Re: Rollback on close [Was: Fwd: [DB-SIG] conn.close() idempotence] - Mailing list psycopg

From Daniele Varrazzo
Subject Re: Rollback on close [Was: Fwd: [DB-SIG] conn.close() idempotence]
Date
Msg-id CA+mi_8YA4Qmdgz99DaGDoypxJ+rzSvE-z020P3t80u1ti9Q3TQ@mail.gmail.com
Whole thread Raw
In response to Re: Rollback on close [Was: Fwd: [DB-SIG] conn.close() idempotence]  (Marko Kreen <markokr@gmail.com>)
List psycopg
On Wed, Oct 19, 2011 at 6:50 PM, Marko Kreen <markokr@gmail.com> wrote:

> Let's clarify that further - DB-API *MUST* say that .close()
> [and also .__del__()] will rollback.  Eg - imagine
> driver accessing .dbf / .csv files directly.
>
> But there is no good reason to require communication
> with backend if the backend will rollback anyway
> on connection close.

Yes, this is our current position. Changing from it could only improve
the interaction between "sloppy code" and middleware, and as you have
pointed out this would make edge case more troublesome.

Thank you for the discussion, I wanted to hear the opinion of somebody
on the other side of the socket :)


-- Daniele

psycopg by date:

Previous
From: Marko Kreen
Date:
Subject: Re: Rollback on close [Was: Fwd: [DB-SIG] conn.close() idempotence]
Next
From: "Frank Kauff"
Date:
Subject: Re: Error: no Python Codec for client encoding