Re: Postgres config file: autocommit = off

From: Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us>
To: Bruce Momjian <pgman(at)candle(dot)pha(dot)pa(dot)us>
Cc: "Nigel J(dot) Andrews" <nandrews(at)investsystems(dot)co(dot)uk>, Rasmus Resen Amossen <rresena(at)hotmail(dot)com>, pgsql-hackers(at)postgresql(dot)org
Subject: Re: Postgres config file: autocommit = off
Date: 2003-06-02 19:11:25
Message-ID: 11226.1054581085@sss.pgh.pa.us
Views: Raw Message | Whole Thread | Download mbox | Resend email
Thread:
Lists: pgsql-hackers

Bruce Momjian <pgman(at)candle(dot)pha(dot)pa(dot)us> writes:
> Tom Lane wrote:
>> Yeah, it's a bit warty, but psql -c is inherently different from a psql
>> script. IMHO it's not reasonable to make (shell) scripts using psql -c

> How is it different, except for having no prompt?

It's different because it's explicitly designed for convenient execution
of a single command. Thus, autocommit off would be useless and
counterproductive. Also, since the -c string is fed to PQexec as a
single query (again different from the psql stdin behavior), if you do
put multiple commands into -c then you get them executed as a single
transaction anyway. So you do not need or want .psqlrc modifying the
behavior in either case.

One of the reasons for taking autocommit control out of the backend and
pushing it up to the client level is exactly to make it feasible to take
these sorts of application-level considerations into account when
choosing the behavior.

regards, tom lane

In response to

Responses

Browse pgsql-hackers by date

  From Date Subject
Next Message Bruce Momjian 2003-06-02 19:14:33 Re: Postgres config file: autocommit = off
Previous Message Peter Eisentraut 2003-06-02 19:07:11 Re: [HACKERS] Are we losing momentum?