Re: Accelerating aggregates - Mailing list pgsql-hackers

From Greg Stark
Subject Re: Accelerating aggregates
Date
Msg-id 87y8mp987h.fsf@stark.xeocode.com
Whole thread Raw
In response to Re: Accelerating aggregates  (Steve Atkins <steve@blighty.com>)
Responses building rpms from source rpm's
List pgsql-hackers
Steve Atkins <steve@blighty.com> writes:

> On Fri, Jun 11, 2004 at 12:17:57PM -0400, Greg Stark wrote:
>
> > no, read committed would see any other updates that have been committed since
> > the start of your transaction. 
> 
> Uhm... only updates within the current transaction. 

No, "read committed" refers to being able to read any updates that are
committed, even if they were committed after the start of your transaction:


For example:



db=> begin;
BEGIN
                                       db=> begin;                                       BEGIN

db=> insert into test values (1);
INSERT 6725927 1
db=> select * from test;a 
---1
(1 row)
                                       db=> select * from test;                                        a
                       ---                                       (0 rows)
 

db=> commit;
COMMIT
                                       db=> select * from test;                                        a
                       ---                                        1                                       (1 row)
 


-- 
greg



pgsql-hackers by date:

Previous
From: Tom Lane
Date:
Subject: Spinlock assembly cleanup
Next
From: David Garamond
Date:
Subject: "An old version of the database format was found."