Re: proposal: set GUC variables for single query

From: Jan Urbański <wulczer(at)wulczer(dot)org>
To: Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us>
Cc: Postgres - Hackers <pgsql-hackers(at)postgresql(dot)org>
Subject: Re: proposal: set GUC variables for single query
Date: 2011-10-16 15:57:06
Views: Raw Message | Whole Thread | Download mbox | Resend email
Lists: pgsql-hackers

On 16/10/11 17:49, Tom Lane wrote:
> =?UTF-8?B?SmFuIFVyYmHFhHNraQ==?= <wulczer(at)wulczer(dot)org> writes:
>> this idea has cropped up last PGCon - the ability to set GUC variables
>> for the duration of a single query. It would work by setting the GUCs
>> for the duration of the query and setting them back to what they were
>> after it has terminated.
> Doesn't SET LOCAL cover this use-case pretty well already?

It does to a certain degree. If you have a bunch of statements in a
transaction and want to execute one of them with a different timezone
setting, you have to do the SET/RESET dance. In theory you should also
first grab the current value to set it back afterwards, in case someone
else did SET LOCAL before you, but I'll admin that's far-fetched.

The main use case would be apps running behing pgbouncer and using
statement pooling, and plain convenience.

I'd find it useful myself, but for now I'm making do with SET LOCAL and
it works fine. I'm bringing it up because it appears in the TODO created
at the PL Summit:

* Further discussion of per-statement config parameters for things
like timezone - Jan Urbanski

Tryin' to do my bit and all ;)


In response to

Browse pgsql-hackers by date

  From Date Subject
Next Message Florian Pflug 2011-10-16 16:10:51 Re: Pushing ScalarArrayOpExpr support into the btree index AM
Previous Message Thom Brown 2011-10-16 15:50:57 Re: proposal: set GUC variables for single query