Request a Demo
Home   >   mailing lists

Re: Keeping track of updates/performances - Mailing list pgsql-novice

From Romain Billon-Grand
Subject Re: Keeping track of updates/performances
Date
Msg-id DUB124-W17216178BA23AAE1F26BE8CE030@phx.gbl
Whole thread Raw
In response to Re: Keeping track of updates/performances  (Felipe Santos <felipepts@gmail.com>)
Responses Re: Keeping track of updates/performances
List pgsql-novice
Tree view
Many thanks for all your helpfull answers. May be I was not accurzte enough in my questions...

So:

´now'::timestamp why not ´simply' current_timestamp?

And..

I have red about inheritance vs Like, but I wanted to know if one of those would have better performance

Date: Tue, 17 Mar 2015 09:18:11 -0300
Subject: Re: [NOVICE] Keeping track of updates/performances
From: felipepts@gmail.com
To: akretschmer@spamfence.net
CC: pgsql-novice@postgresql.org



2015-03-17 5:50 GMT-03:00 Andreas Kretschmer <akretschmer@spamfence.net>:
Romain Billon-Grand <billongrand@hotmail.fr> wrote:

> Hello!
>
> Please give me your advices about the best way to keep in each table those
> informations:
> name of the user who created a row
> name of the last user who modified it
> timestamp for creation
> timestamp for last modification
>
> 1/ I have been looking for built in options in postgres but I guess I have to
> build it myself...?
> Of course if Postgres already does it I am very happy, and all of the following
> quetsions are useless but I did not find this in the documentation...
>
> 2/ Is there some SQL query to return the "usermane" from pg_catalog or
> something

You can use TRIGGER for Insert/Update, the current user can you detect
with current_user.


Andreas
--
Really, I'm not out to destroy Microsoft. That will just be a completely
unintentional side effect.                              (Linus Torvalds)
"If I was god, I would recompile penguin with --enable-fly."   (unknown)
Kaufbach, Saxony, Germany, Europe.              N 51.05082°, E 13.56889°


--
Sent via pgsql-novice mailing list (pgsql-novice@postgresql.org)
To make changes to your subscription:
http://www.postgresql.org/mailpref/pgsql-novice




"What would be the advantages and drawbacks of this :
CREATE TABLE anytable(...) INHERITS (tracks)
vs this: CREATE TABLE anytable (...) LIKE tracks?"


I see them differently.

INHERITS creates a child-table that is still connected with the parent-table. I use this feature to emulate the partition-feature that exists in other RDBMSs. If you add a column to the father table, it will automatically appear on the child-table, and that is exactly what I want. And then I use triggers to do the "partition function/scheme" (in SQL Server words). You can partition by date, by user, by country or any other thing that makes sense to you. Also, if you query the child table you get only the child-table's rows. But if you query the parent-table you get parent-table's rows PLUS all child-tables' rows.

CREATE TABLE LIKE is like making a photocopy of a table to create another one, but just that. The tables are not "connected" and dont have any kind of relation with one another.

Hope it helps

pgsql-novice by date:

Previous
From: Luca Ferrari
Date:
Subject: Re: Keeping track of updates/performances
Next
From: Luca Ferrari
Date:
Subject: Re: [GENERAL] psql sqlstate return code access