Re: It looks like transaction, but it isn't transaction

From: Sergey Samokhin <prikrutil(at)gmail(dot)com>
To: Sam Mason <sam(at)samason(dot)me(dot)uk>
Cc: pgsql-general(at)postgresql(dot)org
Subject: Re: It looks like transaction, but it isn't transaction
Date: 2009-08-27 00:36:36
Message-ID: e42595410908261736i73108ac9w6bc9433f47cfdd90@mail.gmail.com
Views: Raw Message | Whole Thread | Download mbox | Resend email
Thread:
Lists: pgsql-general

Hello, Sam.

Thanks for the answer.

> This behavior seems to be what libpq exposes by default:
>
>  http://www.postgresql.org/docs/current/static/libpq-exec.html

The driver I was talking about isn't written in C and doesn't use
libpq behind the scene like interfaces for Perl/Python/Tcl do, so what
I see is more likely refers to how PostgreSQL itself executes queries.
I haven't found a good explanation of this behaviour yet.

> I always tend to bracket things in an explicit BEGIN+COMMIT, why
> wouldn't you do this?

Probably that is what I will end up with. But anyway I would like to
know for sure what causes queries to be executed in "almost
transactional" context.

I should rejig my question: is this normal for a query made up from
several commands to be completely discarded if an error occurs? If the
answer is "no, it seems unusual" - I should write a bug report to the
author of the driver.

--
Sergey Samokhin

In response to

Responses

Browse pgsql-general by date

  From Date Subject
Next Message Tom Lane 2009-08-27 01:25:56 Re: It looks like transaction, but it isn't transaction
Previous Message Mark Kirkwood 2009-08-26 22:18:01 Re: [Skytools-users] WAL Shipping + checkpoint