Re: Vote totals for SET in aborted transaction - Mailing list pgsql-hackers

From Bruce Momjian
Subject Re: Vote totals for SET in aborted transaction
Date
Msg-id 200204260443.g3Q4hm821818@candle.pha.pa.us
Whole thread Raw
In response to Re: Vote totals for SET in aborted transaction  ("Marc G. Fournier" <scrappy@hub.org>)
List pgsql-hackers
Marc G. Fournier wrote:
> > Yes, let's find out what the others do.  I don't see DROP TABLE
> > rollbacking as totally different.  How is it different from SET?
> 
> SET currently has an "accepted behaviour" with other DBMSs, or, at least,
> with Oracle, and that is to ignore the rollback ...
> 
> DROP TABLE also had an "accepted behaviour", and that was to leave it
> DROPed, so "oops, I screwed up and just lost a complete table as a
> result", which, IMHO, isn't particularly good ...
> 
> NOTE that I *do* think that #1 is what *should* happen, but there should
> be some way of turning off that behaviour so that we don't screw up ppl
> expecting "Oracles behaviour" ... I just think that implementing #1
> without the 'switch' is implementing a half-measure that is gonna come
> back and bite us ...

Yes, I understand, and the logical place would be GUC.  However, if we
add every option someone would ever want to GUC, the number of options
would be huge.

We currently have a problem doing #2.  My suggestion is that we go to #1
and wait to see if anyone actually asks for the option of choosing #3.

--  Bruce Momjian                        |  http://candle.pha.pa.us pgman@candle.pha.pa.us               |  (610)
853-3000+  If your life is a hard drive,     |  830 Blythe Avenue +  Christ can be your backup.        |  Drexel Hill,
Pennsylvania19026
 


pgsql-hackers by date:

Previous
From: Lincoln Yeoh
Date:
Subject: Re: Vote totals for SET in aborted transaction
Next
From: Curt Sampson
Date:
Subject: Re: Block size: 8K or 16K?