for every client, but I think this is
not
a
> >
> > nice solution.
> >
> > I am affraid of lost client....
> >
> > I am intrested in every solution which can solve my problem.
>
> Why not simply add a timestamp column to each row. Then, your
client
> could query for all rows created or modified after a certain
> time/date.
Yes it can be a good solution, but that table can have more than
100000 records.
I don't think a trigger and a separate table would be
> necessary, but it is hard to know without more information.
>
> Sean
>
I think if I made a copy of the new records into a separate table
the
selection would be more efficient. However, somebody has to delete
from that table.
Ferenc
From | Date | Subject | |
---|---|---|---|
Next Message | Sean Davis | 2008-02-29 11:07:28 | Re: Monitoring new records |
Previous Message | Tom Lane | 2008-02-29 04:55:42 | Re: Postgres 8.3 - C function taking and returning arrays |