Re: Question: update and transaction isolation - Mailing list pgsql-hackers

From Hiroshi Inoue
Subject Re: Question: update and transaction isolation
Date
Msg-id 3CABA7A0.5A49A6D0@tpf.co.jp
Whole thread Raw
In response to Re: Question: update and transaction isolation  (Peter Eisentraut <peter_e@gmx.net>)
List pgsql-hackers
Tom Lane wrote:
> 
> Peter Eisentraut <peter_e@gmx.net> writes:
> > Tom Lane writes:

> To my mind, full SERIALIZABLE mode is the only approach that can be
> explained in terms of simple notions like "you see only the data that
> existed at time T".

There's another way. If the current value is different from
that at time T, we may be able to reset the time when the
statement begun, which is equivalent to replaceing the snapshot
(this isn't allowed in serializable mode). Of cource it would
be very difficult to implement(at least effectively).

As I've already mentioned many times SELECT and SELECT ..
FOR UPDATE are alike in appearance but quite different in
nature. For example, the meaning of the snapshot isn't the
same as you've pointed out already in this thread.
It's meaingless for SELECT and UPDATE(SELECT .. FOR UPDATE)
to have a common snapshot.

regards,
Hiroshi Inoue


pgsql-hackers by date:

Previous
From: Bruce Momjian
Date:
Subject: Re: Locale support is now on by default
Next
From: Hiroshi Inoue
Date:
Subject: Re: timeout implementation issues