Re: Decoding of (nearly) empty transactions and regression tests - Mailing list pgsql-hackers

From Andres Freund
Subject Re: Decoding of (nearly) empty transactions and regression tests
Date
Msg-id 20140901115050.GA20990@awork2.anarazel.de
Whole thread Raw
In response to Re: Decoding of (nearly) empty transactions and regression tests  (Tom Lane <tgl@sss.pgh.pa.us>)
List pgsql-hackers
On 2014-06-29 11:24:35 -0400, Tom Lane wrote:
> Andres Freund <andres@2ndquadrant.com> writes:
> > Hm? I think we're misunderstanding each other - I was thinking of tacking
> > ALTER TABLE .. SET (AUTOVACUUM_ENABLED = false) to the tables created in
> > test_decoding/sql/, not to something outside.
>
> Ah, got it.  Yeah, seems like an OK workaround.

I'd tried that back then, but it didn't really work out very
well. Sometimes there's still autovacuums running on tables we don't
want to disable autovacuum on.

My new proposal (patch attached) is a skip-empty-xacts option for
test_decoding. It causes test_decoding not to output BEGIN/COMMIT for
transactions that don't output any rows.

I don't really like that approach, but it's surely better than
continually coloring the CLOBBER animal red. And the patch is, besides
the verbiage of regression test output changes, not that bad.

Unless somebody protests I'll apply that to master/9.4 later today.

Greetings,

Andres Freund

--
 Andres Freund                       http://www.2ndQuadrant.com/
 PostgreSQL Development, 24x7 Support, Training & Services

Attachment

pgsql-hackers by date:

Previous
From: Pavel Stehule
Date:
Subject: Re: Adding 'crosstab' variants returning refcursor?
Next
From: Kyotaro HORIGUCHI
Date:
Subject: Re: alter user set local_preload_libraries.