Re: [patch] Adding EXTRA_REGRESS_OPTS to all pg_regress invocations

From: Michael Paquier <michael(dot)paquier(at)gmail(dot)com>
To: Bruce Momjian <bruce(at)momjian(dot)us>
Cc: Christoph Berg <cb(at)df7cb(dot)de>, Andrew Dunstan <andrew(at)dunslane(dot)net>, PostgreSQL Hackers <pgsql-hackers(at)postgresql(dot)org>
Subject: Re: [patch] Adding EXTRA_REGRESS_OPTS to all pg_regress invocations
Date: 2013-12-10 03:57:12
Message-ID: CAB7nPqQyH062MXC4_Cx1tRd9_FJhNKM5wxHyp=+B07GuapY0XA@mail.gmail.com
Views: Raw Message | Whole Thread | Download mbox | Resend email
Thread:
Lists: pgsql-hackers

On Tue, Dec 10, 2013 at 12:08 PM, Bruce Momjian <bruce(at)momjian(dot)us> wrote:
> On Thu, Dec 5, 2013 at 09:52:27AM +0100, Christoph Berg wrote:
>> > The change is sane in itself. It won't affect anyone who doesn't use
>> > EXTRA_REGRESS_OPTS. Why would we want to make packagers do MORE
>> > work?
>>
>> The patch has been in the Debian/Ubuntu/apt.pg.o packages for some
>> time, for 8.3+. I'm attaching the patches used there.
>>
>> (Sidenote: To enable building of several package flavors in parallel
>> on the same machine we use
>>
>> make -C build check-world EXTRA_REGRESS_OPTS='--host=/tmp --port=$(shell perl -le 'print 1024 + int(rand(64000))')'
>>
>> so pg_regress' static per-version ports do not conflict. But 9.2's
>> contrib/pg_upgrade/{Makefile/test.sh} do not like --port in there, so
>> the 9.2 patch has an extra sed hack in there to remove --port for
>> pg_upgrade. That bit should probably not be applied for general use.
>> The rest is safe, though.)
>
> OK, Christoph has provided a full set of tested patches back to 8.4.
> Should I backpatch these? Peter says no, but two others say yes.
My 2c. Adding a new feature in a maintenance branch is usually not
done, so I'd vote no.

Regards,
--
Michael

In response to

Browse pgsql-hackers by date

  From Date Subject
Next Message Amit Kapila 2013-12-10 04:02:09 Re: plpgsql_check_function - rebase for 9.3
Previous Message Bruce Momjian 2013-12-10 03:42:41 Re: Proof of concept: standalone backend with full FE/BE protocol