Re: serializable read only deferrable

"Kevin Grittner" <Kevin(dot)Grittner(at)wicourts(dot)gov> writes:
> Florian Pflug <fgp(at)phlo(dot)org> wrote:
>> Say you've written a trigger which enforces some complex
>> constraint, but is correct only for SERIALIZABLE transactions. By
>> simply sticking a "SET TRANSACTION ISOLATION LEVEL SERIALIZABLE"
>> at the top of the trigger you'd both document that fact it is
>> correct only for SERIALIZABLE transactions *and* prevent
>> corruption should the isolation level be something else due to
>> a pilot error. Nice, simply and quite effective.

> It would be great to have a way within a trigger, or possibly other
> functions, to assert that the transaction isolation level is
> serializable. What gives me pause here is that the standard allows
> you to specify a more strict transaction isolation level within a
> subtransaction without error, so this way of spelling the feature is
> flirting with rather nonstandard behavior.

Yes. This is not the way to provide a feature like that.

> Is there maybe a better way to check this?

You can always read the current setting and throw an error if you
don't like it.

regards, tom lane