Re: Nested Transactions, Abort All - Mailing list pgsql-hackers

From Simon Riggs
Subject Re: Nested Transactions, Abort All
Date
Msg-id 1089398502.17493.571.camel@stromboli
Whole thread Raw
In response to Re: Nested Transactions, Abort All  (Andreas Pflug <pgadmin@pse-consulting.de>)
Responses Re: Nested Transactions, Abort All
List pgsql-hackers
On Fri, 2004-07-09 at 11:45, Andreas Pflug wrote:
> Simon Riggs wrote:
> 
> >ISTM - my summary would be
> >1. We seem to agree we should support SAVEPOINTs
> >
> >2. We seem to agree that BEGIN/COMMIT should stay unchanged...
> >
> >  
> >
> >>With savepoints, it looks pretty strange:
> >>    
> >>BEGIN;
> >>    SAVEPOINT x1;
> >>    INSERT INTO ...;
> >>    SAVEPOINT x2;
> >>    INSERT INTO ...;
> >>    SAVEPOINT x3;
> >>    INSERT INTO ...;
> >>
> >>    
> >>
> >
> >This isn't how you would use SAVEPOINTs...look at this...
> >
> >BEGIN
> >            display one screen to user - book the flight
> >    INSERT INTO ...
> >    INSERT INTO ...
> >    UPDATE ...
> >    SAVEPOINT
> >            display another related screen - book the hotel
> >    INSERT INTO
> >    DELETE
> >    UPDATE
> >    UPDATE
> >    SAVEPOINT
> >            offer confirmation screen
> >COMMIT (or ROLLBACK)
> >  
> >
> 
> No, SAVEPOINT is not some kind of intermediate commit, but a point where 
> a rollback can rollback to.

Hmmm....I'm not sure what you mean by "No". The SAVEPOINT is somewhere
you can ROLLBACK to, yes - exactly what I'm saying. 

I've not introduced any concept of "intermediate commit"...

Do you agree that my example is valid Oracle SQL? 

Best Regards, Simon Riggs





pgsql-hackers by date:

Previous
From: Christopher Kings-Lynne
Date:
Subject: Re: pg_dump 7.4 bug
Next
From: Simon Riggs
Date:
Subject: Re: PITR Redo Create Database fails