* Bruce Momjian (pgman@candle.pha.pa.us) [020423 12:30]:
>
> 1 - All SETs are rolled back in aborted transaction
> 2 - SETs are ignored after transaction abort
> 3 - All SETs are honored in aborted transaction
> ? - Have SETs vary in behavior depending on variable
>
> Our current behavior is 2.
>
> Please vote and I will tally the results.
#2, no change in behavior.
But I base that on the assumption that #1 or #3 involve serious amounts
of work, and don't see the big benefit.
I liked the line of thought that was distinguishing between in-band
(rolled back) and out-of-band (honored) SETs, although I don't think
any acceptable syntax was arrived at, and I don't have a suggestion.
If this were solved, I'd vote for '?'.
Hmm. Maybe I do have a suggestion: SET [TRANSACTIONAL] ...
But it might not be very practical.
-Brad